ASSESSMENT OF FARMERS EVALUATION CRITERIA AND ADOPTION OF IMPROVED BREAD WHEAT VARIETIES

M. Sc. Thesis

DEREJE HAMZA MUSSA

December 2005 Alemaya University

ASSESSMENT OF FARMERS EVALUATION CRITERIA AND ADOPTION OF IMPROVED BREAD WHEAT VARIETIES

A Thesis Submitted to the Department of Rural Development and Agricultural Extension, School of Graduate Studies ALEMAYA UNIVERSITY

In Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE IN AGRICULTURE (AGRICULTURAL EXTENSION)

By Dereje Hamza Mussa

December 2005 Alemaya University

APPROVAL SHEET OF THESIS

SCHOOL OF GRADUATE STUDIES ALEMAYA UNIVERSITY

As members of the Examining Board of the Final M. Sc. Open Defense, We certify that we have read and evaluated the thesis prepared by DEREJE HAMZA MUSSA and recommend that it be accepted as fulfilling the thesis requirement for the degree of MASTER OF SCIENCE IN AGRICULTURE (AGRICULTURAL EXTENSION) . Name of Chairman Signature Date

. Name of Internal Examiner

Signature

Date

. Name of External Examiner

Signature

Date

Final approval and acceptance of the thesis is contingent up on the submission of the final copy of the thesis to the council of the Graduate Studies (CGS) through the Departmental Graduate Committee (DGC) of the candidates Major Department. I here by certify that I have read this thesis prepared under my direction and recommend that it be accepted as fulfilling the thesis requirement. Ranjan S. Karippai (Ph.D ) Name of Thesis Advisor .. Signature . Date

ii

DEDICATION

I dedicate this thesis manuscript to my wife, TIRUWORK ABATE and my son, SOLOMON DEREJE, for their love and untold-enormous partnership effort in my academic success.

iii

STATEMENT OF AUTHOR

First, I declare that this thesis is my bona fide work and that all sources of materials used for this thesis have been duly acknowledged. This thesis has been submitted in partial fulfillment of the requirements for an advanced M. Sc. degree at Alemaya University and is deposited at the University Library to be made available to borrowers under rules of the Library. I solemnly declare that this thesis is not submitted to any other institution anywhere for the award of any academic degree, diploma, or certificate.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the School of Graduate Studies when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

Name:

DEREJE HAMZA MUSSA

Signature:

Place: Alemaya University, Alemaya Date of Submission: December 2005.

iv

BIOGRAPHICAL SKETCH
Dereje Hamza Mussa was born in Jamma District (Sora-Micha village), South Wollo Zone, and Amhara region on August 6, 1964. He attended his elementary and junior education at Boren -Teklehaimanot and Jamma-Degollo elementary and junior schools (both found in my district) respectively. He also attended his High-school education at Woreilu secondary Senior high school. After completion of his high school education, he joined Awassa Junior Agricultural College (under Addis Ababa University) (now Debub University) to attend a two years Diploma program in Animal Science and Technology. After graduation he was employed in Ministry of Agriculture and has worked for more than 15 years. After this much time service he got an opportunity to join at Alemaya University to attend his degree program education in Agricultural Extension in Mid-career program from 1999 to 2002. After completion and graduation his education he turned back and joins at Alemaya University to attend his M. Sc. degree education in Agricultural Extension Since 2004.

ACKNOWLEDGEMENT

First and for most, I am greatly indebted to Ranjan S. Karippai (Ph.D) my major advisor and Senait Regassa (Ph.D) my co-advisor for their unreserved help, advice, directing, insight guidance, support on the field, critical review of my thesis manuscript, invaluable support and suggestions as without their professional help it was difficult to be successful in my research work and Thesis write up; in addition, my acknowledgement should forwarded to Dr. Ranjan for his professional and critical review and Dr. Senait for her help in SPSS and Limdep computer soft wares as well as Logit, Probit and Tobit, econometrics models, t-test and x2-test statistics uses and application. My sincere thanks should also go to Tesfaye Lema (Ph.D) and Tesfaye Beshah (Ph.D) for their unlimited review of my thesis manuscript help, guide and continues encouragement to be successful in my study and research.

Above all, I am greatly indebted to Ato Zewdu Teferi and his children (Eyu, David and Dani), Seid Ahmed (Ph.D), Solomon Asseffa, (PhD) for their greatest financial and material contribution as well as moral encouragement and all sided help.

My thanks and appreciation should also extend to many individuals, to Belayneh Leggesse (Ph.D), Prof. Panjabin, Asegdew Gashaw with his wife, Mehadi Egi, Ato Walelign and his wife Abaye, Bizuhayehu Asfaw and Amare Berhanu and Tewodros Alemayhu from Alemaya university; to Ato Yishak Berado, Amsalu Bedaso, Admassu Terefe and his wife Belay with their children Eyuti and Mitisha from Alage technical agricultural college; to Ato Chane Gebeyhu, Ato Gebyhu, Alemayehu and Tekle from Akaki-Kality subcity and agricultural unit; to woreilu wereda agricultural office staff members, and Lulseged Bekele, Mohammod Yimer , Abebaw gidelew and (Ato Eshetu Woraei and w/o Toyiba -through their every day pray) from woreilu woreda ; to Ato kasye Afre and his wife Turye Getye with their family, particularly Mamush, Zelalem and Sisaye Kasye-they always accompanied me to and from bus station of Addis Ababa in my every travel to or arrival from Alemaya University for academic and research purposes; to Wondye Kasye, Tesafa Belay, Alemye Argaw, Esubneh

vi

Checolle, Kasye Mohammed and to my brothers Endale and Eshetu Hamza should deserve acknowledgement for their moral, financial and material as well as all sided helps, wishes and encouragements to accomplish my study successfully.

My sincere and special thanks should go to Jifar Tarekegn and Yodit Fekadu for their free charge and complete computer, office provision and all sided co-operations; particularly, Jifar Tarekegn for his additional and unlimited helps in computer and statistical manipulation through out my thesis write up.

My heart felt and deepest thanks should go to Tiruwork Abate and Solomon Dereje my wife and my son respectively who received and paid all suffers and scarifications but the greatest contributors and partnership in my research and academic success.

I would like to extend my thanks to my mother-Ayelech Sebsibe, and my wife's mothers Alganesh Afre, who are always with me in help and wish for my success through their everyday pray.

Several organizations, Alemaya University, School of Graguate Studies, Department of Rural Development and Agricultural Extension of Alemaya University, Debre-Zeit Research Center and Agricultural Unit of Akaki-Kality sub-city, Alage Agricultural College should deserve acknowledgement for their contributions to my study. At last but not the least, I would like to extend my acknowledgement to IFAD (International Fund for Agricultural Development) that offered a budget support for this research through EARO and Debrezeit research center.

vii

LIST OF ABBREVIATIONS

AU B.B.M. DA EA EARO EARI IR PA RKA

Alemaya University Broad Bed Molder Development Agent Extension Agent Ethiopian Agricultural Research Organization Ethiopian Agricultural Research Institute Institution of Research Peasant Association Rural Kebele Administration

viii

TABLE OF CONTENTS

APPROVAL SHEET OF THESIS DEDICATION STATEMENT OF AUTHOR BIOGRAPHICAL SKETCH ACKNOWLEDGEMENT LIST OF ABBREVIATIONS TABLE OF CONTENTS LIST OF TABLES LIST OF TABLES IN THE APPENDIX LIST OF TABLES IN THE APPENDIX ABSTRACT 1. INTRODUCTION
1.1. Background 1.2. Statement of the Problem 1.3. Objectives of the Study 1.4. Significance of the Study 1.5. The Scope and Limitations of the Study 1.6. Organization of the Thesis

ii iii iv v vi viii ix xi xiii xiii xiv 1

1 2 4 4 5 6

2. LITERATURE REVIEW
2.1. Concept and Theoretical Framework of Adoption 2.2. Empirical Studies on Adoption 2.3. Farmers Participation in Agricultural Technologies Development and Evaluations 2.4. Conceptual Framework of the Study

7
7 12 19 23

3. RESEARCH METHODOLOGY
3.1. Description of the Study Area (Akaki) 3.1.1. Location, relief and climate 3.1.2. Agriculture and demographic characteristics of the study area 3.1.3. Institutional services of the study area 3.2. Description of Data Collection and Data Analysis Methods and Procedures 3.2.1. Sources and types of data 3.2.2. Sample size and sampling techniques 3.2. 3. Data collection methods

27
27 27 30 31 34 34 35 36

ix

TABLE OF CONTENTS (Continued)

3.2.3.1. Quantitative data collection methods 3.2.3.2. Qualitative data collection method 3.3. Analytical Models 3.3.1. Logit model 3.3.2. Tobit model 3.3.3. Other Quantitative data analysis methods 3.3.4. Qualitative data analysis method 3.4. Hypotheses Testing and Definitions of Variables 3.4.1. The Dependent variables of logit and tobit models 3.4.1.1. The Dependent variable of logit model 3.4.1.2. The Dependent variable of tobit model 3.4.2. The Independent variables and their definitions used in logit and tobit models

36 37 37 37 40 42 43 43 44 44 44 44

4. RESULTS AND DISCUSSION

50

4.1. Analysis through descriptive statistics 50 4.1.1. Sample Households Demographic Characteristics 50 4.1.2. Respondents` livestock and land ownership 60 4.1.3. Accessibility of respondents to different institutional services 63 4.1.4. Agricultural information sources of the study area 72 4.1.5. Farmers selection and evaluation criteria of improved bread wheat varieties 76 4.2. Analytical results and discussion 79 4.2.1. Analysis of determinants influencing probability of adoption of improved bread wheat varieties and their marginal effect 85 4.2.2. Analysis of determinants influencing intensity of adoption of improved bread wheat varieties and their marginal effects 90

5. SUMMARY AND CONCLUSION

5.1. Summary 5.2. Conclusion and Recommendations

101
101 105

6. REFERENCES 7. APPENDICES
Appendix.1.Information on sample household demographic and socio-economic characteristics Appendix.2. Interview Schedule for data collection from. Farmers

109 114
115 126

LIST OF TABLES Tables Pages

1.The Livestock and crop types in the study area 2.The land use of farmers in the study area 3.The summary of oxen ownership 4.Improved agricultural input distribution of the study area in different years 5. Improved bread wheat seed distribution of the study area in different years 6.Credit Distribution of the study area in various years 7.Sample household heads distribution by Sex, Kebele and adoption category 8.Marital status of respondents 9.Association of adoption of improved bread wheat and sex of sample household head 10. Respondent farmers demographics characteristics 11.Adopters and non-adopters demographic characteristics 12.Reasons given for not using improved bread wheat varieties 13.Level of awareness of improved bread wheat varieties 14.Sample Farmers perception on benefit of fertilizer 15.Beginning time of cultivation of improved bread wheat varieties of sample farmers 16.Health status and adoption of improved bread wheat varieties 17.Sample household educational status 18.Livestock and land ownership of respondents farmers 19.Respondents land ownership in 1996/97 Ethiopian major cropping season 20.Respondents opinion on extension service of the study area 21.Extension support on improved bread wheat varieties and distance of DAs office 22.Summary of respondents opinion on credit 23.Association between credit and market service 24.Summary of households accessibility of off-farm job 25.Respondent farmers reasons for not involvement of their family in off-farm job 26.Rrespondents opinion on decision of off-farm and other household resources 27 Pattern of off-farm income utilization of respondent farmers

29 30 31 32 33 34 51 52 53 53 54 56 56 57 58 58 59 61 62 63 64 65 66 67 68 68 69

xi

28.Family labor utilization of respondent farmers 29.Types of activities and family labor utilization of respondents 30.Respondents accessibility to non-family labor and to off-farm income 31.Respondent farmers labor sources outside their family members 32.Respondents participation in training, field day and demonstration 33.Respondent farmers sources of information 34.Farmers evaluation and selection criteria of improved bread wheat varieties 35.Farmers preference (selection and evaluation criteria) of improved bread wheat varieties disseminated in the study area 36.Variable Inflation Factor for the continuous explanatory variables

70 70 71 72 74 75 77

78 84

37.Contingency Coefficients for Dummy Variables of Multiple Linear Regressions Model 84 38.Factors affecting Probability of adoption of improved bread wheat varieties and the 39.The effects of changes (marginal effect) in the significant explanatory variables on the intensity of adoption of improved bread wheat varieties 92 88

xii

LIST OF TABLES IN THE APPENDIX

Appendix Tables 1.The distribution of sample respondents by age gro 2 Educational statuses of sample house hold head farmer 3 .The sample household family size 4.The sample household family size 5. Total Family members of sample households in age group 6.Respondents farming experience 7.Types of livestock and owners and the number of respondents 8 .Sample house hold oxen ownership 9. Sample house hold land ownership 10.Size of farmland holding of sample household

Pages 115 115 116 116 117 117 117 118 118 119

11.Respondents average land area and yield of wheat crops in 1996/97E.C.cropping season119 12 Respondents farm land ownership and crop type grown in 1996/97 E.C.cropping season120 13.Respondents livestock ownership 14 Respondents livestock ownership in Tropical Livestock Unit (TLU) 15. Conversion factors used to estimate the households livestock ownership in tropical livestock units (TLU) 16. Discrete characteristics of respondents 17. Respondent farmers general information 18. Factors affecting Intensity of adoption of improved bread wheat varieties (Maximum Likelihood Tobit Model Estimation) 19. Household characteristics 20. Land holding & Farm Characteristics of the sample households 21. Livestock ownership 22. Types of crop grown in the survey year 23. Improved bread wheat varieties characteristics 24. Cramers V and Pearsons R values for Discrete and Continuous variables 25.Resondents leadership position 125 126 127 127 128 130 133 133 122 123 124 121 122

xiii

ASSESSMENT OF FARMERS EVALUATION CRITERIA AND ADOPTION OF IMPROVED BREAD WHEAT VARIETIES

Major advisor: Ranjan S. Karippai (Ph.D) Co-advisor: Senait Regassa Bedadda (Ph.D)

ABSTRACT

Wheat is beneficial to man long before the dawn of recorded history. Ethiopia is one of the largest wheat producers in sub-Saharan African next to South Africa. Wheat is one of the most important cereal crops grown in the study area, Akaki. It contributes to the major share of daily consumption and cash source. The objectives of this study were: to identify farmers evaluation and selection criteria of improved bread wheat varieties disseminated in the study area; to assess probability and intensity of adoption of farmers in the study area; and to know and analyze determinants of probability and intensity of adoption of improved bread wheat varieties in the study area. In this study, data were collected and analyzed qualitatively and quantitatively. Quantitative data analysis methods employed in this study were (percentage, tabulation, t-test and X2, Logit and Tobit models) using SPSS and Limdep computer soft ware programs and qualitatively through group discussion and observations.. In farmers evaluation and selection criteria of improved bread wheat varieties disseminated in the study area HAR1685 ranks first, Paven-76 second and HAR-1709 third. White color, large grain size, market demand, straw quality were the most important, germination capacity, cooking quality, better yield performance were the second important, water logging resistance, tillering capacity, good food quality, short maturity date the third important, disease and pest resistance and frost resistance the fourth, storage and harvesting quality were the fifth important quality were identified as a selection and evaluation criteria of improved bread wheat varieties in the study area. Out of the total 150 samples, adopters were 99(66%) and non-adopters were 51(34%). In determining factors influencing probability of adoption through logit analysis, distance of DA-

xiv

office from farmers home, leadership status of respondent farmer, market accessibility and sample farmers experience in extension were identified and (b) intensity of adoption through tobit analysis, house hold sex, age, education, health status, off-farm income, distance of DA office, size of farm land holding and extension service were identified. To enhance probability and intensity of adoption, closer placement of DAs, encouragement of those farmers having less education, female farmers, popularization of improved varieties, improving varieties qualities of characteristics and farmers environmental, economic and social situation should get a serious consideration.

xv

1. INTRODUCTION

1.1. Background

Grain cultivation and the intensive utilization of wild grains in the horn of Africa probably began by or even before 1300 B.C. However, modern agricultural technologies and crop improvement activities to increase grain production have been introduced to the region very recently. Wheat was made beneficial to man long before the dawn of recorded history. Archeological findings and discoveries have indicated that wheat domestication and use as a human food has a long history, for at least 6000 years (Pearson, 1967); as early as 7500 B.C. (Langer and Hill, 1982); that took place between 17,000 and 12,000 B.C. (Tanner and Raemaekers 2001); and for 8000 years (Curtis, 2002).

Wheat is today, one of the most important of all cultivated plants, more nutritious of cereals and continue to be most important food grain source to human nutrition (Pearson 1967; Harlan, 1981; and Curtis, 2002) and its contribution to the human diet puts it clearly in the first rank of plants that feed the world (Harlan, 1981). World wide, wheat is used as human food, seed, livestock feed, and as an industrial raw material (Tanner and Raemaekers, 2001).

Ethiopia is the largest wheat producer in sub-Saharan Africa. Wheat is an important food crop and it is one of the major cereal crops in Ethiopia. Wheat in Ethiopia is ranking fifth in area and production after teff, maize, barley and sorghum and fourth in productivity. Ethiopia endowed with a wealth of genetic diversity, particularly for tetraploid-wheats. Nevertheless, the productivity of wheat has remained very low mainly because, improved production technologies have not been adopted by the farming community (Adugna et.al, 1991). It is grown in the highlands at altitudes ranging from 1500 to 3000 masl. However, the most suitable agro ecological zones for wheat production fall between 1900 and 2700 masl. Major wheat production areas are located in the Arsi, Bale, Shewa, Illubabor Western Harerghe, Sidamo,Tigray, Northern Gonder, and Gojam regions (Bekele et. al , 2000).

Ethiopia began to use improved varieties of bread wheat on a commercial level in 1968. Most of the early improved bread wheat varieties released were developed in Kenya. The first varieties of Mexican origin were released in 1974. The first improved durum wheat was released in 1976, and the first bread wheat varieties developed in Ethiopia were released in 1980 (Adugna et.al, 1991).

Wheat technology demonstrations have been conducted by MOA (Ministry of Agriculture), AUA (Alemaya University of Agriculture, now Alemaya University) and IAR (Institution of Agricultural Research now renamed - EARI (Ethiopian Agricultural Research Institute), since 1958. Through these demonstrations, many wheat technologies have been transferred to farmers, particularly improved wheat varieties. (Getachew et al, 2002).

In the past, variety development and recommendation was made based on on-station trial with testing and selecting of promising genotypes under high external input and optimum crop management practices with low participation of farmers. In most of the cases, varieties developed under such conditions were poor and failed to prove their superiority under on-farm conditions and farmers management practices. This could be due to differences in management levels practiced by researchers and farmers and due to the lack of farmers participation and interaction in the variety evaluation and selection processes. To fill the gap of low adoption of technologies by farmers and increase farmers participation in technology evaluation and recommendation, a participatory research approach through client-oriented research should be employed widely (Getachew et al, 2002).

1.2. Statement of the Problem

Agriculture is the main economic sector in Ethiopia, providing employment for about 85% of the population, and accounting for around 50% GDP. Despite the importance of agriculture in its economy, Ethiopia has been a food deficit country for several decades (Tesfaye, 2004). Available evidence indicates that peasant agriculture in Ethiopia is characterized by

inadequate resource endowment and traditional methods of cultivation and husbandry practices. The majority of small holders in Ethiopia have limited access to land saving agricultural innovations such as high yielding varieties, inorganic fertilizers and chemicals (FAO, 1993).

Wheat is one of the most important cereal crops grown in Akaki, the study area and in the country. It contributes to the major share of daily consumption demand of rural households. In addition, it is used as cash source for a household. Wheat is one of the major products marketed. In the area, farmers grow both the improved and local varieties. Even though there is a tremendous and continuous effort made by agricultural development workers and researchers adoption and the yield increment of improved bread wheat varieties have not reached to the required level. There fore, assessing the level of adoption and the related problems by involving and participating farmers in the study can help to get reliable information that can be useful to facilitate and fasten t the production of improved bread wheat varieties.

Most of the times, in the country as well as in the study area, development, introduction and promotion of improved bread wheat varieties and other agricultural technologies are conducted without due consideration of farmers circumstances, constraints, local environment and their participation. As a result, less achievement in adoption of improved bread wheat varieties as well as in other improved agricultural technologies has been resulted. Therefore, the information revealed in this study on the probability and intensity of adoption and on farmers evaluation and selection criteria of improved bread wheat varieties by involving farmers is believed more reliable to use as an input in promotion of improved bread wheat varieties in the study area and in other areas having similar socio-economic and geographical conditions.

1.3. Objectives of the Study

In general, the objective of this study was to know the status of adoption of improved bread wheat varieties in the study area. However, the study was focused on the following specific objectives:

1.to identify farmers evaluation criteria of improved bread wheat varieties distributed in the study area;

2 .to assess the adoption and intensity of farmers improved bread wheat varieties use in the study area; and

3 .to identify determinants of adoption and intensity of improved bread wheat varieties use in the study area.

1.4. Significance of the Study

The findings and the results of this study could help to strengthen the promotion and production of improved bread wheat in the study area and in other areas having similar geographical and socio-economic characteristics with the study area. Therefore, based on the knowledge generated from the study, policy makers, government officials, NGOs, extension personnel, researchers and other development organizations can use as an input in policy, decision-making, in their development programs and efforts, in order to accelerate, the diffusion, dissemination and yield performance of improved bread wheat varieties as well as to make the quality improvement of the varieties characteristics through their professional efforts. The findings of this study might be also helpful and serve as a springboard for further investigation and research activities. More over, research organizations and extension providers can also use as an input in their activities.

1.5. The Scope and Limitations of the Study

The most important reasons to select and to conduct this research in Akaki area were the interest of the funding agencies EARO and Debrezeit research center. The Extension Division of Debrezeit research center also recommended to be conducted this research in this area and the recommendation got acceptance by EARO head office crop research department. More over in this area there is a wide wheat production practices going on; the area is also one of the sixth on farm research and demonstration sites of Debre Zeit research center; and there is also one research station to conduct on farm verification and adaptation trial in this area. From those six research sites of Debre Zeit research center, only Akaki is the area where wheat and other highland cereal crop research activities are conducted. Since wheat is the first most important crop in this area, the farming population uses this crop as major food and income source. These are some of the basic reasons why this research was conducted in Akaki.

The other limitations of this study were the budget or financial scarcities to cover the payments requested by enumerators for data collection and by respondent farmers for the information they intended to give and for the time they spent during the interview. But to maintain the quality of data, efforts were made and some payment arrangements for respondents per interview and for enumerators per interview schedule were made.

Due to the above-mentioned problems, the study was not conducted in other wheat growing area of the country and was also constrained to cover wider areas and larger sample size even in this area. As a result the study was limited to be conducted in Akaki area and cover only two PAs (Peasant Associations) or RKA (Rural Kebeles Administration) and only 150 respondents randomly selected from the two-selected sample PAs of this area, Akaki. The other limitation of this study might arise due to its closer location to the capital city of the country Addis Ababa since some farmers in this area may not spend their full time on their farm. As a result there is a doubt that they may not provide the real information.

On the other hand, the scope of this study was limited to cover and analyze only those factors influencing adoption and intensity of adoption behavior of farmers such as farmers age, education, health, gender (sex), leadership, extension service, distance of DA office and credit providers institutions from the farmers village, market and credit accessibility farmers farming and extension experience, family size, resource endowment like land, livestock, oxen and labor source. Other factors like farmers perception, knowledge, needs and attitude towards the various characteristics of improved bread wheat varieties were not covered in this study; hence, it is required to conduct further investigations.

1.6. Organization of the Thesis

This thesis is organized into seven major parts. Part one constituted the introduction, which focuses mainly on the background, statement of the problem, objectives, significance, the scope and limitation of the study as well as the organization of the thesis. Part two deals with review of different literatures on adoption of improved technologies and factors affecting adoption and intensity of adoption of improved bread wheat varieties. Part three describes the materials and methods including a brief description of the study area, data collection procedures and analytical techniques. Part four contains result and discussion. Part five constitutes summary and conclusion of the study. The remaining parts of this thesis are reference and appendices, which are covered under part six and seven respectively.

2. LITERATURE REVIEW

2.1. Concept and Theoretical Framework of Adoption

Adoption was defined as the degree of use of a new technology in long-run equilibrium when a farmer has all the information about the new technology and its potential. Adoption refers to the decision to use a new technology, method, practice, etc. by a firm, farmer or consumer. Adoption of the farm level (individual adoption) reflects the farmers decisions to incorporate a new technology into the production process. On the other hand, aggregate adoption is the process of spread or diffusion of a new technology within a region or population. Therefore, a distinction exists between adoption at the individual farm level and aggregate adoption, within a targeted region or within a given geographical area (Feder et. al., 1985).

If an innovation is modified periodically, the adoption level may not reach equilibrium. This situation requires the use of economic procedures that can capture both the rate and the process of adoption. The rate of adoption is defined as the proportion of farmers who have adopted new technology overtime. The incidence of adoption is the percentage of farmers using a technology at a specific point in time (e.g. the percentage of farmers using fertilizer). The intensity of adoption is defined as the aggregate level of adoption of a given technology, e.g., the number of hectares planted with improved seed. Aggregate adoption is measured by the aggregate level of use of a given technology with in a geographical area (Feder et. al., 1985).

Diffusions scholars have long recognized that an individuals decision about an innovation is not an instantaneous act. Rather, it is a process that occurs over a period of time and consists of a series of actions (Rogers and Shoemaker, 1971). Adoption is not a sudden event, but a process. Farmers do not accept innovations immediately; they need time to think things over before making a decision. There are several well-known schemes for explaining the adoption process. A popular one involves awareness, interest, evaluation, trial and adoption; and

another; knowledge, persuasion, decision and confirmation (Adams, 1982; and Rogers and shoemaker, 1971).

They elaborated these four stages of adoption process as follows (1) Knowledge: - when the individual learns of the existence of the innovation and again some understanding of its function. (2) Persuasion: - when the individual forms a favorable or unfavorable opinion of the innovation (3) Decision: - when the individual engages in activities that lead to a choice between adoption and rejection, (4) Confirmation: - when the individual makes a final decision to accept or abandon the innovation. According to their expression, it is well known that some people are more innovative (responsive to new ideas) than others. Therefore, adopters have been subdivided in to categories on the basis of the relative time they take to adopt innovations (innovators, early adopters, early majority, late majority and laggards). Innovativeness generally can be related to other personal characteristics; background, social status, affiliations, attitudes, etc. Research has shown that adoption of innovations often follows a bell shaped or normal curve when plotted against time.

Innovations are new methods, ideas, practices or techniques, which provide the means of achieving sustained increases in farm productivity and income. It is the extension workers job to encourage farmers to adopt innovations of proven value. It is an idea or object

perceived as a new by an individual. The innovation may not be new to people in general but, if an individual has not yet accepted it, to that person it is an innovation. Some Innovations originate from agricultural research stations, others from farmers. Innovations relate to objects social acts and abstract ideas. Generally, innovations may be classified in to technical and social innovations (Adams, 1982).

Innovations are also classified into process and product innovation (Adams, 1982). A process innovation is an idea input to a production process, while product innovation is a material input to the production process. The term innovation and technology are used interchangeably. Adoption and diffusion are distinct but inter-related concepts. Adoption refers to the decision to use a new technology, method, practice, etc. by a firm, farmer or consumer. The concept of

diffusion refers to the temporal (of time) and spatial (of area) spread of the new technology among different economic units (firms, farmers, and consumers). These two concepts defined by many researchers belongings to different academic disciplines (Legesse, 1998).

Among the many definitions that suggested by Rogers (1983) has been used in several adoption and diffusion studies. He defined aggregate adoption (i.e. diffusion) behavior as the process by which a technology is communicated through certain channels over time among the members of a social system. This definition encompasses at least four elements:(1) Technology, which represents the new idea, practices or objects being diffused (2) Channel of communication, which represents the way information about the technology flows from change agents (such as extension workers or technology suppliers to final users or adopters, (3) Time which represents the period over which a social system adopts a technology and (4) Social system, which is comprised of individuals, organizations or agencies and their adoption strategies (Kundson, 1991, in Legesse, 1998). Rogers defined adoption as use or non-use of new technology by a farmer at a given period of time. This definition can be extended to any economic units in the social system (Legesse, 1998).

With regard to the measurement of intensity of adoption, a distinction should be made between technologies that are divisible and technologies that are not divisible. The intensity or extent of adoption of divisible technologies can be measured at the individual level in a given period of time by the share of farm area under the new technology or by the per hectare quantity of input used in relation to the research recommendation (Legesse, 1998).

Feder et al., (1985) suggested that this measure might also be applied at the aggregate level for a region. In the case of non-divisible agricultural technologies such as tractors and combine harvesters, the extent of adoption at the farm level at a given period of time is dichotomous (adoption or non-adoption) and the aggregate measure becomes continuous. Thus, aggregate adoption of lumpy technology can be measured by calculating the percentage of farmers using the new technology within a given area.

There is also a great difference between the agricultural sectors of developing and developed countries. Agriculture in developing countries is heavily dependent on natural phenomena, while the effects of natural factors are, to some extent, mitigated by the application of modern technology and improved weather forecasting systems in developed countries. Moreover, farmers in developing and developed countries do not face the same types of constraints and opportunities. Therefore, conclusions concerning technology adoption cannot be drawn for agriculture in developing countries based on experience of the agriculture in developed countries (Legesse, 1998).

All individuals in a social system do not adopt a technology at the same time. Rather they adopt in ordered time sequence. Based on the time when farmers first begin using a new technology five possible adopter categories can be identified in any social system: innovators, early adopters, early majority, late majority, and laggards (Rogers, 1962 and Rogers and Shoemaker, 1971, in:Legesse, 1998).

In describing the characteristics of these groups, Rogers (1962, cited in: Largesse, 1998)) suggested that the majority of early adopters have expected to be more educated, venturesome, and willing to take risks. In contrary to this group the late adopters are expected to be less educated, conservative, and not willing to take risks. A practical aspect of the classification of adopters into adopter categories has been in the field of deliberate or planned introduction of innovation. Nevertheless, the usefulness of this categorization is restricted as there is evidence indicating possible movement from one category to another, depending on the technology introduced (Runguist, 1984, in: Legesse, 1998).

Attention has also been given to explain the mode (or approach) and sequence of agricultural technology adoption. Two approaches seem to appear in agricultural technology adoption literatures. The first approach emphasizes the adoption of the whole package and the second one stresses the sequential or stepwise adoption of components of a package. The technical scientists often advocated the former approach while the latter has advocated by the field practitioners, especially by farming system and participatory research groups. There is a great

10

tendency in agricultural extension programmers to promote technologies in a package form whereby farmers are expected to adopt the whole package. Experiences of integrated agricultural development projects such as CADU, in Ethiopia, however, show that this approach has not brought needed technical change because of resource limitation (Legesse, 1998).

The adoption of agricultural innovations in developing countries attracts considerable attention because it can provide the basis for increased production and income. That means farmers will adopt only technologies that suited their needs and circumstances (Nanyeenya et. al., 1997). In efforts to increase agricultural productivity, researchers and extension staffs in Ethiopia have typically promoted a technological package consisting of a number of components. However, because of capital scarcity and risk considerations, farmers are rarely adopting complete packages (Million and Belay, 2004).

Agricultural development implies the shift from traditional methods of production to new, science-based methods of production that include new technological components and/or even new farming systems. For farmers to adopt these new production technologies successfully, they must first learn about them and how to use them correctly in their farming system (Swanson and Claar, 1984).

The transfer of technology approach grounded on the diffusion model focuses on technology generation by scientists then handed over to extension to pass on to farmers. In this model, farmers considered as passive receivers and extension as technical delivery conveyer belts. The new roles of farmers, the new participatory approaches and methods and the new learning environments all imply new roles for agricultural scientists and extension (Kiflu and Berhanu, 2004). They have to learn from farmers and develop technologies that can serve the diverse and complex farmers situations. Farmers show great interest to evaluate the promising varieties that could be suitable to their local situation (Mergia, 2002).

11

Adoption of a new technology must be preceded by technology diffusion, e.g., the act of making new technology known to the potential adopters. Diffusion is therefore the link between research and development and adoption. Effective diffusion is an essential but not sufficient condition for adoption. The farmers of a given target category must not only be made aware of an available technology, they must also be convinced that adoption is in their best interests and above all, they must be able to adopt the proposed technology (Andersen, 2002 and Arnon, 1989). Adoption studies in developing countries started two to three decades ago following the green revolution in Asian countries. Since then, several studies have been undertaken to assess the rate, intensity and determinants of adoption. Most of these studies focused on the Asian countries where the green revolution took place and was successful. In Africa, new agricultural technologies have only been introduced recently (Roy, 1990 and Rukuni, 1994, cited in: Legesse, 1998).

The effectiveness of agricultural extension work highly depends on the availability of extension professionals who are qualified, motivated, committed and responsive to the everchanging social, economic and political environment. Adoption of technology by farmers can be influenced by educating farmers about improved varieties, cropping techniques, optimal inputs use, price and market conditions more efficient methods of production management storage, nutrition, etc. (Anderson and Feder, 2002).

2.2. Empirical Studies on Adoption

Feder, et al (1985) estimated the relationships among technologies already adopted by maize growing farmers in Swaziland by using factors analysis. They found farmers adopted the technologies investigated in three independent packages (1) improved maize verity, basal fertilizers and tractor ploughing (2) top dressing fertilizers, and chemical (3) planting date, and plant population (density). These empirical findings do not support a sequential or stepwise adoption process. They reported that farmers in Swaziland tend to adopt a package of technologies and the social system adopts a technology, which is comprised of individuals, organizations, or agencies with their adopting constraints.

12

Jha et al (1991) made a study in eastern province of Zambia to evaluate how small holders respond to interventions that promote the use of improved biochemical (seeds and fertilizers) and mechanical (animal traction) technologies. In eastern province, farmers were adopting both labor saving and yield increasing techniques. Agro ecological factors play a critical role even with in a relatively small region. Farm size affects the use of fertilizer in eastern province of Zambia. The age and gender of head of the household significantly influence the adoption of hybrid maize. Heads of household that are older and females are not likely to adopt hybrid maize. Extension makes only a small contribution to the process of adopting and diffusing technology. It contributes to specialized commodity-oriented programs but not to maize, the main crop.

A study conducted in Sierra Leone by Adesina and Zinnah (1992) on technology characteristics, farmers perceptions and adoption decision using tobit model analysis. The result of tobit analysis demonstrated that apart from age, farm size, extension service and experience were positively related to adoption decisions.

The study done by Legesse (1992) in Aris Negelle area on adoption of new wheat technologies indicated that experience, credit, expected yield, expected profit, cash availability for down payment, participation in farm organization as a leader, and close exposure to technology were the factors which significantly influenced the probability of adoption of improved varieties and intensity of adoption of fertilizer and herbicide. He found that the probability of adoption of improved varieties increases with an increase in farming experience. Farmers with higher experience appear to have often full information and better knowledge and were able to evaluate the advantage of the technology is considered. The study also revealed that credit is a crucial factor affecting the probability of adoption of improved varieties. And the quantity of fertilizer farmers applied was found to be sensitive to access of credit. The coefficient of the variable expected yield was significant and shows the intensity; of fertilizer application on wheat and maize is related to its expected profitability. In his study, farm size was not found to be important factor affecting probability of adoption of improved varieties and intensity of fertilizer application. However, the variable farm size per person significantly and negatively

13

influenced the intensity of herbicide adoption for weed control in wheat. The role of direct extension visits (as represented by frequency of visit by extension agent) was not found to be a significant factor affecting adoption. This can be attributed to the limited frequency of direct extension agent visits to non-contact farmers. On the other hand, the variable close exposure to technology was found to significantly affect the probability of adoption of improved varieties.

Chilot (1994) conducted a study using Probit and Tobit analytical models to identify factors influencing the three dependent variables such as rate of adoption of new wheat varieties, intensity of fertilizer adoption and intensity of adoption of 2.4.D weed control chemical under the title of adoption of new wheat technologies, by hypothesizing of eleven independent variables and this research result showed that several factors were affecting adoption of new wheat technologies in two extension centers, Wolemera and Addis Alem areas. As the study revealed, access to timely availability of fertilizer, perceived related profitability of the improved variety, number of extension contacts and wealth position were positively and significantly related to new improved wheat variety adoption. None of the household characteristics were significantly related to variety adoption. With regard to the intensity of fertilizer use, timely availability of fertilizer, number of livestock owned and perceived profitability were positively and significantly related to the intensity of fertilizer use. Literacy, wealth position of the farmers, exposure to improved technology and timely availability of 2, 4 D were significantly and positively related to the intensity of 2, 4, - D use. The result showed that only one variable, distance of extension agent office from farmers home was the common influencing independent factor affecting inversely and significantly all the three dependent variables namely the rate of adoption of new wheat varieties, intensity of fertilizer adoption and intensity of adoption of 2.4.D weed control chemical.

Another study made by Bisadua and Mwangi (1996) in southern high lands of Tanzania Mbeya region showed that farmers were at various components of the recommended package of improved maize production. Besides, farmers have adopted these components gradually. The four major factors that contributed this gradual adoption were cost of technologies, environmental factors, timely availability of inputs, and source of information of new

14

technologies. Technologies which require little cash out lay such as row planting are easily taken up by farmers because it was less costly and had an added advantage of simplifying weeding. Environmental stress affected the adoption of some of the recommendations especially where maize is planted during the day season, which utilizes residual moisture in the soil. Farmers who dry planted their maize did not apply basal fertilizer. This might be because of the fear of scorching their maize seed due to low soil moisture. Others did not perform the second weeding, apparently because rigorous weed germination will be suppressed by the moisture conditions. Lack of timely availability of inputs was widely cited as constraint to use them. In availability of improved maize seed was considered bottleneck to its use.

Other study in Sudan highlands also suggested unavailability of inputs as major constraint to their use (Lyimon and Temu, 1992, In: Bisauda and Mwagi, 1996). Giving the extension service is charged with the responsibility of extending information on new technologies; their low rates of contact with farmers may be acting as a constraint to the use of these technologies (Bisauda and Mwangi, 1996).

A study conducted on factors affecting adoption of maize production technologies in Bako area, Ethiopia, by Asfaw et al., (1997) using logit analytical model by hypothesizing seven independent variables to influence three dependent variables namely adoption of fertilizer, improved variety and row planting. The result of the model analysis showed that only one variable, namely extension activities was significantly influence adoption of improved varieties. Among the seven proposed independent variables, only two independent variables affected fertilizer adoption and two independent variables influenced the adoption of row planting significantly. In addition, only one independent variable namely extension activity, was influence all the three dependent variables in common among the seven proposed independent variables. Though, all of the hypothesized independent variables were expected to influence significantly the identified three dependent variables. Among hypothesized variables, one independent variable namely credit was excluded from further model analysis due to its less important to affect adoption of row planting as mentioned as a possible reason

15

in the report of the study but included in model analyses of fertilizer and improved variety adoption.

An assessment of the adoption of seed and fertilizer packages and the role of credit in small holder maize production in Kakamega and Vihiga districts, Kenya by Salsya et al (1998) showed that the age of household head, primary education, cash crop area, farm size, and credit were not significantly correlated with adoption. Secondary education, cattle ownership, use of hired labor, and access to extension significantly influenced the adoption of improved maize varieties. The use of hired labor and manure, cattle ownership and membership in an organization were significant factors affecting the adoption of fertilizer. Livestock ownership serves as a source of wealth to purchase inputs that affect significantly and positively in adoption of fertilizer. Farmers who use manure had lower probabilities of adopting fertilizer. Membership in an organization increased the likelihood of adopting fertilizer. Farmers who belong to an organization are likely to benefit from better access to input and to information on improved farm practices.

Farmers' participation in leadership of farmers' organizations seems to be the best prediction of adoption behavior of the farmer characteristic variables. The relationship between technologies may be independent, sequential or simultaneous and the patterns of adoption follow the domicile Period of time by the share of farm area under the new technology or by the per hectare quantity of input used in relation to the research recommendation (Rauniyer and Goode 1996, in:Legesse ,1998).

Another study conducted on adoption of soil conservation technologies in philppines uplands of two areas namely Cebu and Claveria by Lucila, et al. (1999) using probit model by hypothesizing nine independent variables to influence adoption of soil conservation practices in these two areas. The result showed that in Cebu and Claveria in each area only three independent variables were significant to influence the dependent variable adoption of soil conservation practices. But the common significant independent variable among those nine

16

hypothesized independent variables to affect the mentioned dependent variable in the two study areas was only one variable, which was the percent of land slope.

A study conducted on adoption of wheat technologies by Bekele et al (2000) in Adaba and Dodola woredas of Bale highlands of Ethiopia using tobit analysis model demonstrated that adopters of improved wheat technologies were younger, more educated, those who had larger families and farm, hired more labor and owned more livestock.

Another study conducted by Tesfaye and Alemu (2001) on adoption of maize technologies in northern Ethiopia shows that applying chemical fertilizer, access to credit, access to extension information, distance from development center, distance from market center and family size were factors affecting adoption of improved maize positively and significantly. The level of education was found to have no significant influence on the adoption decision of farmers for improved maize. Attendance of field day and access to extension information were negatively and significantly related to the adoption of decision of chemical fertilizer use. In this study, farm size, though positive, was not found to have a significant influence on the adoption of chemical fertilizer. Access to credit and use of improved maize are the most important factors found to positively and significantly influence the adoption decision of chemical fertilizer.

A study conducted by Tesfaye et al, (2001) on adoption of high yielding maize variety in maize growing regions of Ethiopia indicated that use of chemical fertilizer, access to credit, attendance of formal training on maize production and other agricultural techniques, access to extension information, distance to the nearest market center, family size and tropical live stock unit had significant and positive influence. On the side of adoption decision of chemical fertilizer, access to credit, level of education, farm experience, total farm size, use of improved maize, use of community labor were found to have a significant and positive influence.

Another study conducted on adoption of improved wheat technologies by small scale farmers in Mbeya district, southern highlands of Tanzania by Mussei et al, (2001) clearly indicated

17

that farm size, family size, and the use of hired labor significantly influenced the probability of land allocation to improved wheat varieties. Farm size, family size, the use of hired labor and credit significantly influenced the probability of fertilizer use. A unit increased in farm size among adopters decreased the probability of adopting fertilizer by 2.4% family size by 9.7%, use of hired labor by 4.7% credit by 5.9% has increased the probability of adopting fertilizer among adopters. Credit enables farmers to purchase inputs and increased the probability of adopting fertilizes among adopters by 5.1 %.

Another study conducted by Lelissa and Mullat (2002) on determinants of adoption and intensity of fertilizer use in Ejera district West Shoa Zone, Ethiopia, using probit and tobit analytical models and the result of probit model analysis indicated owning of draught power, credit access, owning of large farm size, access to extension service affect adoption of fertilizer positively. But age affects negatively and education has no significant influence on fertilizer adoption. Regarding the result of Tobit model analysis on the determinants that influence the adoption and intensity of fertilizer use; family size, education, draught power, access to credit and extension service have influenced positively.

A study conducted by Tesfaye (2004) on adoption of in organic fertilizer on maize in Amhara, Oromia, and southern regions, shows that on the adoption of chemical fertilizer, farm experience, access to credit, use of improved crop varieties, use of farm yard manure, family size, level of education, total farm size were considered significant. The larger the farm size the greater the probability of adopting of chemical fertilizer. In this study, family size was found to have a positive and significant impact on the adoption decision of chemical fertilizers. Access to credit can relax the financial constraints of farmers and allows farmers to buy inputs. The result of the study revealed that credit availability has significantly and positively impacted up on chemical fertilizer adoption. Educational level has increased the probably of adoption of chemical fertilizer. Use of improved variety of crops also influenced the decision of farmers to use chemical fertilizer positively and significantly.

18

A study conducted in Gumuno area of southern Ethiopian by Million and Belay (2004) to identify determinants of fertilizer use (adoption decision) shows that age of household head, access to credit, frequency of development agent visit, livestock holding and off- farm in-come influenced the adoption of fertilizer positively and significantly.

A study conducted by Adam and Bedru (2005) on adoption of improved haricot bean varieties in the central Rift Valley of Ethiopia, using logistic analytical model found that sex, total livestock unit, credit, and participation in extension service affect adoption of haricot bean varieties but dependent family members and land size affect negatively and significantly.

Another study was conducted in the central highlands of Ethiopia, on adoption of chickpea varieties by Legesse, et al, (2005) using logistic analytical model. The result of analysis demonstrated that the level of education of household head, farm size, access to extension service proportion of chick pea area and access to seed affect positively and significantly the adoption of chick pea varieties.

2.3. Farmers Participation in Agricultural Technologies Development and Evaluations

According to Hanson (1982), farmers, millers, bakers, and consumers differ in their concepts of desirable qualities in wheat. To farmers, a variety of wheat has quality if it resists diseases, matures at the proper time, doesnt topple over before harvest, and gives a good yield of plump grains without shattering (grain falling, to the ground before harvest). The miller is concerned with the grain. The Kernels should be uniform, the grain should be free of foreign matter, the moisture content should be low and the protein content high and the yield of flour per 100 kg of wheat should be high. The baker who produces leavened bread looks for flour that produces dough with desirable characteristics, the dough should be able to hold gas bubbles and yield a large loaf with good internal texture and color. The consumer does not see what grain before it is milled, but he or she has strong preferences regarding the appearance, texture, aroma and flavor of the breads, biscuit, cakes and other products that trace their

19

character partly to the wheat Kernels. These different viewpoints of farmers, millers, bakers, and consumers must all be considered to raise the wheat production.

It is useful to examine several features of small holders farming system in Ethiopia, and in the third world in general, and their implications for agricultural researchers. First, farmers are economically rational, that is, they adopt new practices that are in their interests and reject those that are not. When farmers resist a new technology, it is probably because it is not compatible with their objectives, resources or environment and not because of their backward ness, irrationality or management mistakes. Moreover, the small holders farming system is complex; small holders allocate their limited resources of land, labor and capital among many enterprises in a manner determined by their agro-ecological and socio- economic environment.

Farmers need to compromise enterprises to increase productivity. Farmers consider both technical and socio- economic aspects when deciding whether to use a new technology. Researchers are need to obtain an accurate and balanced assessment of the performance of the varieties, using both scientific and farmers own criteria. Farmers rarely adopt complete packages all at once, that is, complete set of recommended technological components concerning how to mange an enterprise. Instead, farmers usually use a step-by step approach, testing components individually and incorporate the successful ones into their system. Therefore, researchers need to evaluate new technologies individually or in simple combinations under farmers own management conditions. The greater the farmers participation in the designing and testing of a technology, the greater is the chance that they adopt it (Franzel, 1992).

A study conducted in West Shoa Zone, Ambo Woreda Birbisa and Cherech servicecooperative by Ethiopian Rural Self Help Association (ERSHA) in (2000) to evaluate bread wheat technologies on the farmers farm condition using farmers criteria. According to the study, farmers have formulated the criteria to evaluate the bread wheat varieties at different growing stages. The criteria formulated by farmers were crop stand (uniform germination, strong and healthy, deep green and many tillers), flowering (uniform flowering), heading

20

(panicle size, number of spike lets per head, resistance to lodging, frost and disease), yield (superior in yield, easy to thresh, stored for long time), Grain quality, (size, color, full body), baking (dough quality, baking quality and taste). Farmers evaluation criteria judging varieties during vegetative growth stage in order of importance were: tillering capacity (many tillers per plant), head size (panicle size, head length), frost and disease tolerant (healthy leaf and shoot, uniform germination and crop stand, resistance to lodging and shattering. Farmers evaluation criteria for judging grain quality characteristics were: yield per unit area, grain size (fill body, no shrink seeds and deformed seeds), baking quality (dough quality and being good bread), and color (important for market) and easy to thresh.

These days, it is well known that farmers participation in agricultural research and development processes are increasingly improved by realizing that the socio- economic and agricultural conditions of small-scale farmers are too complex, diverse and risk prone. Conventional approaches, which are well known by station-based researches followed by topdown technology transfer system, are not often adopted in a sustainable manner. Hence, building a partnership and management with farmers is needed throughout the cycles of diagnosis, experimentation and technology dissemination. This increases the understanding of the opportunities and constraints faced by farmers on top of their technical knowledge. This in turn enhances the prospects of technological development and its adoption rates (Mergia, 2002).

It was realized that farmers have their own priorities in their production strategy and often accepts those technologies, which they consider as most advantageous to their production system. Close engagement with farmers through the cycle of diagnosis, experimentation and dissemination increases understanding of conditions, of the opportunities and constraints farmers face, and of their own technical knowledge. The package-testing program also helped to get the assessments and evaluations of the technologies from the beneficiaries themselves. The approach has considerably contributed in increasing the understanding of the biological researches towards the farmers complex and linked circumstances and constraints. It has also

21

contributed in improving the linkage between research, extension and farmers as compared to previous approaches (Mergia, 2002).

A study conducted on the use of B.B.M (Broad Bed Maker) technology on vertisol, Sheriff (2002) shows that farmers in the study area got an opportunity to identify and select different crop varieties and grow the crop they preferred that can best meet their needs, interests and the corresponding agronomic practices of their specific agro-ecological conditions. Farmers do not operate according to the assumption of policy makers and scientists. It, moreover teaches us that agricultural knowledge varies and is accorded different social meanings depending on how it is applied in the running farms. This leads to differential patterns of farm management style, cropping patterns and levels of production. Farmers are heterogeneous and they are indeed knowledgeable and capable actors who consciously pursue various objectives.

Technological patterns of development should refer back to various resources and farmers capacity. The achievement of these objectives is influenced by the images they have of various aspects involving institutions. These call for the negotiation of values and resulting unintended consequences that could be referred to as counter development. We therefore, need to learn that technology transfer has to address different farmers needs, perceptions and strategies. We need to intervene with redesigning their use of technologies (Sherif, 2002).

Technology is not always a product of scientific institutions. Human beings are inherently capable of modifying their environment in the process of adaptation, where by technology is created and subsequently utilized. The struggle between the environment and people never stops, though under some circumstances, a long time may pass before intended changes are achieved. For various reasons, some societies adhere to certain technologies for centuries where as others pass comparable level of technology in a relatively shorter period of time. For instance, the revolution of farming tools for different operations in developed countries and the stagnation of the same in a developing country such as Ethiopians explain this observation (Tesfaye, 2003).

22

2.4. Conceptual Framework of the Study

In general, it could be inferred that agricultural technology adoption and diffusion patterns are often different from area to area. The differences in adoption patterns were attributed to variations in agro-climatic, information, infrastructures, as well as environmental, institutional and social factors between areas. Moreover farmers adoption behavior, especially and in lowincome countries, is influenced by a complex set of socio- economic, demographic, technical, institutional and biophysical factors (Feder et al, 1985).

Understanding and considering these factors when analyzing and interpreting farmers response to agricultural innovations has, there fore, become important both theoretically and empirically. Adoption rates were also noted to vary between different groups of farmers due to differences in access to resources (land, labor, and capital) credit, and information and differences in farmers perceptions of risks and profits associated with new technology. The direction and degree of impact of adoption determinants are not uniform; the impact varies depending on type of technology and the conditions of areas where the technology is to be introduced (Legesse, 1998).

Farmers decision to adopt or reject new technologies can also be influenced by factors related to their objectives and constraints. These factors include farmers resource endowments as measured by (1) size of family labors, farm size and oxen ownership, (2) farmers socio economic circumstance (age, and formal education, etc) and (3) institutional support system (available of inputs) (CIMMIYT, 1993).

In many developing countries it has become apparent that the generating new technology alone has not provided solution to help poor farmers to increase agricultural productivity and achieve higher standards of living. In spite of the efforts of National and International development organizations, the problem of technology adoption and hence low agricultural productivity is still a major concern (CIMMIYT, 1993).

23

The inability of farmers to achieve high yield levels has been blamed on many different sources on extension services side, for not properly disseminating the research stations technologies, on input supply systems side, for failing to make the new technologies available, on policy decision makers side, for making the new technologies unprofitable to use due to policy distortion and on farmers themselves, who are alleged (assumed) to be too conservative. However, many studies point to another cause of low adoption rate the research center recommendations that are irrelevant to the small farmers priorities, resource constraints, and the physical, cultural and economic environment (Winkelmann, 1977, in: Mulugetta et al, 1994).

On agricultural technology adoption and diffusion determinant factors in different countries across the world, Africa including Ethiopia, several and various studies have been conducted and many researchers have obtained various findings. The researchers lack of understanding of the farmers problem and the conditions under which they operate may result in the development of inappropriate technologies and low rates of technology adoption (Fresco, 1984, in: Mulugeta et al, 1994).

In this study efforts were made to revealed factors affecting adoption and intensity of adoption, the pattern and direction of adoption of improved bread wheat varieties (part of agricultural technologies) that varied according to farmers resource endowment, environmental situations, technological development, personal characteristics, accessibilities to different services such as credit, extension, information market and the importance, suitability, management and cost of the technologies.

Moreover literatures, practical experiences and field observations have confirmed that technologies adoption by farmers can be fasten, enhanced and make sustainable by understanding those factors influencing the pattern, degree and direction of adoption and by designing and establishing technologies diffusion and adoption pattern strategies through

24

farmers empowering, making farmers access to infrastructure, information, technologies, credit, field support how to utilize new technologies.

Other factors should be also included in agricultural technologies disseminations and adoption. Farmers participation in technologies development, selection and dissemination strategies as well as result evaluation should be considered, because farmers have a long year of farming and environmental experience. The need and interest of farmers towards agricultural innovations also varies depending on farmers farming environment, their belief, experience, economic status and their personal background and characteristic. Therefore, disseminating improved agricultural technologies without consultation of farmers most probably ends with failure.

Several literatures, practical experiences and observations of the reality have been showed that one factor may enhance adoption of one technology in one area at one time and may hinder it in another situation, area and time. Therefore it is difficult to develop a one and unified adoption model in technology adoption process because of the socio economic and ecological variations of the different sites, and the various natures of the determinant factors. Hence, the analytical framework presented in the below figure shows the most importance variables expected to influence the adoption and intensity of adoption of improved bread wheat varieties in the study area, Akaki.

25

Asset endowment and other income source - Livestock -Farm land - Off-farm

Institutional variables - Extension service - Credit access -Market access - Distance of extension office -Distance of credit provider Institutions

Decision to adopt improved bread wheat and to increase size of farm land for improved bread wheat production

Household socioeconomic characteristics - Sex - Age -Health - Education -Experience in extension

Labor Sources -Oxen -Labor source -Family size

Note = the above Figure shows the chart of conceptual frame work of the study

26

3. RESEARCH METHODOLOGY

It is well known, that, there are two research methodologies classified under the broad headings: the qualitative and quantitative research methodologies.. Methods are the tools of data generation and analysis techniques practically; methods are the tools of the trade (job) for social scientists and are chosen on the basis of criteria related to or even dictated by the major elements of the methodology in which they are embedded, such as perception of reality, definition of science, perception of human beings, purpose of research, type of research, type of research units and so on.

As many people described the basic objective of a sample is to draw inferences about the population from which such sample is drawn. This means that sampling is a technique, which helps us in understanding the parameters or characteristics of the universe or population by examining only a small part of it. Therefore, it is necessary that the sampling technique be reliable. (In general, a study on relatively small number of units, are the sample, should be representative of the whole target population. Sampling is, thus, the process of choosing the units that could be included in the study, determine the sample size and the sample selection procedures. A sample design is a definite plan, completely determined before any data are collected for obtaining a sample from a given population. In this study under this chapter the study area description and sample farmers demographic, resource ownership and institutional services has conducted.

3.1. Description of the Study Area (Akaki)

3.1.1. Location, relief and climate

The socio-economic and environmental factors of the area play a great role for better performance of any activity done in that particular area. There fore it is highly valuable to

27

describe the area where the activity is planed to be under taken. . In addition, its accessibility and the budget constraint of the research were some of the factors to fix and conduct this research in this area.

This research activity was decided to undertaken at Akaki area, which was selected by its wide growing and demonstration of improved wheat crop varieties and wide utilization of other improved agricultural technologies in this area The reasons to conduct this research, in Akaki area due to wide wheat production practices and high-improved agricultural inputs utilization as well as wide demonstration practices on agricultural inputs applications and utilizations in this area.

Akaki, located at South East of Addis Ababa and it is the rural part of Addis Ababa and Akakikaliti sub-city, one of the sub-cities of Addis Ababa. It is bounded by Oromia region to the east and southern part. The study area, Akaki, constituted 9 Peasant Associations (PAs) or Rural Kebele Administrations (RKAs) in the mean time when this research survey was carried out. But at the end of this research survey, these Rural kebele Administrations /Peasant Associations were reorganized into four reduced number of Rural Kebele Administrations as a result of a new restructuring program of Addis Ababa administration.

The agro-ecological zone of the study area, Akaki, is 100% high land and its altitude ranges from 2100 2300 masl. In the study area there are different soil types. The most important and dominant soil type in area coverage is heavy ver ti sol or black clay soil. Except some few hilly landscapes of the study area, Akaki, virtually is plain.

Therefore, soil erosion by water is not a problem in the study area. But water logging is a very serious problem resulted from its flat landscape.

28

Table 1.The Livestock and crop types in the study area No. Types of crop grown in the study area Land Coverage in (Ha) 3580 1930 1560 90 604 80 80 201 44 199 30 104 42 4360 0.70 2.38 0.96 100 Poultry Bee-in-Hive 17807 No data Land coverage in percent 82.1113.85 Types of Livestock reared in the study area Cattle Oxen (Cow and Others) Sheep Goat Pack Animals (Horse, Mule And Donkey) 13211 10380 3064 Number of Livestock 17269 4058

Cereal crops Wheat Teff

Others Pulse crops Faba bean peas Chick pea lentils

5012

3 4 5

others Oil crops Vegetables Others (Fenu greek) Total

(Source: Akaki Agricultural Unit Office, 2005)

The study area, Akaki, is characterized by the high land climate. It has the main and small rainy seasons. Farmers in the study area rely on the main rainy season known as kiremt or Meher rainy season for their agricultural production activities. There is no a practice and experience of crop production using the rain of small rainy season known as Belg rainy season and irrigation production. The small rainy season extends some times starting from January or February and ends some times in May or June. But the main rainy season is similar like otherparts of Ethiopia that extends from end of or mid of June to most of the time mid of September, according to Akaki agricultural unit office. Some of the major crops grown in the study area are Wheat, Teff, Faba bean, Chickpea and Lentils.

29

3.1.2. Agriculture and demographic characteristics of the study area

The total farming population of the study area is 14519. Of which, 7626 were male, and 6893 female. The number of household heads of the farming population is 2490 male and 265 female with a total of 2755 household heads. The average family size of the study area is 5.27 per household. In Akaki, agriculture, which includes crop and livestock production, is the main stay of the farming community like other parts of Ethiopia. The types of crops, the farm land coverage by each crop type and the types of livestock and the size of livestock population reared by farming community in the study area are summarized in Table 1.

Table 2.The land use of farmers in the study area No 1 2 3 4 5 Types of land use Cultivated land Grazing land Forest land Village and construction Others Total Coverage in (Ha) 4360 151 80 2496 22 7109 Percentage 6 1.33 2.12 1.13 35.11 0.31 100

(Source: Akaki Agricultural Unit Office)

As indicated in table 2, the larger proportion of land in the study area is used for cultivation, which is 4360 (61.33 %.). Of which the major proportion goes to cereal crop production particularly for wheat production followed by Teff production. In the study area, there is a very serious grazing land scarcity, greatly affecting the livestock production, resulted from high population pressure and extended farming practice that shrinks grazing land and compete with livestock production. The farming society in the study area used crop by product for animal feed though it is poor nutritionally.

30

Table 3.The summary of oxen ownership No. 1 2 3 4 5 6 Oxen Ownership With no oxen With one oxen With two oxen With three oxen With four oxen With five % above Total (Source: kaki Agricultural Unit Office) Number of owners 1272 243 555 139 442 104 2755 Percentage 46.17 8.82 20 .15 5.05 16.04 3.77 100

In the study area as summarized in table 3, around 46.17% farmers are with out ploughing oxen, according to the Akaki Agricultural Unit office. In the study area Akaki, there is a fuel wood scarcity resulted from unwise practice of deforestation for long time. There fore the farming population has forced to use cow dung as their source of energy for heat and food preparation. Using of cow dung for fuel can affect the utilization of compost to improve soil fertility for better crop yield.

3.1.3. Institutional services of the study area

Effective Agricultural Extension services have paramount importance to farmers to get timely advices and information on the availability, use and application of new, improved and modern agricultural inputs, technologies and practices. The Akaki Agricultural Unit office is responsible to offer agricultural extension service in the study area. Under this unit office different expert with different professions has assigned at the Unit Office level and Development Agent (DA) center level. The 9 Extension Agents / Development Agents at the center level were assigned and responsible to give extension service to the farming community. They were accountable to Akaki Agricultural Unit. But at the end of this research survey and data collection process due to the new restructuring program of Addis Ababa Administration some DAs from their center and other professionals from Akaki Agricultural

31

unit office has transferred to other unrelated duties with their professions that can affect negatively the extension services and rural development efforts of the study area.

Table 4.Improved agricultural input distribution of the study area in different years Crop Season 1997 1998 1999 2000 2001 2002 2003 2004 Total DAP Quit. 5224.5 4695.5 3759 4172 3866 3868 3994 167.5 29746.50 Urea Quit. 3000.5 2506.5 2026.5 2235.5 2034.5 2182 2340 167.5 16493 Teff Quit. 4.5 5.5 11.85 12 13.05 19.5 66.40 Pesticide K.g. 10 5 5 5 4 4 4 37 Pesticide Lit 30 30 20 10 10 4 7 111
Weedi-cide

Lit 50 76 48 33 35 5 50 297

(Source: Akaki Agricultural Unit Office)

Availability of improved agricultural inputs to use and credit service to purchase agricultural inputs is very vital for technology adoption. In the study area different agricultural inputs and credit were distributed in different time for farmers. Table 4 showed the agricultural input distribution of the study area in different years.

Table 5 showed the improved bread wheat varieties seed distribution and table 6 showed the improved credit distributions in different years to the farming community of the study area. According to the Akaki Agricultural unit office information the major inputs distributed in the study area were fertilizer (DAP and Urea) and improved bread wheat varieties as well as improved teff varieties. The major input distribution of the study area from 1995 to 2004 were, fertilizer in quintal, (DAP=29746.50 and 16493), improved bread wheat varieties 732.75 and improved Teff 66.40 quintals.

32

Table 5.Improved bread wheat seed distribution of the study area in different years No 1 2 3 4 5 6 7 8 Crop Season (Years G.C.) 1997 1998 1999 2000 2001 2002 2003 2004 Total Improved Land covered Bread Wheat (Quit.) (Ha.) 43.5 29 74.25 49.5 48 32 72 82.5 120 145.5 147 732.75 50 55 80 97 98 490.50 No.of Participant Farmers 58 99 64 100 110 160 194 196 981

(Source: Office of Akaki Agricultural Unit office)

As it is presented in table 4 and table 5, the average annual DAP fertilizer distribution were 3718.3 quintals, urea 2061.63, improved bread wheat seed 8.3 quintals and that of improved Teff was 4.625 quintals. The fertilizer distribution was almost satisfactory. But the improved bread wheat and teff seed distribution was very low quantity.

Regarding the credit distribution of the study area as presented in Table 6, the larger credit for the last seven years a total of 376478.06 Ethiopian birr for fertilizer and improved grain crop seed purchase was distributed .The larger proportion of the credit were used for fertilizer purchase. This is because farmers in the study area got credit in the form of fertilizer.

33

Table 6.Credit Distribution of the study area in various years Credit for Fertilizer and Improved seed (ET.Birr) 27,858.50 54,076.60 16,275.75 60,552.70 25,446.61 94,218.75 98,049.15 376478.06 Livestock Production (ET.Birr) 47,800 118,885 166685

No 1 2 3 4 5 6 7

Crop season (Year G.C.) 1997 1998 1999 2000 2001 2002 2003 Total

(Sources: Office of Akaki Agricultural unit office).

In the study area credit were also distributed for livestock production as showed in Table 6.For this sector of the economy credit was distributed before five years ago. The credit distribution was covered only for two consecutive years. The credit service limitations in amount, type and facilities can affect negative the adoption of improved agricultural technologies, agricultural development and the over all rural community lively hood living situation improvement.

3.2. Description of Data Collection and Data Analysis Methods and Procedures

3.2.1. Sources and types of data

It is very helpful for researchers to anticipate and think over in advance about the sources and types of data that are relevant to the research and, therefore, need to be gathered. This help to avoid confusion and unnecessary time, labor, finance and other resources wastages. The types of data, primary and secondary, were collected to answer and fulfill this research questions and objectives. All information about determinants of adoption and intensity of adoption and

34

farmers evaluation and selection criteria of improved bread wheat varieties, demographic, socio-economic, environmental situations, wheat production, credit facilities, extension service and others relevant data to the study were gathered from primary sources quantitatively through interview schedule and qualitatively through group discussion and observation.

Data also were gathered by examining secondary sources such as documents, reports and records of DAs (Development Agents), and other related agricultural offices and research centers. All these data in the process of the study were gathered using different methods and techniques based on the nature, types and characteristics of the data. Both quantitative and qualitative data were gathered through different data collection methods from primary and secondary sources.

3.2.2. Sample size and sampling techniques

This study was determined to conduct in Akaki area, which is the rural part of Addis Ababa Administration. In this study sample size was determined by taking different factors such as research cost, time, human resource, accessibility, availability of transport facility, and other physical resource accessibilities. By taking these factors into account, it was fixed to cover two Peasant Associations out of 9 PAs and 150 household head respondents from the total 2755 household head population of the study area, Akaki. Through out sample selection processes simple random selection method has employed. The two stage sampling techniques were applied in sample selection processes. First, the two Peasant Associations (PAs) or Rural Kebele Administrations (RKAs) namely Koye and Gelan-Edero involved in improved bread wheat production were selected out of nine PAs using simple random selection method.

Second 150 sample household head farmers were selected from total wheat growers of the two samples Pas. About 65 (43%) Sample farmers from Gelan-Edero PA and from Koye sample PA 85 (57%) sample farmers were selected proportionally. From Gelan-Edero PA 36 were adopters and 29 were non-adopters and from Koye PA 63 were adopters and 22 were non-

35

adopters. Out of 150 respondents (132) 88% were male and the remaining (12%) was female. From total respondents (99) 66% were adopters and the rest 51 (34%) were non-adopters. From total adopters 93% were male adopters and the remaining 7% were female adopters. Concerning non-adopters 78% were male and 22%were female as presented in Table7. All sample selection processes were carried out in pursuing of statistical procedures and with consultation of DAs, Akaki Agricultural Unit Office professionals and PA leaders of the study area.

3.2. 3. Data collection methods

Data for this study were gathered from sample household head farmer respondents through interview, group discussion and observations. Both qualitative and quantitative data were gathered using the mentioned methods. Secondary data that were relevant to this study were also gathered through examining of published and unpublished data that was gathered and organized by other bodies for other purposes .In this process care was taking in taking and selection of the relevant data suitable and relevant for this study.

3.2.3.1. Quantitative data collection methods

In this regard, primary data were collected through personal and face-to-face interview using structured and pre-tested interview schedule that were filled up by recruited and trained enumerators under the close supervision of the researcher. Totally, 150 randomly selected samples household head farmer respondents were covered under the survey. Also, secondary data were gathered by examining secondary sources such as records, reports, and research results and other documents and publications from office of agriculture, research centers and other respective offices.

36

3.2.3.2. Qualitative data collection method

In the study area primary qualitative data on improved bread wheat varieties selection and evaluation criteria of farmers were gathered through group discussion and individual discussions conducted with farmers and professionals. Researchers personal observation and transect walk were also used in this data gathering processing. Data gathering through these methods were continued to the point of saturation, to crosscheck, triangulate, elaborate and enrich the information on both qualitative and quantitative data to increase the reliability and trustworthiness of the information. The group members and individuals were familiarized to the discussion points and encouraged to forward their opinion they felt with out any reservation. In this process, recording, coding, reorganizing and arrangements, refining expanding of information was conducted.

3.3. Analytical Models

3.3.1. Logit model

Several models are available to analyze factors affecting technology adoption. The choice of one may depend up on several factors. Some of these alternative models are the discrete regression models in which the dependent variable assumes discrete values. The simplest of these models is that in, which the dependent valuable Y is binary (it can assume only two values denoted by 0 and 1).

The three most commonly used approaches to estimate such models are the linear probability model (LPM), the logit model and the probit model .The linear probability model has an obvious defect in that the estimated probability values can lies outside the normal range 0-1 range. The fundamental problem with the LPM is that it is not logically a very attractive model because it assumes that the marginal or incremental effects of explanatory variables remain constant, that is Pi=E (Y=1/x) increases linearly with x (Maddala, 1997 and Gujaratti, 1998).

37

The authors suggested that the sigmoid or S-shaped curve were very much resembles the cumulative distribution function (CDF) of random variable is used to model regressions where the response variable is dichotomous, taking 0-1values.

The cumulative distribution functions (CDFs), which are commonly chosen to represent the 01response models, are the logit (logistic CDF) model and the probit (normal CDF) model. Logit and probit models are the convenient functional forms for models with binary endogenous variables (Tohnston and Dianardo, 1997 cited in Techane, 2002).

These two models are commonly used in studies involving qualitative choices. To explain the behavior of dichotomous dependent variables we will have to use a suitably chosen cumulative Distribution Function (CDF). The logit model uses the cumulative logistic function. But this is not the only CDF that one can use .In some applications the normal CDF has been found useful. The estimating model that emerges from normal CDF is popularly known as the probit model (Gujarati, 1999).

The logistic and probit formulations are quite comparable the chief difference being that the logistic has slightly flatter tails that is the normal curve approaches the axes more quickly than the logistic curve. There fore, the choice between the two is one of mathematical convenience and ready availability of computer programs (Gujrati, 1988 cited in Techane, 2002). A relevant model offers better explanation on the underlying relation ship between adoption decision and factors influencing it. The most widely used qualitative response models are the logit and the probit models (Amemiya, 1985).

Both the probit and logit models yield similar parameter estimates and it is difficult to distinguish them statistically (Aldrich and Nelson, 1984). How ever, because of the fact that binomial logit model is easier to estimate and simpler to interpret. Therefore a logit model is used in this study to determine the relation ship between adoption decision and factors affecting the adoption of improved bread wheat varieties in the study area, Akaki.

38

The specification of the logit model is as follow: = ( = 1) = exp (Zi) 1+exp(Zi)

Where denotes the probability that the i

th

farmer will fall in the adopters class (yi=1) and

exp (Zi) stands for the irrational number e to the power of Zi. The un observable stimulus index Zi, assumes any value.

However, the Logistics transformation guarantees each corresponding value of Pi to fall inside the 0-intervals. The stimulus index Zi, also called the Log of the odds ratio, in favor of improved bread wheat varieties adoption, is actually a linear function of factors influencing adoption decision as specified hereunder: =0+11+22i + .+ i+e Z= ln [
pi 1-p

] i

Where: X1, X2, X3,...........Xp=explanatory variables 0, 1, 2, 3, 4, ,p = Logit Parameters to be estimated, ei = the error term
= 0+11i+22i+33i +.ppi +

In reality, the significant explanatory variables do not all have the same level of impact on the adoption decision of the farmers. Therefore, the impact of each significant variable on the probability of adoption was calculated by keeping the continuous variables at their mean values and the dummy variables at their most frequent values (zero or one).

The estimated coefficients of the Logit model of improved bread wheat adoption are listed in the table 20. The likelihood ratio statistics is significant at 10 percent probability level and implies that the independent factors taken together influenced improved bread wheat varieties adoption. The model correctly predicted 81.33 percent of the adopters and non-adopters.

39

3.3.2. Tobit model

Adoption studies based up on dichotomous regression model have attempted to explain only the probability of adoption versus non-adoption rather than the extent and intensity of adoption Knowledge that a farmer is using high yielding variety may not provide much information about farmer behavior because he /she may be using some percent or 100 percent of his /her farm for the new technology. Similarly, with respect to adoption of fertilizer is, a farmer may be using a small amount or a large amount per hectare area. A strictly dichotomous variable often is not sufficient for examining the extent and intensity of adoption for some problems such as fertilizer (Feder et al., 1985).

There is also a broad class of models that have both discrete and continuous parts. One important model in this category is the Tobit. Tobit is an extension of the probit model and it is really one approach to deal with the problem of censored data (Johnston and Dinardo, 1997 cited in Techane, 2002).

When examining the empirical studies in the literatures, many researchers have employed the Tobit model to identify factors influencing the adoption and intensity of technology use. For example Nykonya et al (1997) ; Lelissa (1998) ; Bezabih (2000); Croppenstedt et al. (1999) as cited in techane (2002) used the Tobit model to estimate the probability and the intensity of fertilizer use ( Adesina and zinnah ,1993; in Techane ,2002).

The econometric model applied for analyzing factors influencing the intensity of adoption of improved bread wheat varieties is the Tobit model shown in equation (1). This model was chosen because, it has an advantage over other adoption model (LPM, Logistic and Probit Models) in that, and it can reveal both the probability of adoption of improved bread wheat varieties and the intensity of use of the varieties. The Tobit model can be defined as:
Yi = X i + U i , Yi = Yi if Y* > 0 = 0 if Yi * 0 (1)
* *

i = 1,2,...n

40

Where, Yi= the observed dependent variable, in our case the land size in hectare covered with improved bread wheat variety. Yi*= the latent variable which is not observable
Xi=

vector of factors affecting adoption and intensity of fertilizer use

i= vector of unknown Parameters Ui= residuals that are independently and normally distributed with mean zero and a common variance 2. Note that the threshold value in the above model is zero. This is not a very restrictive assumption, because the threshold value can be set to zero or assumed to be only known or unknown value (Amemiya, 1985). The Tobit model shown above is also called a censored regression model because it is possible to view the problem as one where observations of Y* at or below zero are censored

The model parameters are estimated by maximizing the Tobit likelihood function of the following formula (Maddala, 1997 and Amemiya, 1985).

1
L = C

Yi >0

f(

Yi

i X i

) Y * 0 F ( i

i X i

(2)

Where, f and F are respectively the density function and cumulative distribution function of Yi*. IIYi*< 0 means the product over those i for which Yi* < 0, and IIYi*<0, and IIYi>0 means the product over those i for which Yi*>0.

It may not be sensible to interpret the coefficients of Tobit in the same way as one interprets coefficients in an uncensored linear model. Hence one has to compute the derivatives to predict the effects of changes in the exogenous variables.

1. The marginal effect of an explanatory variable on the expected value of the dependent variable is:
(Yi ) = F (Z ) i X i (3)

41

Where

i X i is denoted by z,

2. The change in the probability of adopting a technology as independent variable Xi change is:
F ( Z ) = f(Z) i X i
(4)

3. The change in intensity of adoption with respect to a change in an explanatory variable among adopters is:
(Yi / Yi > 0) f (Z ) f(Z) 2 = i [1 Z ( ) ] X i F (Z ) F(Z)
*

(5)

Where, F (z) is the cumulative normal distribution of z, f (z) is the value of the derivative of the normal curve at a given point (i.e., unit normal density). Z is the z score for the area under normal curve, B is a vector of Tobit maximum likelihood estimates and 0* the standard error of the error term.

Parameter estimates of the Tobit model for the intensity of adoption of improved bread wheat varieties (measured in terms of size of land in hectare used for growing of improved bread wheat varieties over the total wheat land in hectare) as shown in Table 22. And the results are discussed under section (5.1.2.). The Tobit model was used or applied to analyze the factors that determine the intensity of adoption of improved bread wheat varieties because the mean proportion of land allocated to improved bread wheat varieties is a continuous variable but truncated between zero and one. This model is relevant to predict the intensity of adoption of improved bread wheat varieties by farmers when the dependent variable is continuous.

3.3.3. Other Quantitative data analysis methods

In this study, in addition to econometrics models, Logit and Tobit models described in the above, descriptive statistics such as percentage, tabulation, mean, standard deviation, t-test and 2 - test data analysis methods were employed in quantitative data analysis of the study.

42

3.3.4. Qualitative data analysis method

The qualitative data analysis has conducted to strengthen the evidences obtained through quantitative data collection methods or survey method. In this study, the qualitative data obtained from the group, individual formal and informal discussions and through the researchers personal observation regarding farmers selection and evaluation criteria, their priorities of improved bread wheat varieties disseminated in the study area were summarized using the criteria established by the group members by analyzing the characteristics of these varieties to what extent these varieties satisfy and fit their needs, interests and to their environmental situations. The qualitative data were analyzed through explanation of idea, opinion, and concept explanation method. Researchers personal observations and transect walk watching were analyzed through, further explanation of the real world under observation.

3.4. Hypotheses Testing and Definitions of Variables

In this study the variables were selected and hypothesized using literatures, by considering farmers production practices, area situations and objectives of the study. In this study it was decide to concentrate the research effort and limited resources on socio economic and environmental conditions and constraints that was expected to influence probability and intensity of adoption because as the Ethiopian extension history shows that in this area extension service was provided for long years using different methods such as demonstration farmers field day, on farm field visit and support. More over, farmers in this area have better access to different information sources .As a result, were expected to have better understanding, knowledge, and attitude towards improved agricultural technologies. As a result, more emphasis was given to exogenous socio economic variables than internal variables to hypothesize, test and analyze using the Logit and Tobit Analytical models.

43

3.4.1. The Dependent variables of logit and tobit models

3.4.1.1. The Dependent variable of logit model

The dependent variable of the binomial logit models log-odds ratio is the probability of adopting or not adopting the improved bread wheat varieties which can used to identify factors determining adoption of improved bread wheat varieties is the natural logarithm of the ratio of the probability that a farmer adopts the improved varieties (Pi) to the probability that he /she will not (1-Pi). The log-odds ratio is a linear function of the explanatory variables.

3.4.1.2. The Dependent variable of tobit model

The dependent variable of Tobit model has continuous value, which should be the intensity, the use and application of the technology. As observed in different empirical studies this variable can be expressed in terms of ratio, actual figure and log form depending on the purpose of the study. For example in their study of factors influencing adoption of fertilizer, Nkonya et.al, (1997) as cited in Techane (2002) considered fertilizer applied per hectare as the dependent variable of the tobit model. Likewise Shiyani et al., (2000) as cited in Techane considered the proportion of area under chickpea varieties in their study of adoption of improved chickpea varieties. Consequently, in this study the ratio of actual land size under improved bread wheat varieties to total wheat land size was taken as a dependent variable of the tobit model.

3.4.2. The Independent variables and their definitions used in logit and tobit models

Adoption literatures provide a long list of factors that may influence the adoption of agricultural technologies. Generally these factors can be grouped into demographic personal, socio economic, physical and institutional factors (Million and Belay, 2004). There fore,

44

farmers decision to use improved bread wheat varieties and the intensity of the use in a given period of time is hypothesized to be influenced by a combined effect of various factors such as household characteristics, socio-economic and physical environments in which farmers operate. Based on Feder et.al, (1985) that extensively reviewed factors affecting adoption of agricultural technologies in low income countries, and on the brief literature review in this study the variables mentioned below are hypothesized to explain improved bread wheat varieties adoption and the intensity of the use of these varieties by the sample households.

In the course of identifying factors influencing farmer's decision to use improved bread wheat varieties, the main task is to analyze which factor influence how and by how much. It is hypothesized that adoption and intensity of adoption are influenced by the combined effect of various determinants. There fore, in the following section potential variables that are supposed to influence adoption and intensity of adoption of improved bread wheat varieties in the study area will be explained. More specifically, the following potential explanatory variables hypothesized to influence the adoption and intensity of adoption of improved bread wheat varieties in the study area, on a priori grounds is indicated below:

1. Farmers age (HHHAGE) As the farmers age increases it was expected that farmer become conservative. Then it is hypothesized that the farmers age, adoption, and intensity of adoption of improved bread wheat varieties are inversely correlated. Therefore, in this study it was assumed that the lesser age group could adopt improved bread wheat varieties more than the older age group farmers. Then, in this study farmers age and adoption are expected to relate negatively. As farmers age increase probability of adoption/intensity of adoption is expected to decrease.

2. Gender/Sex (HHHSEX) - It is hypothesized that male household headed farmers are expected to adopt improved bread wheat varieties more than female headed ones. Because it is expected that male-headed farmers have a better opportunity to access to credit and extension service. In this study gender/sex was coded if the household is male 1 and 0, otherwise.

45

Adoption/Intensity of adoption was expected to increase and correlate positively as the farmer being male.

3. House hold head Education (EDUHHH) This represents the level of reading and writing and formal schooling attended by the household headed. It is expected that educated farmhouse hold head can make better decision to adopt improved bread wheat varieties than non-educated ones. Here, education extends from read and write to attending regular school education. In this study this variable was treated as a dummy variable and has coded if the house hold head can read and write as well as attending the regular school education as 1 and 0, otherwise. Adoption was expected to correlate positively as education increases.

4. Family size (FAMILYSI) - household heads with large family size are less likely to adopt improved bread wheat varieties due to risk aversion. In this study it had expected that the family size, adoption, and intensity of adoption would have related inversely. As family size increase adoption/intensity of adoption has expected to decrease.

5. Extension-services (GEXSERVE) - the more frequent DA visit, using different extension teaching methods and training, attending demonstrations and field day can help the farmers to adopt a new technology and can also increase the intensity of adoption. If the farmers get better extension services are expected to adopt better-improved bread wheat varieties than others. In this study this variable had treated as a dummy variable in that if the farmer get extension service is coded as 1 and 0, otherwise. As extension service increase adoption/Intensity of adoption was expected to increase and correlate positively.

6. Off- farm income (HHOFFINC) - the household head that have off farm income are expected to adopt improved wheat varieties better than who have not off farm income. This variable also was treated as a dummy variable that if the farmer has off-farm income coded 1, otherwise, 0. As the number of farmers number increases to involve in off-farm work it was expected to increases adoption positively.

46

7. Access to credit (GECRSERV) - It is expected that those who have better access to credit can adopt improved bread wheat varieties than other who do not have access. Because it is expected that credit can solve the financial limitation of farmers. The variable in this study was treated as dummy variable in that, if the farmer gets credit service coded as 1 and 0, otherwise. As credit service increase adoption/Intensity of adoption was expected to increase and correlate positively.

8. Livestock ownership (TOTLIVUN) house holds that have large number of livestock are likely to adopt improved bread wheat varieties better than others who have less number of livestock Because those who have better number of livestock can have better opportunity to get credit. In this study it was assumed that livestock ownership and adoption would be related positively. As livestock ownership increases adoption/intensity of adoption was expected to increase and correlate positively.

9. Labor accessibility (OTSOLA1) those farmers who have access to labor are expected to adopt improved bread wheat varieties than those who lack labor accessibility since improved varieties required more labor. The variable has been treated as a dummy variable in that if the farmer has an access to labor coded 1, otherwise 0. As labor accessibility increases adoption/Intensity of adoption was expected to increase and correlate positively.

10. Social/leadership status of the respondent (PRTILEDE) - those farmers who have experience of leadership and better social status previously or currently are likely to adopt wheat technologies than others. Because, it is expected that they have an opportunity to get and interpret the information they get about improved bread wheat variety. The variable was coded as 1 and 0, otherwise. As the number of farmers increase to involve in leadership position adoption was expected to increase and to correlate positively.

11. Distances from extension agent office (DISDAOF1) those who are closer to extension agent are expected to adopt improved bread wheat varieties than others as a result of

47

accessibility. The variable was coded as 1 if the farmer is close to the DAs office and 0, otherwise/far. As distance of DA office increase adoption/intensity of adoption was expected to decrease and correlate negatively. As distance of DA office decrease the correlation will be vise versa.

12.Distance from input and credit supply institutions (CRINFAR1) those farmers closer is likely to adopt improved bread wheat varieties than those who are not close since they can easily facilitate and follow up the credit process. This variable was treated as dummy variable and had coded as 1 if the farmer is close, 0 if not close or if far. The far distance might affect negatively and the close distance affects adoption and Intensity of adoption positively.

13. Oxen ownership (OXTLU) those who have oxen for ploughing is likely to adopt improved bread wheat varieties since they could solve ploughing power problem. Then, oxen ownership and adoption were expected to relate positively. As the number of oxen owned by farmers increased, adoption/Intensity of adoption was expected to increase.

14. Farmers experience in any extension activities (YEXPEXTS)- Farmers who have long involvement in any agricultural extension activities is expected to use improved bread wheat varieties than with less experience. Then, this variable was hypothesized to correlate and influence positively improved bread wheat varieties adoption and the intensity of adoption.

15. Access to market (CRINMFF1) - Access to market was hypothesized to be positively related to the probability of adoption of improved bread wheat varieties in that if the house holds near to market tend to buy improved agricultural inputs including improved bread wheat seed and they can have easy access to dispose of and sell their production in the market. Therefore, the variable was treated as a dummy variable in that if the house hold has an access to market has coded as 1and 0, otherwise. As market distance increases adoption and intensity of adoption was expected to decrease.

48

16. Health status of the household head (HEALSTAT)-this is a dummy variable, which takes a value 1 if the household head was healthy and 0, otherwise. As farmers health statuses improve adoption and intensity of adoption of improved bread wheat varieties was expected to increase and correlate positively.

17. Farm land holding (SUMOWRE)-farmer who has large farm size is likely to adopt improved varieties than those who have lesser farm size. Because farmers with large farm size can distribute the yield loss risk and better land ownership serve as insurance to get credit which can use to purchase improved agricultural inputs. As farmers farmland holding increases adoption/intensity of adoption might increase and correlate positively.

49

4. RESULTS AND DISCUSSION

4.1. Analysis through descriptive statistics

This study was intended to examine the farmers evaluation and selection criteria of improved bread wheat varieties and to identify factors affecting adoption and intensity of adoption of improved bread wheat varieties in the study area Akaki as well as to know the effect of hypothesized independent variables on the dependent variables. In this section of analyses descriptive statistics such as mean standard deviation, percentage, frequency tabulation, t-test and chi-square test will be employed using SPSS- computer soft ware program.

In this study, adoption of a technology refers to a continued use of the technology on an area of land, which is large enough to contribute to the economy of the household. Here, the respondents who have cultivated improved bread wheat varieties and continued growing at least one of the distributed improved bread wheat varieties in the study area during the survey year and in any one of the years before the survey year of this study are considered as adopters. Farmers who never adopted and those who discontinued from growing of improved bread wheat varieties are categorized as non-adopters.

4.1.1. Sample Households Demographic Characteristics

In order to understand the sample households, it is very important to describe their demographic characteristics. The number of household head respondents was from two selected Rural Kebele Administrations or Peasant Associations namely Koye and GelanEdero. The sample house hold heads covered in this study from Koye PA/RKA were (76) 89.41% male and (9) 10.59 % female with a total of 85 constituted 56.67% which of the total sample and (56) 86.15% male and (9) 13.85% female with a total of 65 were from GelanEdero PA/RKA which constituted (43.33%) of the total sample as presented in Table 7.

50

Out of the total 150 respondents in the sample, adopters were 99 (66%) and non-adopters were 51 (34%). From 150 sample respondents, 132 (88%) were male and 18 (12%) were female respondents. From 132 male respondents 92 (69.70%) were adopters and 40 ((30.30%) were non-adopters. From total 18 female sample respondents 7 (38.89%) were adopters and 11 (61.11%) were non-adopters as presented in Table 7.

Table 7.Sample household heads distribution by Sex, Kebele and adoption category Adopters (99) Samples Male-sample Female-Sample Total Koye-Kebele Gelan-Edero Total Koye Kebele Male Female Total Gelan-Edero Kebele Male Female Total 32 4 36 88.89 11.11 100.00 24 5 29 82.76 17.24 100.00 56 9 65 86.15 13.85 100.00 58 5 63 92.06 7.94 100.00 17 5 22 77.27 22.73 100.00 76 9 85 89.41 10.59 100.00 N 92 7 99 63 36 99 Percent 92.93 7.07 100.00 63.64 36.36 100.00 Non-Adopters (51) N 40 11 51 22 29 51 Percent 78.48 21.57 100.00 43.14 56.86 100.00 Total (150) N 132 18 150 85 65 150 Percent 88 12 100.00 56.67 43.33 100.00

(Source: Computed from own survey data, 2005)

Out of the total 99 adopters 92 (92.93%) were male and 7 (7.07%) were female and from the total 51 non-adopters 40 (78.43%) were male and 11 (21.57%) were female. The numbers of sample household heads from Koye PA were 63 adopters and 22 non-adopters, where as from

51

Gelan-Edero PA, the number of sample household head adopters was 36 and that of nonadopters were 29 as indicated in Table 7.

In similar studies in the past, the major reasons for farmers to adopt improved and new technologies are technical, institutional, social, and economical reasons. Farmers do not adopt a technology if they are not convinced of its benefits, costs and risk associated with it. By seeing their fellow farmers, by attracting of high yield performances of improved varieties, market demand as well as DAs information and extension support farmers might urged and motivated to use and adopt improved varieties. On the other hand, the major reasons for those non-adopters might shortage of improved varieties, credit problem, or cash, land, labor or other farm resource constraints (Legesse et al., 2005).

Table 8.Marital status of respondents Marital status Married Un-married Divorced Widow/er Adopters (99) N 85 1 2 11 Percent 85.86 1.01 2.02 11.11 Non-Adopters (51) N 36 2.00 4.00 9.00 Percent 70.59 3.92 7.84 17.65 Total (150) N 121 3 6 20 Percent 80.67 2.00 4.00 13.33

(Source: Computed from own survey data, 2005)

The marital statuses of respondents are summarized in Table 8 as (121) 80.67% married, (3) 2% unmarried or single, (6) 4% divorced and 20 (13.33%) widow/er. The proportion of married respondents was much larger than the remaining marriage categories. As indicated in Table 8, the married adopters were 85.86 percent and that of non-adopters were 70.59 percent. The remaining categories of respondents constituted fewer proportions of respondents both in adopters and non- adopters.

52

Table 9.Association between adoption of improved bread wheat varieties and sex of sample household head

House hold head sex Female Male Total

Adopters 7 (7.1%) 92 ((92.9%) 99 (100%)

Non-adopters 11 (21.6%) 40 (78.4%) 51 (100%)

Chisquare

C.Coef

df

Sig.

Total 18 (12%) 132 (88%)

6.700***

0.207

0.01

150 (100%)

(Source: Computed from own survey data, 2005),

*** significant at 1%level

Table 10. Respondent farmers demographics characteristics Adoption Category Adopters Summary of statistics Minimum Maximum Range Minimum Maximum Range Minimum Maximum Range House hold head Age 19 80 61 20 80 60 19 80 61 House hold family size 1 11 10 1 10 9 1 11 10 Farming House hold head experience in extension experience 2 2 20 18 1 9 8 1 20 19 55 53 4 60 56 2 60 58

Nonadopters

Total

(Source: Computed from own survey data)

In adoption of new agricultural technologies, farmers age has an influential effect as it was observed in many adoption studies. The maximum and the minimum ages of total respondents were 80 and 19 years respectively. The adopters maximum age was 80, which was equal to the non-adopters maximum age. The minimum age of adopters was 19 years and that of nonadopters was 20 years. The age variation between maximum and minimum age of adopters, non-adopters and that of total respondents were 61, 60 and 61 respectively as presented in Table 10.

53

The respondents' maximum years of experience in extension were 20 for adopters 9 for nonadopters and 20 years for total samples. The minimum years of experience in extension for adopters was 2 years, for non-adopters 1 year and for total respondents was 1 year. The variations between maximum and minimum years of experience in extension were 18 years for adopters, 8 years for non-adopters and 19 years for total respondents as presented in Table 10. The maximum and minimum farming experience for adopters were 55 and 2 years, for nonadopters 60 and 4 years and for that of total respondents were 60 and 2 years. The variation between maximum and minimum total farming experience of adopters was 53, non-adopters were 56 and that of total respondents was 58 years as indicated in Table 10.

The maximum and minimum family size of adopters respectively were 11 and 1 for adopters, 10 and 1 for non-adopters and 11and 1 for total samples. The variation between maximum and minimum family size was 10 for adopters, 9 for non- adopters and 10 for that of total respondents as indicated in Table10.

Table 11.Adopters and non-adopters demographic characteristics Adopters (99) Characteristic Age Family size Experience in Extension (years) Farming Experience (years) Mean 46.10 5.85 7.87 21.90 SD 13.256 2.192 4.787 11.08 Non-Adopters (51) Mean 46.47 5.10 3.765 20.80 SD 14.53 2.385 1.784 10.98 T-test 0.157 1.927* -0.907*** -0.596 Significance Level (2-tailed) 0.876 0.056 0.000 0.552

***and* significant at 0.01and 0.10 p-value respectively.

The average age of adopters, non-adopters and total respondents were, 46.10, 46.47 and 46.23 years respectively. The S.D (Standard Deviation) of adopters, non-adopters and total respondents ages were 13.256, 14.53 and 13.655 respectively as indicated in Table 11.T-test

54

statistics was run to check whether there is a significant mean difference in age between adopters and non-adopters. The result of t-test showed that there was no statistically significant mean age difference.

The respondents average/mean and S.D (Standard Deviation) family size of adopters, nonadopters and total respondents were 5.85, 5.10 and 5.59 respectively as indicated in Table 11. T-test statistics was run to know whether there is statistically significant variation in average family size between adopters and non-adopters. The result of t-test analysis showed that there was statistically significant difference in average family size at 10 percent probability level as indicated in Table 11.

The respondents average (mean) and S.D (Standard Deviation) of experience in extension and total farming experience in years is presented in Table 11. The average years of experience in agricultural extension as well as the total farming experience of adopters were 7.87 and 21.9 that of non-adopters were 3.765 and 20.8 and the total respondents experience in extension were 4.485 and 11.023 years respectively. T-test was conducted to see the variation in average years of experience in agricultural extension and in total farming experience between adopters and non-adopters. The result of t-test analysis showed that there is a significant difference in average years of experience in agricultural extension participation involvement at 1 percent probability level as indicated in Table 11.

But the total farming experience was not significant in t-test analysis. Because those farmers who have better experience in extension could got better extension service that help them to adopt better improved bread wheat varieties.

From the total sample respondents 103 (68.67%) were involved in improved bread wheat production, while the remaining 47 (31.33%) respondents were not involved in improved bread wheat production during the survey year. Their reasons why they were not involved had summarized and presented in Table 12 from the response they gave during the interview.

55

Table 12.Reasons given for not using improved bread wheat varieties No 1 2 3 4 5 6 7 8 Reasons limiting to involvement in improved bread wheat production Farm land shortage Lack of information Fertilizer shortage High price of fertilizer Lack of extension support Labor problem Seed scarcity Lack of ploughing oxen N (47) 15 9 6 6 4 3 2 2 Percent Rank 32.00 19.15 12.77 12.77 8.51 6.38 4.25 4.25 1st 2nd 3rd ,, 4th 5th 6th ,,

(Source: Computed from own survey data, 2005)

Farmland shortage and lack of information were the two most important reasons that limit farmers in the study area. The remaining reasons were less important for farmers to adopt improved agricultural practices as indicated in Table 12.

Table 13.Level of awareness of improved bread wheat varieties Improved bread wheat Aware varieties N Percent HAR-1685 (Kubsa) 123 82 HAR-1709 (Mitike) Paven-76 54 115 36 76.67 Not aware N Percent 27 18 96 35 64 23.33 Total N Percent 150 100 150 150 100 100

(Source: Computed from own survey data, 2005)

56

Concerning respondents awareness of improved bread wheat varieties, interviews were conducted. About (123) 82% respondents knew HAR-1685, (115) 76.67% knew Paven-76

and 54 (36%) knew HAR-1709 variety. As indicated in Table 13, HAR-1685 was known by larger proportion of respondents variety by respondents. followed by Paven-76 and HAR-1709 was the least known

Table 14.Sample Farmers perception on benefit of fertilizer 1 1.1 1.2 1.3 1.4 1.5 2 Perception on fertilizer benefit Low profit No loss or no profit High profit Very high profit Encountered loss Total Perception on fertilizer Problems 2.1 High price (high interest rate) 2.2 Un-timely and lately arrival 2.3 Credit scarcity and credit service related problems to purchase fertilizer Total N 98 26 14 9 3 150 N 75 56 19 150 Percent 65.33 17.33 9.33 6.00 2.00 100 Percent 50 37.33 12.67 100

(Source: computed from own survey data, 2005)

Respondents were interviewed to know their opinion based on their experience about the benefits obtained from fertilizer use. About 65.33% low profit, for 17.33% reported no loss or no profit for 9.33% high profit, for 6% very high profit could be obtained and 2% said encountered loss. In this study the larger proportion of farmers reported the low profit from fertilizer as indicated in Table 14.

Respondents

were also interviewed to get their idea on problems related to fertilizer in their

area. About 50% of respondents have reported high price, 37.33% reported un-timely and late arrival and about 12.67 % reported credit scarcity and credit service related problems to

57

purchase fertilizer as indicated in Table15. The problems arisen due to weakness of the credit provider institutions and less attention of the government as observed during data collection time.

Table 15.Beginning time of cultivation of improved bread wheat varieties of sample farmers

No 1 2 3 4 5 6

Starting Years Before During ,, ,, ,, ,, 2000 2000 2001 2002 2003 2004 Total

N 9 10 27 28 27 2 103

Percent 8.74 9.71 26.26 27.30 26.26 2.02 100

(Source: computed from own survey data, 2005)

In this study to know their commencement or beginning time of using improved bread wheat production, respondents were interviewed and their responses were summarized in Table 16.

Table 16.Health status and adoption of improved bread wheat varieties Health status Non- adopters Un-healthy Healthy Total 7 (13.73%) 44 (86.27%) 51(100%) Adopters 7(7.07%) 92 (92.93%) 99(100%) 2-test df Co coef Sig. Total 14(9.33%) 136(90.67%) 150(100%)

1.762

0.108

0.184

(Source: Computed from own survey data, 2005); Co coef =contingency coefficient

To accomplish the agricultural activities as required, the farmers need to be healthy. In this study, it was tried to assess the household head respondents health situation. The respondents were grouped into healthy and un- healthy farmers (those who face a health problem) to

58

accomplish their day-to-day agricultural activities. From total adopters the healthy farmers were 92.93% and that of unhealthy were 7.07%. In the case of non-adopters 86.27% were healthy and 13.73% were unhealthy. Out of 150 respondents, (136) 90.67% were fully healthy and the remaining (14) 9.33 % had health problem. To check the relationship of the health situation of the respondents and adoption of improved bread wheat varieties, a chi-square test was conducted and the result showed that the relationship between health status and adoption of improved bread wheat varieties was not statistically supported and insignificant as indicated in Table 16.

Table 17.Sample household educational status Education Illiterates Literate Total NA 33(64.70%) 18(35.30%) 51(100%) Ad 64(64.65%) 35(35.35%) 99(100%) 0.000 1 0.001 0.994 X2 df Co. coef Sig. Total 97(64.67%) 53(35.33%) 150(100%)

(Source: Computed from own survey data, 2005); * NA=Non-adopters, Ad=Adopters; *Co coef =contingency coefficient.

Education is very important for the farmers to understand and interpret the information coming from any direction to them. Farmers education is also pivotal for the effective work of extension personnel because if the farmer has better education status he/she can has a capability to understand and interpret easily the information transferred to them from Extension Agent (EA). From total non-adopters 35.30% were literates and 64.70% were illiterates. In the case of adopters 35.35% were literates and that of 64.65 % were illiterate. The proportion (percentage) of illiterate adopters and non-adopters as well as that of literate adopters and non-adopters was almost equal as indicated in the Table 17.

59

In this study the literacy was extended from read & write to attending regular school education. To see the relationship and the intensity of relationship, the chi-square- test was conducted. But the result of chi-square- test was not statistically significant as indicated in Table 17. This means there is no any significant difference in adoption between adopters and non-adopters due to education.

4.1.2. Respondents` livestock and land ownership

In the study area mixed farming is practiced with crop and livestock production. Each household owns at least one or more types of livestock and a piece of land for crop and livestock production.

Livestock in the study area provides traction and manure and also serves as a source of income through sale of livestock and livestock products. Livestock also serves as a source of fuel in the study area. Crop residue and by-products serve as livestock feed source.

As it confirmed in many studies farmers who have better livestock ownership status are likely to adopt improved agricultural technologies like improved bread wheat varieties; because, livestock can provide cash through sale of them and their products and draught power for agricultural operations. In this study, it was revealed that the average livestock ownership of adopters and non-adopters in TLU were 6.834 and 5.02 respectively.

To know whether there is a variation in average livestock ownership between adopters and non- adopters and as a result if there is any significant difference due to the resource position, t-test was conducted. The result of t-test showed that there is a significant variation in average livestock ownership between adopters and non-adopters at one percent probability level as indicated in Table 18 and the average oxen ownership of adopters was also significantly larger (3.03) than non-adopters (2.157) at 5 percent probability level as indicated in Table 18. As ttest indicated, adopters had larger livestock and oxen ownership as compared to non-adopters.

60

This implied that large ownership of oxen and livestock can help farmers to adopt agricultural innovations by solving the power force need for improved wheat production practices and cash constraint by providing income from sale of livestock and their by products to purchase agricultural inputs.

Table 18.Livestock and land ownership of respondents farmers

Characteristics Livestock ownership (TLU) Oxen owner ship Improved Bread wheat land (ha) Total wheat land Total farm land (ha) Total land (ha)

Ads Mean 6.8340 3.0300 0.9621 1.2400 2.7141 3.01

SD 3.2724 1.6317 0.4579 0.6358 1.08 1.25

NAs Mean 5.0200 2.1570 0.0400 0.8530 2.0147 2.28

T-test SD 3.2120 1.6294 0.1759 0.3846 1.121 1.25 -3.236*** -3.107** -13.859*** -3.983*** -3.710*** -3.382***

Significance (2-tailed) 0.001 0.002 0.000 0.000 0.000 0.001

*** and ** Significant at 1 and 10 percent probability level

Sample farmers vary in their adoption of improved bread wheat varieties by their livestock and oxen ownership. Adopters average livestock ownership was significantly larger than nonadopters. This indicate that livestock ownership help farmers to adopt improved bread wheat varieties since the income from livestock obtained through selling of the animals or their by products can help to solve their financial limitation s to purchase inputs.

Land is the main asset of farmers in the study area. Farmers in the study area use both their own land and rent farm land for crop production and grazing land for livestock production .All 150 sample households have their own land and only (24) 16% and (2) 1.33% respondents rented cultivated and grazing land respectively. The average land holding of adopters was 3.01 hectares total average land holding, 2.7141 hectares total average farmland, 1.24 hectares total average wheat land and 0.9621 hectares was average farm land used for improved bread wheat

61

production. In the case of non- adopters the total average land holding was 2.28 hectares, 2.0147 hectares was average farmland, 0.8530 hectares was average wheat land and only among improved bread wheat growers of non-adopters was 0.0400 hectares in average was used for improved bread wheat production. To know whether there is the mean land holding variation, between adopters and non-adopters, t-test analysis was carried out and the result showed that there were the significance differences in all types of land holding at one percent probability level as indicated in Table 18.The result showed that farmers who have better land ownership can adopt improved bread wheat varieties better than non-adopters.

Table 19.Respondents land ownership in 1996/97 Ethiopian major cropping season Area of land ownership in hectares Total farm Land Total wheat land Total improved B.W.L. Max 6.5 4.5 3.25 Min 0.25 0.25 0.25 Range 6.25 4.25 3 .00 Average 2.44 1.125 1 .00 St.D 1.13 0.605 0.457

(Source: Computed from own survey data, 2005) * N.B=Improved B.W.L. (Improved Bread Wheat Land)

The total farmland ownership of respondents ranges from 0.25 hectares to 6.5 hectares. The total wheat land ownership of sample households ranges from 0.25 hectares to 4.5 hectares. The improved bread wheat land holding of sample household ranges from 3.25hectares. 0.25 hectares to

On the average sample households owned 2.44 hectares total farmland, 1.13 hectares total wheat lands and 0.96 hectares used for improved bread wheat production as presented in Table 19. In the study area respondents farmers allocated most of their farmland for wheat production as presented in table one.

62

4.1.3. Accessibility of respondents to different institutional services

In this study respondents were interviewed to get their opinion about the importance of extension services based on their experiences. About 84.67, 3.33% and 12% respondents have reported important, not important and do not have any opinion respectively.

The respondents have also been interviewed to give their opinion about the extension support they obtained. About 46.67% reported extremely weak due to un-availability of development agent, about 27.33% reported very weak even though the Development Agents are available around. The remaining 26% responded that the extension service they got in their area becomes extremely weak due to unknown reasons for them as indicated in Table 20.

Table 20.Respondents opinion on extension service of the study area

No 1. 1.1. 1.2. 1.3. 2. 2.1. 2.2. 2.3.

Respondents opinion on extension service About importance of extension service Important Do not have any idea Not important Total Status of extension service of the study area Extremely weak due to un-availability of DA Very weak even though the DA available Extremely weak due to un-known reasons for them Total (Source: Computed from own survey data, 2005)

N 127 18 5 150 70 41 39 150

Percent 84.67 12 3.33 100 46.67 27.33 26 100

Data were collected regarding the type extension service obtained by the respondents as indicated in Table.20. The whole non-adopters and 94.95% of adopters did not get extension service during the survey year on improved bread wheat variety. As indicated in Table 21 only

63

five respondent farmers reported that they got extension support. The respondents

were also

interviewed to get their opinion on the distance of DAs office from their home. As indicated in table 22 about 90% reported far and the remaining 10% reported close to their home.

Table 21.Extension support on improved bread wheat varieties and distance of DAs office Service accessibility Extension service on bread wheat Responses No Yes Total Far Close Total NAs 51 51 48 3 51 Ads 94 5 99 87 12 99 X2 df Sig. C.coef Total . 145 5 0.138 150 135 0.098 15 150

2.665*

0.103

Distance of DA office

1.456

0.228

*Significant at 10 % probability level. ; Ccoef = Contingency coefficient *NAs =Non-adopters, Ads = Adopters, df =Degree of freedom.

To know the association of extension service and distance of DA office with adoption of improved bread wheat variety, chi-square analysis was conducted. The result of chi-square analysis (2.665) showed that there is a significant association between extension service and adoption of improved bread wheat varieties at 10 percent probability level. But the chi-square test result of distance of DA office and adoption of improved bread wheat varieties was not significant as indicated in Table 21.

As it has indicated in many literatures, credit is considered as one of the favorable factors for improved agricultural technologies adoption because it can solve financial constraints of farmers to purchase and use improved agricultural inputs. Respondent farmers have reported about credit institution services and related problems in their area based on their experience. Of that, 83 (55.33%) have reported that there is scarcity, 26 (17.33%) reported that there is a complex and boring procedures and the remaining 41(27.34%) reported that there is a high interest rate problems as indicated in Table 22.

64

Table 22.Summary of respondents opinion on credit

No 1 2 3

Responses on credit related problems Scarcity High interest rate Complexity of procedures Total Responses on the importance of credit

N 83 41 26 150

Percent 55.33 27.34 17.33 100.00

1 2

Credit is important Credit is not important Total Suggestions for better credit services Easy and Reduced procedures Low Interest rate Total

101 49 150 59 42 101

67.33 32.67 100 58.42 41.58 100

1 2

(Source: Computed from own survey data, 2005)

Respondent farmers were also interviewed to provide their opinion about the importance of credit and their suggestions for better credit service in the future. The respondents responses on these issues were summarized and presented in Table 22. About 67.33% respondent farmers reported that credit is important and the remaining 32.67% reported that credit is not important.

From those 101 respondent farmers who supported the credit service as important also provided their opinion for better credit service. About (59) 58.42% reported reduced processes and procedures and about (42) 41.58% suggested to reduce the interest rate as indicated in Table 22.

65

As presented in Table 22 from the total 150 sample respondents, only 26 adopters and 8 nonadopters with a total of 36, which constituted 24% of the total respondents, were credit users particularly from individual credit sources in the mean time when this study conducted. In credit utilization, adopters were larger in proportion than non-adopters. Almost all of the credit users of sample respondents have reported that the major credit sources for them were informal and private lenders.

Table 23.Association between credit and market service Accessibility Credit Service Response NA No Yes Total Far Close Total 43(84.31) 8(15.69) 51(100) 43(84.31) 8(15.69) 51(100) Ad 71(71.72) 28(28.28) 99(100) 89(89.90) 10(10.10) 99(100) 0.994 1 0.319 0.081 X2 df Sig. Co.coe Total 114 36 150 132 18 150

2.928*

0.087

0.138

Market Access

*Significant at 10% probability level; Numbers in brackets are in percentage

To see the association between adoption of improved bread wheat varieties and credit service, a chi-square test was carried out. The result showed that there is a significant relationship between adoption of improved bread wheat varieties and credit services at 10 percent probability level as indicated in Table 23. Market accessibility is also another important factor for farmers to adopt improved agricultural inputs. If farmers are closer and having access to credit services they can easily purchase improved agricultural inputs and sell their agricultural outputs without moving long distances. Farmers also motivated to use improved agricultural inputs if they have access to attractive market for their output to sell in good price. In this study respondent farmers were interviewed to provide their idea regarding the market accessibility. About 84.3% non- adopters and about 89.90% have reported far from market and the remaining 15.69% non-adopters and 10.10% adopters reported close to market access as

66

indicated in Table 23. A chi-square-test analysis was carried out to check the association between market access and adoption of improved bread wheat varieties. The result showed that the relationship was not statistically significant as indicated in the Table 23.

Table 24.Summary of households accessibility of off-farm job

No 1

Respondent farmers access to off-farm job Have access to off-farm job Adopters Non-adopters

N 26 17 9 124 82 42 150

Percent 17.33 65.38 34.62 82.67 66.13 33.87 100.00

Have not access to off-farm job Adopters Non-adopters Total (Source: computed from own survey data, 2005)

Income from off-farm job can play a great role in adoption of improved agricultural technologies. Because, it has hypothesized that the income from off-farm can solve farmers financial constraints to purchase and use improved agricultural inputs. In this study about 17.33% sample households reported that one of their family members has off-farm job and the remaining 82.67% do not have family members who have off-farm job. From the total 26 sample households that have off-farm job 65.38% were adopters and the remaining 34.62% were non-adopters as indicated in Table 24.

67

Table 25.Respondent farmers reasons for not involvement of their family in off-farm job No 1 2 3 4 5 Reasons Under and over aged Students Work on the house hold farm Do not have family members Less income from off-farm job Total (Source: Computed from own survey data, 2005) N 42 12 52 16 2 124 Percent 33.87 9.68 41.94 12.90 1.61 100

From the total 124 sample respondents whose family members were not involved in off-farm activities had reported their reasons during the interview why the house hold members did not involve in off-farm job. As indicated in Table 25, about 33.87% have reported that their family members couldnt involve in off-farm job due to the under and over age, 9.68% reported that their family members are students, 41.94% described the time constraint since the household members would work on the house hold farm, 12.9% reported that they do not have family members and 1.61% reported less income from off-farm job.

Table 26.Rrespondents opinion on decision of off-farm and other household resources No 1 2 3 4 Total : (Source: Computed from own survey data, 2005) Respondents opinion on house holds resources decision maker Husband Husband and wife Wife House hold members together N 121 27 1 1 150 Percent 80.66 18.00 0.67 0.67 100.00

68

Concerning the decision on the off-farm and other agricultural income, the household head respondents were interviewed. All of them were reported that the house hold member who involved directly in the off-farm job make a decision for what purpose the income from offfarm job need to be used. But concerning the other agricultural incomes, from the total 150 respondents about 80.66%, 18%, 0.67% and 0.67% reported that the decision were made by husband, by husband and wife, by wife and by the house hold members together, respectively as indicated in the Table 26. In this study, almost all decision on the agricultural resources of the farming household made by the husband.

Table 27 Pattern of off-farm income utilization of respondent farmers No 1 2 3 4 5 Total Use of off-farm income Household food consumption Cloth purchase Health treatment Input purchase Labor hiring N 7 10 5 3 1 26 Percent 26.92 38.46 19.23 11.54 3.85 100.00

(Sources Computed from own survey data, 2005)

On the use of off-farm income, the Total 24 household respondent farmers who themselves and their family members had off-farm job reported about their and their family members offfarm income utilization. About 26.92 %, 38.46 %, 19.23%, 11.54% and 3.85% reported for household food consumption, cloth purchase, health treatment, input purchase, and labor hiring purposes respectively as indicated in Table 27. In this study the order of importance in off-farm income utilization from higher to the lower were, for cloth purchase, food consumption, health treatment, labor hiring and input purchase. Allocation of off-farm income to agricultural input purchase took the least proportion.

69

Table 28.Family labor utilization of respondent farmers No 1 2 Family labor utilization Utilized family labor Not utilized family labor Total Ad NA Total Percent 91.33 8.67 100

93(93.94) 44(86.27) 137 6(6.06) 99(100) 7(13.73) 51(100) 13 150

(Source: Computed from own survey data, 2005)

In this study the respondent respondents` labor source and their family labor utilization were revealed through interview. From the total 150 respondents, about (137) 91.33% have reported that they used their family labor on their farm activities for weeding, harvesting, threshing, plowing and sowing as indicated in Table 28.

Table 29.Types of activities and family labor utilization of respondents No 1.1 1.2 1.3 1.4 1.5 No 2.1 2.2 2.3 1.Activities on which family labor used Ploughing Sowing Weeding Harvesting, Threshing 2.Critically labor required activities Weeding Sowing Ploughing N 5 2 120 7 3 Total 137 N 109 28 13 Total 150 Percent 3.65 1.46 87.59 5.11 2.19 100 Percent 72.66 18.67 8.67 100

(Source: Computed from own survey data, 2005)

70

Out of these sample households who used their family members labor on the household farm, about 67.9% were adopters and the remaining 32.10% were non-adopters. Respondents were also interviewed to describe the type of agricultural activities they used their family labor. From the total 137 respondents who used their family labor on their farm, about 87.59%, 5.11%, 2.19%, 3.65%, and 1.46% reported for weeding, harvesting, threshing, plowing and sowing respectively. Concerning the critical labor requirements of the respondents labor

requirement were about 8.67% were for plowing and sowing, 18.67% for weeding, and 72.66% were reported for harvesting and threshing as indicated in Table 29.

Table 30.Respondents accessibility to non-family labor and to off-farm income Access to Labor outside the house hold labor Access to off-farm income Response No Yes Total No Yes Total NA 48(94.12) 3(5.88) 51(100) 42(82.35) 9(17.65) 51(100) Ad 87(87.88) 12((12.12) 99(100) 82(82.83) 17(17.17) 99(100) 1.456 1 0.228 0.098 X2 df Sig. C.coef Total 135 15 150 124 26 150

0.005

0.942

0.006

(Source: computed from own survey data, 2005) * Numbers in brackets are in percentage

To check the association between off-farm income of the sample household and adoption of improved bread wheat varieties chi-square analysis was carried out and the result showed that there is no a systematic association statistically supported between adoption of improved bread wheat varieties and off-farm income as revealed in this study as the result presented in Table 30. To see the labor source and accessibility of respondents to labor outside the house hold

labor and its relation ship with adoption of improved bread wheat varieties, chi-square test was conducted and the result showed that the two variables, adoption and the utilization of labor outside the household labor is not statistically significant as indicated in Table 30.

71

Table 31.Respondent farmers labor sources outside their family members

No 1 2 3 4

Types of labor sources Employed labor Exchange labor Relatives and colleagues support Not used labor outside their family labor Total

N 80 41 15 14 150

Percent 53.333 27.333 10 9.333 100

(Source: Computed from own survey data, 2005

Respondents were also interviewed for their labor source other than their family labor. About 53.33%, were reported that they used hire or employed labor, about 27.33% reported as they used exchange labor, 10% were report that they used support labor from relatives and colleagues and 9.33% were reported that they do not used labor outside their family labor source. The respondents labor source outside the family members labor, employed and exchange labor was very important. In the study area the agricultural activities required the higher labor are harvesting, threshing and weeding as indicated in Table 31.

4.1.4. Agricultural information sources of the study area

Access to information or extension messages as well as various extension services was one of the institutional characteristics hypothesized to influence farmers decision to adopt a new technology. One can gain access to information about new technologies through various means such as attending field days, visiting demonstration fields, participating training, listening to agricultural programs on radio, through contact with Extension or Development Agents, and through various forms of communication with neighbors, relatives, other colleague farmers and leaders of community, religious and PA (Peasant Associations) and through other means (Tesfaye et al, 2001).

72

As shown in various literature, different extension methods such as training, demonstration, farmers field day, farmers meetings (mass-meeting), group discussions, posters, mass communication methods and other extension methods have described that can be employed to transfer extension messages to the farmers. Using and practicing these extension methods properly to transfer extension messages can facilitate diffusion and adoption of improved agricultural inputs.

In this study the improved bread wheat grower respondents were interviewed to give their opinion how they got extension messages regarding the utilization and application of improved bread wheat varieties production and management. Out of 103 respondent farmers who grew improved bread wheat during the survey year, only (4) 3.88% were non-adopters and the remaining (99) 96.12 were adopters. From 99 total adopters only (3) 3.03% reported that they got training on improved bread wheat varieties production and management as presented in Table 32.

As shown in the Table 35 about 100 of improved bread wheat growers (adopters and non adopters) who do not got training were interviewed how they precede the improved bread wheat varieties production. About 89% of improved bread wheat growers reported by seeing other grower farmers, (copying mechanism or farmer to farmer extension), about 8% reported by trial and error method and the remaining growers reported by asking the help of Development Agent and other educated people living in their area.

To know the field day and demonstration program participation of Improved bread wheat growers, their interview responses had summarized as indicated in Table 32., only 11.77% respondents told that they got an opportunity to attend field day and demonstration program. But the remaining (91) 89.23% had not have it. Out of these 12 respondents only 8.33% were non-adopters and 91.67% were adopters.

73

Table 32.Respondents participation in training, field day and demonstration No 1. 1.1. 1.2. 2. 2.1. 2.2. 3. 3.1. 3.2. 3.3. Training, field day and demonstration participation Adopters (99) Non-adopters (4) Total (103) Total (103) 3 (2.91) 100 (97.09%) Total 103 12 (11.77%) 91 (88.23 %) Total 150 17 (11.33%) 67 (44.67%) 66 (44 %)

Training participation Attend training 3 (3.03%) Not attending training 96 (96.97%) Field day and Demonstration program participation Attending training 11 (91.67%) Not attending training 88 (96.7) Attending extension meeting called by DA Feel happy to attend Un-happy No feeling (Source: Computed from own survey data, 2005)

4 (100%)

1 (8.33%) 3 (3.3%)

The respondent farmers were also interviewed to know their feeling when called by DA for extension meeting Regarding their feeling they were show when they receive DAs call for extension meeting; about 11.33%, 44.67% and 44% were reported that they feel happy, unhappy and did not have any feeling on this issue respectively as indicated in Table 32.

In the study area respondent farmers were interviewed to provide their idea regarding their agricultural information sources. As it is presented in Table 33, neighbors and colleague farmers, DA, community leaders, farmers field day, PA leaders, demonstrations, radio, newspaper/news letters, publications, posters, training programs, TV and religious leaders have served as sources of general and agricultural information sources for them.

74

Table 33.Respondent farmers sources of information

Information Sources Neighbors and colleague farmers DA Community leaders Farmers field day PA leaders Demonstration Radio News paper/News letter Other publication Poster Training TV Religious leaders (Source: Computed from own survey data, 2005)

N 138 136 133 126 123 123 123 122 122 107 105 104 101

Percent 92 90.67 88.67 84 82 82 82 81.33 81.33 71.33 70 69.33 67.33

Rank 1st ,, 2nd 3rd 4th ,, ,, 5th ,, 6th 7th 8th 9th

As in Table 33 presented farmers neighbor and colleagues are the major and the firs important farmers source of information. This survey result is similar with the result of group discussion conducted in this study. According to this study DA serve as the second information source. The survey result showed that the third and fourth sources of information are community leaders and farmers field day respectively. As showed in the Table 33 PA leaders, demonstration and radio serve as fifth source of information. Newspaper/news letter and other publications serve as sixth information source. The remaining, poster, training TV and religious leaders serve as seventh, eighth, ninth and tenth sources of information respectively as indicated in Table 33.

75

In this study, it was also tried to summarize the agricultural information sources of the farmers in the study area through group discussion. During the time of group discussion the group members were familiarized to the discussion point and were expected to identify and prioritize the agricultural information sources of farmers in their area .The group members took care in listing of all alternative sources of information available in their area using brain storming method and tried to refined, summarized and prioritized the listed alternative information sources listed through brain storming method.

The result of the group discussion showed that; a neighbor stands first and the most important and TV stands the last and least important. The result of the group discussion findings showed that farmers got more information easily from their neighbors than other sources available in their area. The second most important information sources of farmers in the study area were religious and community leaders. PA leaders and DAs serve as third and fourth respectively as sources information. Demonstration and field day training and posters serve as fifth, sixth and seventh sources of information respectively. The remaining, publications, radio and TV serve as eighth, ninth and tenth sources of information respectively for the farmers in the study area.

4.1.5. Farmers selection and evaluation criteria of improved bread wheat varieties

Varieties characteristics play a vital role in adoption of improved varieties if their characteristics satisfied the need, interest and in line with the environmental situations of the farmers. The information on evaluation and selection criteria of improved bread wheat varieties of the farmers in the study area was analyzed through personal interviews and group discussion. The procedure to analyze the information through group discussion was conducted as; first make familiar farmers to the discussion agenda, and let them to establish and set evaluation and selection criteria of improved bread wheat varieties disseminated in their area. In the process of setting and establishing the criteria the group were applied the method of brain storming and list down all the ideas provided and forwarded by the group members .The group continued to refined the ideas forwarded by the group members and set or established

76

the evaluation and selection criteria with a common agreement. In this study, the result of the group discussion showed that the best improved bread wheat variety should constituted the white grain color, large seed size, and high disease, pest and frost resistance, good food quality, good straw quality as animal feed and attractive market demand characteristics. This study is in line with the study of (Ethiopian Rural Self Help Association /ERSHA, 2000).

Table 34.Farmers evaluation and selection criteria of improved bread wheat varieties

No Variety Characteristics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 White grain color Large grain Size Straw quality Market demand Germination capacity Cooking quality Better yield performance Water lodging resistance Tillering capacity Food quality Short maturity date Disease resistance and pest resistance Frost resistance Harvesting quality Storage quality

N 140 140 140 140 139 139 139 138 138 138 137 135 133 97 97

Percent 93.33 93.33 93.33 93.33 92.67 92.67 92.67 92.00 92.00 92.00 91.33 90.00 88.67 64.67 64.67

Rank 1st ,, ,, ,, 2nd ,, ,, 3rd ,, ,, 4th 5th 6th 7th ,,

(Source: Computed from own survey data, 2005)

The results in evaluation and selection of improved bread wheat varieties disseminated in the study area has showed in Table 34 that white grain color, large grain size, market demand and straw quality were the first and most important criteria. The traits such as better yield performance, cooking quality and germination capacity got the second rank. Food quality, tillering capacity and water lodging resistance got the third rank Short maturity date, pest and

77

disease resistance and frost resistance were got the fourth, fifth and sixth ranks respectively. Harvesting and storage qualities were got the seventh rank by farmers judgment.

Table 35.Farmers preference (selection and evaluation criteria) of improved bread wheat varieties disseminated in the study area No Variety Characteristics 1 Market demand 2 Cooking quality 3 Water logging rsistance 4 Straw quality 5 Storage quality 6 Frost resistance 7 Seed size 8 Yield performance 9 Rain shortage and Drought 10 resistance Weed resistance HAR-1685 (123) HAR-1709 (54) N Percent N Percent 97 86 86 84 83 80 79 77 77 76 78.8.6 69.92 69.92 68.29 67.48 65.04 64.23 62.60 62.60 61.80 57.73 56.91 53.66 12 23 27 36 33 27 26 38 32 42 37 32 40 405 31.154 3rd 22.22 42.60 50 66.67 61.11 50 48.15 70.37 59.26 77.78 68.52 59.26 74.04 Paven-76 (115) N Percent 41 41 37 30 34 43 45 35 41 32 42 45 44 510 39.231 2nd 35..65 35.65 37.12 26.10 29.57 37.40 39.13 30.43 35.65 27.82 36.52 39.13 38.26 -

11 Grain color 71 12 Food quality 70 13 Disease and Pest resistance 66 Sum 1032 Average 79.385 Rank 1st

(Source: Computed from own survey data, 2005)

Respondent farmers were interviewed to get their idea on evaluation and selection criteria of improved bread wheat varieties and their responses has summarized in Table 35.The respondent farmers responses showed that the improved bread wheat varieties should constitute the characteristics mentioned in Table 35. These evaluation and selection criteria are the most important criteria for the farmers in the study area. The respondent farmers have given their preference of improved bread wheat varieties distributed in their area. The larger

78

proportion of respondent farmers selected HAR-1685 improved bread wheat variety. The remaining varieties Paven 76 and HAR, 1709 ranked second and third respectively. This survey result was also supported and similar result was obtained from group discussion conducted in this study.

4.2. Analytical results and discussion

The purpose of this section is to identify the most important hypothesized independent variables that influence the dependent variables namely the probability of adoption for nonadopters using logit model and the intensity of adoption for adopters using tobit model analysis of improved bread wheat varieties in the study area, Akaki. Before conducting the model analysis selection, screening and verification of hypothesized variables were conducted by considering various situation to get best variables, those can fit with the analytical models, describe the sample groups, environmental and practical situation of the study area. This was done in consultation of professionals and experienced people, based on literatures, practical situations, observation and experience of the researcher and the relevance as well as the importance of the variables. As a result, the variable, distance of credit provider institutions was dropped because the major credit source for the farmer in the study area were the private individual credit providers in time when this research was conducted. These individual credit providers do not have a specific place and including this variable in model analysis was not relevant.

In the case of significant level of hypothesized independent variables, independent sample test between the groups using t-statistics or t-test for continuous variables to describe the pattern of sample data and to test the significance of a given independent variable on adopters and nonadopters groups as well as to check the mean values differences of continuous variables in the two groups and the chi-square test also to test the differences between the two groups for discrete variables in relation to dependent variables (Lind and Mason, 1994 as cited in Adane, 2002). In the analysis some independent variables might show significant and others might show insignificant relationship with dependent variables. The insignificant association doesnt

79

guarantee about the strength or direction of relationship between those insignificant hypothesized independent variables and the dependent variables. The reason for the insignificant relationship of some of the independent variables is mainly because of the fact that there is a drawback with any univariate approach in that it ignores, but there could be a possibility in the collection of variables analysis, each of which is weakly associated with the univariate outcomes can become an important predictor of out come when taken together. Therefore, we should consider them as candidates to be indicated in the multivariable models analyses along with all known important variables (Hosmer and Lemeshow, 1989 as cited in Adane, 2002).

Moreover, there are several literatures and previous research works (Chilot, 1994; Bekele, 2000; Adane N.F., 2002; Adane N.M., 2002; Techane 2002; Endrias, 2003; Yitayal, 2004; Adam and Bedru, 2005) conducted in a similar way which can substantiate this study. These previous research results showed that those hypothesized independent variables were included in econometrics model analyses regardless of the significant or insignificant results of these hypothesized independent variables in descriptive analysis. The model analyses results might show significant or insignificant results differently or similarly to descriptive statistics results. Regarding multicollinearity or high degree of association problem among selected and screened hypothesize independent variables were primarily checked before including in the models as well as before running the models analyses.

Secondly, prior to running the Logit and Tobit models, the presence or absence of correlations or associations between hypothesized independent and dependent variables were checked. The presence or absence of correlation or association, that is, whether or not there is a correlation between the variables in question (Sarantakos, 1998).

Existence, direction and strength of correlation are demonstrated in the coefficient of correlation. A zero correlation indicates that there is no correlation between the variables. The sign in front of the coefficient indicates whether the variables change in the same direction (positive correlation) or in opposite direction (negative correlation), except for nominal

80

measures, where the sign has no meaning, in which case coefficient describe only the strength of the relationship (a high or a low association) between the variables of the study. The value of the coefficient shows the strength of the association with values close to zero meaning a weak correlation and those close to 1 a strong correlation. A correlation of +1is just as strong as one of -1; it is the direction that is different (Sarantakos, 1998). Therefore, in this study, the presence or absence of association or correlations of hypothesized independent variables with the dependent variable, adoption of improved bread wheat varieties were assessed to identify and drop from model estimation if hypothesized independent variables do not have any relationship with dependent variable, adoption of improved bread wheat varieties. The Cramers v coefficient for discrete variables and Pearsons correlation coefficient for continuous variables was calculated using SPSS computer program.

In addition, the direction, range, intensity or degrees of strength of association of each hypothesized variables with dependent variables were assessed. The result showed that there was no total absence of association between hypothesized independent variables and dependent variables. Rather, association between hypothesized independent and dependent variables exist with various degrees of association ranging from moderate to weak. As a result, it was decided to include all selected, verified, screened hypothesized independent variables, those have various degrees of relationship with dependent variables, in models analyses to see their combined effect they have on dependent variables namely probability and intensity of adoption. From these total selected independent variables, only farmland showed moderate correlation with the dependent variable. But the rest showed weak association as indicated in Appendix table 24.

Thirdly, before including the hypothesized variables and running the model analyses the existence of a serious of multicollinearity or high degree of association problem among independent variables for all continuous and discrete variable were checked. There are two measures that are often suggested to test the existence of multicollinearity or association problems among independent variables. These are: Variance Inflation Factor (VIF) for multicollinearity problem among continuous independent variables and contingency

81

coefficients for existence of high degree of association among independent dummy variables. The technique of variance inflation factor (VIF) was employed to detect the problem of multicollinearity for continuous variables. VIF shows how the variance of an estimator is inflated by the presence of multicollinearity (Gujarati, 2003).

It is obvious that multicollinearity problems might arise when at least one of the independent variables shows a linear combination of the others; with the rest that we have too few independent normal equations and, hence, cannot derive estimators for all our coefficients. More formally, the problem is that a high degree of multicollinearity results in larger variances for the estimators of the coefficients. A larger variance implies that a given percentage (eg.95%) confidence interval for the corresponding parameter will be relatively wide; a large range of values of the parameter, perhaps including the value zero, will be consistent with our interval. This suggests that, even if the corresponding independent variable problem may make it quite difficult for us to estimate accurately the effect of that variable. Consequently, we may have little confidence in any policy prescriptions and biased on these estimates (Kelejian and Outes, 1981).

Very often the data we use in regression analysis cannot give decisive answers to the questions we pose. This is because the standard errors are very high or the t-ratios are very low. This sort of situation occurs when the explanatory variables display little variation and/or high inter-correlations. The situation where the explanatory variables are highly inter -correlated is referred to as multicollinearity (Maddala, 1992).

According to Maddala (1992), VIF can be defined as: VIF (xi) =

1 1 Ri2

Where Ri2 is the square of multiple correlation coefficients that results when one explanatory variable (Xi) is regressed against all other explanatory variables .A statistical package known as SPSS was employed to compute the VIF values. Once VIF values were obtained the R2 values can be computed using the formula. The larger the value of VIF, the more will be trouble-some or the collinear of variable Xi. As a rule of thumb, if the VIF of a variable

82

exceeds 10, there is multicollinearity. If Ri2 exceeds 0.90, that variable is said be highly collinear (Gujarati, 2003). The VIF values displayed in Table 36 have shown that all the continuous independent variables have no multicollinearity problem.

Similarly, contingency coefficients were computed from survey data to check the existence of high degree of association problem among discrete independent variables. Contingency coefficient is a chi-square based measure of association .A value of 0.75 or more indicates a stronger relationship (Healy, 1984 as cited in Destaw, 2003).

The contingency coefficients are computed as: C=

2 N + 2

Where, C= Coefficient of contingency

2 = Chi-square random variable and

N = total sample size.

Which assumes a value between 0 and 1 to indicate the degree of association between the discrete variables as indicated in Table 37.The decision rule for contingency coefficients says that when its value approaches 1, there is a problem of association between independent discrete variables. As indicated in Table 37 that there is no a problem of high degree of association among independent discrete variables.

83

Table 36.Variable Inflation Factor for the continuous explanatory variables Variables Age TLU Farm land holding Oxen ownership Experience in extension Family size R2i 0.047 0.637 0.342 0.638 0.036 0.231 Variance Inflation Factors (VIF) 1.049 2.754 1.520 2.759 1.036 1.301

(Source : Own Computation)

Table 37.Contingency Coefficients for Dummy Variables of Multiple Linear Regressions Model HHH SEX HHHSEX EDUHHH HEALSTAT PRTILEDE HHOFFINC DISDAOF1 CRINMFF1 OTSOLA GEXSERVE GECRSERV Source: Own computation 1 EDU HHH 0.184 1 HEAL STAT 0.023 0.003 1 PRTI HHOF LEDE 0.128 0.044 0.155 1 FINC 0.154 0.030 0.035 0.088 1 DIS DAOF1 0.055 0.033 0.046 0.172 0.082 1 CRIN MFF1 0.073 0.058 0.23 0.130 0.048 0.442 1 OTS OLA 0.14 0.196 0.046 0.106 0.094 0.110 0.055 1 GEXS GECR ERVE SERV 0.068 0.018 0.068 0.097 0.085 0.062 0.046 0.183 1 0.063 0.024 0.034 0.080 0.406 0.021 0.174 0.031 0.155 1

As it has indicated in many studies and literatures, if there will be serious multicollinearity or a high degrees of association problems among independent variables, these situations can

84

create difficulties to differentiate the separate effects of independent variables on dependent variables and also seriously affect the parameter estimate because of strong relationship among them. Hence, should not be included in the model analysis (Maddala, 1983; Kathari, 1990 as cited in Adane, 2002 and Gujarati, 1995). But since there is no a serious multicollinearity or high degree of association problem among independent variables in this study all the screened variables were decided to be included in the models analyses.

After conducting and passing all these steps, all screened and verified independent variables were included in logit model analysis using SPSS computer software program. But a problem faced in tobit model analysis to include all these screened and verified independent variables in tobit model analysis using Limdep computer soft ware program due to the limitation of this soft ware program to accommodate all variables included in logit analysis. Therefore, there need to select and choose the variables that can be accommodated by the Limdep soft ware program and most important independent variables for the analysis than others. As a result, from those independent variables included in the logit analysis only leadership position and credit service were dropped from tobit model analysis based on practical and actual situations, researchers observation, relevance of the variables and by employing Limdep computer program to check the number of significant variables those can affect the dependent variable, intensity of adoption of improved bread wheat varieties.

In this process the number of significant independent variables were increased when the above two independent variables were dropped individually or together. At last, the remaining screened and verified hypothesized independent variables were included in tobit model analysis.

4.2.1. Analysis of determinants influencing probability of adoption of improved bread wheat varieties and their marginal effect

To identify factors among hypothesized independent variables that significantly influencing the probability of adoption of improved bread wheat varieties in the study area, Akaki, SPSS

85

computer soft ware program and the binomial econometric analytical model (the binary Logit model) was employed. In fitting the logistic estimation model, the higher significance of chisquare statistics (80.187) was taken as a measurement of goodness-of-fit. This indicates that the explanatory variables together influence the probability of adoption of improved bread wheat varieties in the study area. In addition, the model correctly classified the respondents into adopters and non-adopters at 81.33% of correct prediction percentage. The maximum likelihood estimate of the parameters and the direction of relationship and the effect of independent variables on probability of adoption were analyzed and presented in Table 38.

As indicated in the methodology and other previous sections, a number of independent explanatory factors were postulated to influence the probability of adoption of improved bread wheat varieties in the study area. Among the selected hypothesized explanatory variables and considered by the model, only four variables were found to have significantly affect farmers adoption decision of improved bread wheat varieties. The variables affecting probability of adoption were distance of DA-office from the farmers home (DISDAOF1), house hold social/leadership status (PRTILEDE), market accessibility (CRINMFF1), and house hold farmers experience in extension (YEXPEXTS) as indicated in the Table 38.

Among those significant variables, only one variable, which was market access, related with adoption of improved bread wheat varieties negatively and the sign was different from the expectation but statistically significant at 5 percent probability level. In this study, the negative relationship of market access and adoption of improved bread wheat varieties showed that those farmers in the study area who do not have access to market are more likely to adopt better the improved bread wheat varieties than those farmers who have a better access to market.

The possible reason for this situation might be, those farmers who have better and closer access to market area might create other income opportunity from their farm and they may give more attention and priorities to these other alternatives, production activities and other

86

substitutions, which may bring better income to them than using their whole wheat farm to produce improved bread wheat. But those farmers far away from market since they may not have any other alternatives they give more attention for improved bread wheat production and to their farming occupation. The other possible reason to these farmers who far away from market make them to adopt the improved bread wheat varieties than those who are closer to market might be the better production performance of the varieties to provide food for their family through out the year since they might not have other food sources of alternatives and means .On the other way, environmental situations, the soil fertility frost problem variations might be the possible reasons.

The remaining three significant explanatory variables namely (leadership position/status, experience in extension and distance of DA-office from the farmers home) related with adoption of improved bread wheat varieties significantly and positively, as of the expectations, at 1, 1 and 10 percent probability levels respectively as indicated in the Table 38.

87

Table 38.Factors affecting Probability of adoption of improved bread wheat varieties and the marginal effect of the significant explanatory variables Variables HHHSEX-sex HHHAGE-age EDUHHH-education HEALSTAT-health PRTILEDE leadership position HHOFFING-off-farm income DISDAOF1-DA-office TOTLIVUM -livestock SUMOWRE- farm land CRINMFF1-market OTSOLA1-labor OXTLU-oxen GEXSERVE -extension YEXPEXTS-extension experience GECRSERV -credit FAMILYSI family size Constant B 0.657 -0.023 -0.156 0.942 2.217 -0.688 1.490 0.076 0.361 S.E 0.777 0.021 0.536 0.901 0.756 0.945 0.910 0.146 0.301 Wald df Sign. 0.714 1 0.398 1.194 1 0.274 0.094 1 0.772 1.092 1 0.296 8.610 1 0.003*** 0.531 2.683 0.272 1.435 1 1 1 1 1 1 1 1 1 1 1 0.466 0.101* 0.602 0.231 Exp (B) 1.928 0.977 0.856 2.565 9.181 0.502 4.436 1.079 1.435

-1.636 0.819 0.3.993 0.110 1.228 0.008 0.103 0.278 0.137 7.811 23.647 0.109 0.475 0.200 0.024 -4.053 0.113 0.788 0.135 1.744 17.690 0.064 0.030 5.399

-0.046** 0.195 0.928 1.117 0.712 1.108 0.741 2467.236 1.608 1.221 1.024 0.017 0.800 0.862 0.020

1 0.000***

Notes: Exp (B) shows the predicted changes in odds for a unit increase in the predictor *Omnibus Tests of model coefficients: Chi-square=74.97, Sign.0.000; * Percentage of correct prediction=81.30; and *, **and ***Significant at 10%, 5%, and 1% Significant level.

The variable leadership position affects adoption significantly in the study area as indicated in Table38. Farmers who have a leadership position in the society might give a better opportunity to access resources and inputs such as labor, fertilizer, seed, to contact with DA for better information, better access to credit providers, as a result of their leadership position and, hence, are likely to adopt improved bread wheat varieties better than those who did not have leadership position in the society this result is in line with the result of Rauniyer and Goode (1996) as cited in Legesse (1998). This implies that there need to give attention and identify what those farmers who do not have leadership position lack due to their lower leadership

88

position to design a strategy to provide access, support, encourage them to achieve better adoption of improved bread wheat varieties by them.

The variable, experience in extension influences adoption of improved bread wheat varieties. Farmers who have longer years of experience in extension have adopted better-improved bread wheat varieties than those who have the lower years of experience in extension participation. This showed that the farmers with longer years of experience in extension may use their experience to using and taking the advantages obtained from new agricultural innovations or technologies and also they may develop, the confidence in handling the risk, skills in technology application, and may developed better economical status and better income from out put of using of these improved agricultural technologies.

Regarding, the distance of DAs office, from farmers home showed influential effect in adoption of improved bread wheat varieties in the study area as revealed in this study. The farmers who are nearby the DAs office are likely to adopt the improved bread wheat varieties than those who are far. This implies that the near by farmers to the DAs office would have an opportunity to get better and up dated information on the availability and benefit of improved varieties easily and better than those far farmers. As a result, they can use these opportunities to adopt the improved bread wheat varieties than those farmers far away from DA office.

The remaining hypothesized independent variables were not statistically significant to influence the probability of adoption of improved bread wheat varieties at less than 10% significant level as indicated in Table 38. Even though they were not significant below 10% significant probability level in logit model analysis practical and experience situations, literatures and many research works as well as the test statistics of this study showed that they have influential impact on adoption of improved technologies and innovations. The result of the logit analysis and their change or marginal effect of explanatory variables on dependent variable, probability of adoption of improved bread wheat varieties showed and presented in Table 38.

89

The marginal effect of the variable of the distance of DA-office closer to farmers home by one unit might improve the probability of adoption of improved bread wheat varieties by a factor of more than four times fold. In the social system the farmers leadership position can improve farmers agricultural technologies as observed in this study. In this study farmers leadership position improves the probability of adoption .As there is a change of farmers leadership position from non-leadership to leadership, there is an improvement of adoption of improved bread wheat varieties by the factors of nine times fold. The farmers experience in any of extension activities and use of improved technologies and innovations, can also improve the probability of adoption of improved bread wheat varieties. The changes and improvement of farmers experience in extension participation by one year or by one unit can increase adoption of improved bread wheat varieties by the factor of 61%.

In this study, it was revealed that the market access has a negative relationship with adoption of improved bread wheat varieties. When farmers are closer to market access by one unit, there is a decrease of probability of adoption of improved bread wheat varieties by a factor of 0.2 or by a factor of 20 percent. This implies that as mentioned in the above of this section, farmers who are closer to the market centers and facilities might be influenced and attracted by other substitution factors created by the market center facilities and might inclined to involve in these activities and business tasks with out totally leaving the farming occupation. As a result, they become reluctant to adopt improved bread wheat since improved wheat demand intensive management and labor work.

4.2.2. Analysis of determinants influencing intensity of adoption of improved bread wheat varieties and their marginal effects

Parameter estimates of the Tobit model for the intensity of adoption of improved bread wheat varieties (measured in terms of size of land in hectare used for growing of improved bread wheat varieties over the total wheat land in hectare). The Tobit model was used or applied to analyze the factors that determine the intensity of adoption of improved bread wheat varieties

90

because the mean proportion of land allocated to improved bread wheat varieties is a continuous variable but truncated between zero and one.

The main purpose of this section is to identify the hypothesized independent variables among the selected and proposed to include in the tobit model analysis that significantly influence the dependent variable, intensity of adoption. The result of this study indicated and presented in Table 39. From the total hypothesized independent variables, only eight explanatory variables were significantly influencing and affecting the intensity of adoption of improved bread wheat varieties as presented and indicated in Table 39. These significant variables were, households sex (HHHSEX), age (HHHAGE), education (EDUHHH), health status (HEALSTAT), offfarm income (HHOFFINC), home distance from DA office (DISDAOF1), farmland holding (SUMOWRE) and extension service (GEXSERVE) were statistically the most important explanatory variables affecting intensity of adoption of improved bread wheat varieties in the study area.

The variable household sex was related with the intensity of adoption of improved bread wheat varieties positively and significantly at 5 percent probability level. The sign was in line with that of the expectation. AS it was indicated in the identification of hypotheses, probability and intensity of adoption was expected to relate positively with male sample and negatively related with female sample. Hence, the positive sign indicates that the male-headed households were better in intensity of adoption of improved bread wheat varieties than female farmers. This result showed that male farmers are more likely to allocate larger farmland to improved wheat than female farmers in the study area. This result is in conformity with the finding of (Thechane, 2002).

91

Table 39.The effects of changes (marginal effect) in the significant explanatory variables on the intensity of adoption of improved bread wheat varieties

Variables HHHSEX HHHAGE EDUHHH HEALSTAT FAMILYSI HHOFFINC DISDAOF1 TOTLIVUN SUMOWRE CRINMFF1 OTSOLA1 OXTLU YEXPEXTS GEXSERVE

Coefficient 0.1854436505 0.3862778245 0.1696807312 0.3813275935 0.1544088566 0.1409564409 0.1684878586 -0.8907270032 -0.3801474933 0.1719597193 -0.2466170095 -0.3023853019 0.4834808429 0.2589730719

Standard Error

b/St.Er

P (/Z/>z) 0.0237 0.0132 0.0026 0.0000 0.1845 0.0395 0.0580 0.4131 0.0983 0.9858 0.7590 0.8911 0.3779 0.0196

0.81995580 2.262** 0.15594257 2.477** 0.56390468 3.009*** 0.72099682 5.289*** 0.11636011 1.327 0.68463253 2.059** 0.88875721 1.896* 0.10883167 0.22993869 0.96915193 0.80390554 0.22090082 0.54835151 0.11095968 -0.818 -1.653* 0.018 -0.307 -0.137 0.882 2.334**

(Source Computed from own survey data, 2005) *, ** And***Significant at 10,5 and 1 percent probability level

From these significant explanatory variables only one variable namely size of farmland holding related with the intensity of adoption of improved bread wheat varieties negatively and significantly at 10 percent probability level. This variable has the different sign from that was hypothesized. The remaining of the seven significant explanatory variables namely household sex, age, education, health status, off-farm income, extension service, distance of DA office from farmers home showed statistically significant and positively related with intensity of adoption of improved bread wheat varieties in the study area, at 10 percent probability level.

92

The variable household age was also related with the intensity of adoption of improved bread wheat varieties positively and significantly at 10 percent probability level. As the result of this study showed that older farmers may have already developed better experience, face exposure opportunities with using of large size of improved agricultural technologies through their life experience and might develop experiences how to manage risks and taking of the first benefits from newly released varieties. This might help them to develop confidences to allocate larger farmland to improved bread wheat varieties production more than those lesser and younker age group farmers in the study area. The sign was different from that of hypothesized. The hypothesis formulation and establishment was conducted based on literatures, experiences and observation of actual, practical and existing situations.

Literatures showed that farmers expected to be reluctant to new innovations as their age increased. But in the context of this study area, it is different from that of literatures. In history of Ethiopian extension farmers in the study area have better exposure of opportunities to new agricultural innovations than other areas of Ethiopia. As a result they developed better experience through their life experience better than other areas of farmers who do not get the opportunities like farmers in the study area. This finding agreed with the finding of (Chilot, 1994).

Education was also has a positive and significant relationship with the intensity of adoption of improved bread wheat varieties at 1 percent probability level. In this regard, the proportion of farmland used for growing of improved bread wheat varieties by farmers who are literate is likely to be greater than farmers who were illiterate. This suggests that being literate would improve access to information, capable to interpret the information, easily understand and analyze the situation better than illiterate farmers. So, farmer who are literate were likely to allocate larger size of farmland proportion than those illiterate farmers. The sign was as expected. This result has supported by other previous studies such as the findings of Lelissa (1998), Techane (2002), Lelissa and Mulate (2002), Yitayal (2004).

93

The variable health status had a positive and significant influence at 1 percent probability level relationship with the intensity of adoption of improved bread wheat varieties in the study area.. The result of this study showed that farmers who have better health status are likely to allocate larger farmland size to improved wheat varieties production. It is obvious from practical and actual situation of the ground that managing and operating the improved agricultural innovations demanded intensive labor and management practices. Then, health farmers can do these practices than unhealthy farmers. There fore healthy farmers are likely to allocate larger farmland size than unhealthy farmers. The sign was as expected. Low intensity of adoption by un-healthy farmers may be due to the shortage of labor and the problem to conduct intensive management that the improved bread wheat demanded.

The explanatory variable, off-farm income influenced the dependent variable, intensity of adoption of improved bread wheat varieties positively and significantly at 5 percent probability level as hypothesized under the section of hypotheses description. As it is true that farmers who have better income can adopt new agricultural innovations because their income allowed them to purchase the new technological inputs, can with stand risks if appear and can cover labor costs. Off-farm income is one of the alternatives to improve farmers income. From these grounds of realities farmers who have off-farm income can adopt new technologies in larger proportion than those who do not have off-farm income.

The size of farmland holding, affected the dependent variable, intensity of adoption of improved bread wheat varieties in the study area negatively and significantly at 10 percent probability level. The sign was different from that of postulated. This finding is in conformity with the finding of (Bekele et al, 2000; and Chilot, 1994).

As it is supported by many literatures, those farmers who have larger land size are expected to adopt improved and new agricultural technologies in larger proportions than those farmers who have lesser farmland. Since these farmers have larger farm land they do not have fear of risks, can get credit because they are believed that they can pay their credit, or some part of their land may serve as mortgage to take credit, seed loan from other farmers and can adopt

94

new agricultural technologies than those who have lesser land size. But in this study the result showed the different situations from many literatures even though it is in line with some few literatures as mentioned in the above. The possible reason for this result can be that the actual situations in the study area different from other areas in that the situations of the area where this study has conducted is closer to Addis Ababa, the capital of the country where in the mean time of this study highest construction investment on land were conducted.

Farmers who are closer to the town lease their part of farm land and might reluctant to increase their farm land allocation to improved bread wheat land since they got better income from land contract than they got from improved wheat land production. Another possible reason also for this situation might be that people at the edge boarder of Addis Ababa and the rural part of the study area have the experience of producing crop for their consumption and profit purpose by contracting land from the surrounding farming community. As result of the existence of this situations in the study area farmers might contract their land for these types of par time farmers due to many situations like for better income than they used for improved wheat production or for the reason they may face different problems and cash constraint that could not give some time in the future. Then, those par time farmers might have their own interest of crop type production and objectives. As a result those first owners and adopters of improved bread wheat varieties might unable to increase their farmland allocation for improved bread wheat varieties.

And also in the study area there is an introduction and promotion of white check pea, which has high price in market. This also shift the wheat adopters to allocate their wheat farm land to this new crop variety rather than increasing of their land allocation to the improved bread wheat varieties since wheat land can use for check pea crop interchangeably. It was also observed that credit and input providers greatly reduced their service provisions due to the reluctant effect of farmers to return their previous credit loan as a basic reason of the highest interest rate of the loan. These are some of the possible reasons for the inverse relationship of the independent and dependent variables in this case.

95

Extension service influenced dependent variable, intensity of adoption of improved bread wheat varieties positively and significantly at 5 percent probability level as hypothesized. The finding of this study is in agreement with the findings of Adesina and Zinnah (1992), Chilot (1994), Techane (2002), Lelissa and Mulate (2002) and Yitayal (2004). Theoretical and practical realities showed that extension services provided to the farmers in different forms like training demonstration field day DA visit on the field and on spot field support can motivate, empowers kill and knowledge, increase information access and create interest to improve farmers use of improved agricultural technologies.

The independent variable, distance of DA office from the farmers home, influenced the independent variable, intensity of adoption of improved bread wheat varieties in the study area positively and significantly at 10 percent probability level. The sign was as postulated. The intensity of adoption of improved bread wheat varieties is higher to the farmers who are closer to the DA office than those farmers who found far. This result is in line with the result of (Chilot, 1994). This is also true from theoretical, practical and experience realities when the Das assigned closer to the farmers village farmers can easily and from near by distance can get the required information such as availability of inputs, credit services, market situation government and other development organization supports on time and sufficiently.

The results of the Tobit model analysis also showed the effects of changes or marginal effects in the explanatory variables on the dependent variable, intensity of adoption of improved bread wheat varieties, in the study area as indicated and presented in Table 39.

Literatures showed that adoption and intensity of utilization of improved agricultural innovations has relations with gender. As it was hypothesized male farmers were likely expected to show better intensity of farmland allocation for improved bread wheat production than female farmers. The marginal effect of Tobit model analysis showed that male farmers were better in allocation of farm land as compared to female farmers. The intensity of farmland size allocation for improved bread wheat varieties production by male farmers was larger by a factor of 19 % than female farmers.

96

Age is one factor to influence intensity of adoption. The marginal effect of tobit analysis showed that as age of adopters of improved bread wheat increase by one unit, intensity of farmland size allocation for improved bread wheat varieties production can improve by the factor of 38.63%. As mentioned in the above, since farmers in this area have a long years exposure to extension services than other areas of farmers, they developed better extension experience in the process of their lifetime experience that plays a great role in intensity of adoption of improved bread wheat varieties in the study area.

Education plays a positive and significant role in the intensity of adoption of improved bread wheat varieties in this study. An improvement in education or a change by a unit i.e. from illiterate to literate can improve farmers allocation of farm land for improved bread wheat production from their total wheat land can improved by a factor of 17%. In the other way, as there is an improvement in educational level of adopters of improved bread wheat varieties by one unit, there will be an increased allocation of farm land for production of improved bread wheat varieties by 17%.

In this study it was identified that as farmers health situations improved from unhealthy to healthy situation, the intensity of adoption of improved bread wheat varieties can increases by a factor of 38%, because the health farmers can directly involve in every activities of improved bread wheat production and can them selves manage their farm. As a result the allocation of intensity of farmland for improved bread wheat production can increase by a factor of 38%.

The variable household farmers off-farm income contributes its own part in the intensity of adoption of improved bread wheat varieties in the study area positively and significantly. As the involvement of farmers in off-farm income and consequently their income improved by one unit, their allocation of farmland for improved bread wheat production can increase by the factor of 14%.

97

When the farmers income improves by one unit from off-farm income source, the allocation of farmland by them for improved bread wheat production can improve by a factor of 14%. This is due to the fact that the off-farm income can solve farmers financial constraints and increase their purchasing power of improved bread wheat seed, other agricultural inputs such as fertilizer and other production means relevant to the production of improved bread wheat. Consequently, farmers might be encouraged in allocating larger area of their wheat farmland for the production of improved bread wheat varieties than those who do not have off-farm income.

The distance of DA office plays its role in intensity of adoption of improved bread wheat varieties in the study area positively and significantly. As the distance of DA office decreases or closer to the home of farmers by one unit, intensity of adoption could increase by a factor of 17%. This implies that farmers who are closer to the DAs office can get easy access to extension support and agricultural information that can give a chance to analyze situations and allocate their larger farmland for growing of improved bread wheat varieties than those who are far from DAs office.

As discussed in the above, the independent explanatory variable, farmland holding related with the intensity of adoption of improved bred wheat varieties negatively and significantly as indicated and presented in Table 39. As revealed in this study, when the size of the farm land holding of farmers increased by one unit intensity of adoption of improved bread wheat varieties decreased by a factor of 38%. This may be due to the fact substitution of some part of their wheat land to other highly market demanded crops like for example white chick pea production, land contracting by receiving larger amount of money, and the farmers themselves might involve in other activities and reluctant to allocate increased farm land for improved bread wheat production as a result the result of tobit model analysis showed the inverse relationship between farm land and intensity of adoption of improved bread wheat varieties in the study area.

98

As it is generally true, the better the extension service can improve the intensity of adoption and utilization of improved agricultural technologies. The tobit analysis in this study showed that the change in improving the extension service by one unit can improve the allocation of farm land for improved bread wheat production by adopter farmers can improve by the factor of 26% as indicated in Table 39. This implies that when farmers get support from extension agent, in various forms such as information provision, practical support on the spot of the field or in the form of demonstration, field day and skill development, can improve farmers knowledge, interest, motivation and confidence to allocate larger extent of farm land than those who do not get or who got less extension support.

To summarize the two analytical model results, that the purpose of data analyses using the econometrics models as discussed through out this section and in the previous sections is to know which independent variable most important and powerful to affect the intended dependent variable to which they hypothesized to influence. In this study two-econometrics models logit for identification of factors affecting probability of adoption of improved bread wheat varieties for those non-adopter farmers need to adopt in the future and tobit model for estimation of factors to influence adopters intensity of adoption or allocation of farm land size intensity for improved bread wheat production.

As mentioned at the beginning of this analytical section due to various reasons such as theoretical, actual, practical, technical reasons some hypothesized independent variables were dropped from further analyses like for example leadership position and credit service due to tobit model limitation to accommodate all independent variables included in Logit model analysis did not included in Tobit model analysis. The result of logit model analysis for probability of adoption and result of tobit model analysis for intensity of adoption of improved bread wheat varieties indicated and presented in Tables 38 and 39 respectively.

The most important and significant independent variables below 10% probability level to influence probability of adoption of farmers who did not adopt improved bread wheat varieties in the past but expected to adopt in the future, those identified by logit model analysis were

99

four namely distance of DA-office from the farmers home, house hold head social/leadership position, market accessibility, and house hold farmers experience in extension as indicated in Table 38. Regarding the result of tobit model analysis as indicated in Table 39, eight independent variables namely sex of household head, age, education, health status, off-farm income, distance of farmers home from DA office, farmland holding and extension service were statistically significant below 10% probability level and most important to influence adopters intensity of farm land allocation to improved bread wheat varieties production from their total wheat farm land as indicated in Table 39.

In this study, the two models used for two purposes as mentioned in the above. Though the purpose of this study is not to identify the significant common independent variables among those variables used in the two models analyses to identify influencing significant factors for the two dependent variables as mentioned in the above, it is very important to see whether there is a common influencing independent variables that affect significantly the two mentioned dependent variables. As a result, it was identified that there was only one independent variable, distance of DA-office from farmers home that commonly and significantly affected both probability of adoption and intensity of adoption below 10% probability of significant level. It doesnt mean that the remaining independent variables totally do not have any relationship with the dependent variables rather they are not statistically significant below 10% significant level.

100

5. SUMMARY AND CONCLUSION

5.1. Summary

In this study to identify factors influencing probability and intensity of adoption of improved bread wheat varieties among smallholder farmers the study area, Akaki was selected based on its wide practices of improved bread wheat production and its suitability for this research .In this area, agricultural extension and rural development activities like other rural parts of the country, conducted by Agricultural and Rural Development Unit which comprises several agricultural professionals in different disciplines at office level and Development Agents (DAs) at Center and Peasant Association (PA) level. According to the structural framework of the Addis Ababa Administration, the unit is accounted to and organized under Akaki-kality sub-city.

In this study, data were obtained from 150 randomly selected respondents through personal interview schedule conducted by employed and trained enumerators using pre-tested interview schedule and from group and individual discussions, as well as the researchers personal observations. The respondents, involved in the interview were selected randomly and proportionally from two sample Peasant Associations (PAs), constituted 99 (66%) adopters and 51 (34%) non- adopters.

Data were analyzed, and presented quantitatively using different statistical methods such as percentage, frequency, tabulation, Chi-squaretest (for dummy /discrete variables) and (t-test for continuous variables), Logit, Tobit models and qualitatively through interpretation, explaining, summarizing of ideas and concepts. T-test and Chi-square test were employed to test the variation of the sample group they have towards adoption and also used to describe the patterns of the sample data. Logit and Tobit econometrics models to estimate the effects of hypothesized independent variables they have on dependent variables, probability and

101

intensity of adoption. Computer soft ware package programs such as SPSS and Limdep were employed for statistical analyses.

Among the hypothesized independent variables, sex, health, education, extension service, DAoffice, market access, labor source, off-farm income, leadership, were treated as discrete variables and tested using chi-square-test. In this test the independent variable health, distance of DA-office, labor source, access to off-farm income were not significant below 10% significant level. And family size, years of extension experience, age, livestock ownership, oxen ownership, and farmland holding were considered as continuous variables and tested using t-test. The t-test result showed that except others only age was not significant below 10% significant level.

The t-test and chi-square test results showed that there were variations between adopters and non-adopters sample category in family size, extension experience, livestock ownership, oxen ownership, farm land holding, extension service, sex (gender) and leadership position in adoption of improved bread wheat varieties. According to the result of test statistics male are better in adoption of improved bread wheat varieties. On the other hand adopters have larger family size, livestock ownership, oxen ownership and farmland and they got better extension service than non-adopters. Due to their variation in these independent variables sample farmers vary in their adoption behavior in relation to dependent variables.

Except those hypothesized independent variables dropped due to various cases as mentioned in previous section all screened and verified independent variables were subjected to Logit model analysis. In the case of Tobit model analysis, all verified hypothesized independent variables included in Logit model analysis were not included due to the limitation of the model to accommodate all these independent variables. As a result, leadership position and credit services were dropped from further Tobit model analysis due to their less importance in the study as compared to other independent variables.

102

The Logit model result of this study showed that the significant independent variables affecting probability of adoption were distance of DA office, leadership position of household head, market access and years of house hold heads experience in extension and those independent variables significantly influencing intensity of adoption of improved bread wheat varieties were, household heads sex, age, education, health status, off-farm income, distance of DA office, size of farmland holding, and extension service resulted from tobit analysis. The distance of DAs office from farmers home was the only explanatory variable influencing both adoption and intensity of adoption of improved bread wheat varieties in this study.

The farmers selection and evaluation criteria of improved bread wheat varieties and ranking of the improved bread wheat varieties disseminated in the study area were also conducted through the summary of the survey data, group and individual discussions as well as researchers observation.

In this respect, white grain color, large grain size, straw quality, market demand were the first most important characteristics; germination capacity, cooking quality good yield performance were the second most important; water lodging resistance, tillering capacity, good food quality were the third most important; short maturity date fourth; disease and pest resistance fifth; frost resistance sixth; harvesting quality and storage quality ranks seventh most important characteristics as grouped and ranked based on the result of the survey data group and individual discussions and researchers observation.

Based on the selection and evaluation criteria, the result of the survey summary and group discussion the ranking result of improved bread wheat varieties disseminated in the study area has presented as HAR-1685 variety ranks firs, Paven 76 second and HAR-1709 ranks third.

Farmers in the study area got agricultural information from different sources. The most important information sources as summarized were, neighbors and colleague farmers got the 1st rank, DA and Community leaders the 2nd rank, farmers field day 3rd, PA leaders,

103

demonstration and radio 4th, News paper/News letter and other publications 5th, poster 6th, Training 7th, TV8th and religious leaders ranks 9th sources of information.

104

5.2. Conclusion and Recommendations

In this study several issues were observed and revealed in relation to adoption of improved bread wheat varieties disseminated in the study area, Akaki. The result, description and interpretation of the data were mainly depended on, the context of the research objectives and the situation of the study area.

The truthfulness of the information provided by the sample farmers for this study was also depended on the sample farmers voluntaries and credibility. Since the study area closer to Debrezeit research center, Addis Ababa (the capital of the nation) and subjected to long years of extension services in the past, the result of the study should be seen from this perspectives. This study may serve as an initial input for further study in this and other similar areas of the country.

Like other parts of the country, several agricultural innovations were disseminated in the previous years and extension services were offered to the farmers that have an influential impact on adoption and use of the disseminated agricultural innovations. From those disseminated technologies in this area, improved bread wheat varieties was the one on which this study was focused to identify factors affecting adoption of improved bread wheat varieties by non-adopter farmers, and to identifying other factors influencing the adopter farmers to increase the intensity of farm land size allocation to improved bread wheat production from their total wheat farm land.

Determinants that limit probability of adoption of improved bread wheat varieties were identified using descriptive statistics (t-test and Chi-square test) and logit model analysis. They were gender, extension service; leadership position, market access, farmers extension experience and distance of DA office from farmers home were the influencing factors affecting non-adopters to adopt improved bread wheat varieties in the study area.

105

According to the findings of this research, it is necessary to establish appropriate extension strategy to bring those non-adopters to adopt improved bread wheat varieties. In this regard, attention should be given; to encourage, support and motivate female and less extension experienced farmers to achieve their adoption decision behavior. Some farmers in this area

who have leadership position are better adopters since their position allowed for better access to information, resources and innovations. Therefore, there need to give attention to support those people who do not have resources access opportunities.

As it is confirmed in this study distance of DA office from the farmers home has an influential effect on adoption and intensity of adoption. Therefore, attention should be given to the close assignment and placement of DAs to the rural villages where the farmers can get them easily for extension advises and supports for better adoption.

In the study area, when farmers closer to the market showed reluctant behavior to adopt the improved bread wheat varieties. Some of the possible reasons may be due to weak extension service provided for them or due to fear of intensive management and labor requirement to operate practices, or may be due to substitution effect and their involvement in other par time works etc., created by the market facilities. The market in this area showed a negative effect rather than motivation farmers to adopt improved bread wheat varieties. Therefore, there must be efforts to formulate appropriate extension service for this area, improve market situation for bread wheat and improvements of the varieties qualities for better market demand.

The other aspect of this study was to identify factors influencing those adopters to increase and extend their improved wheat production by allocating larger area of farm land for improved bread wheat production from their total wheat farm land. According to the result of descriptive statistics and tobit model analysis gender, extension service, family size, experience in extension, livestock and oxen ownership, farm land holding, age, education, health, off-farm income, distance of DA-office were factors affecting intensity of adoption of improved bread wheat varieties. Attention should be given to improve farmers intensity of adoption by

106

designing of compatible extension strategy by considering the findings of this research as an input.

Regarding farmers variety selection and evaluation criteria, it is advisable to involve farmers through various techniques like for example using group discussion to evaluate and identify the best suitable varieties that can fit their interest, farming system and environmental situations. In-group discussion farmers from different angels such as gender, age, ecological area and educational levels should be involved to get various ideas and opinions. The idea reflected during group discussions should get attention and need to be incorporated and used in agricultural technologies development, extension programs formulation and policy preparations.

In the study area there is a shift of farmers to involve in improved chickpea production as a result of high price of improved chickpea. There fore, it is necessary to give attention to improve the quality of improved bread wheat varieties that can bring high market demand through breeding and genetics improvement programs. It is also necessary to improve the market facilities for improved bread wheat varieties.

Agricultural information and extension communication are powerful and crucial to achieve better adoption and intensity of adoption of improved agricultural innovations like improved bread wheat varieties in this case. Appropriate and timely information should reach to the intended farmers group to achieve better adoption and intensity of adoption of improved agricultural technologies. Appropriate information and communication strategy compatible with farmers and the study area should be designed and practiced.

Suitable strategies for better extension service are another important issue that should get proper attention. In the study area as observed and the survey data showed, the extension service is at lower and weak position due to various reasons such as transfer of DAs to other lateral offices, low motivation, poor credit service low educational background of extension

107

workers. In these respect it need attentions to solve these problems for better improvements of agricultural technologies adoption and production growths that can bring better living standard of the farmers in the rural areas. Attention also should be given to the research and extension linkages, to the empowerment and training of extension people and farmers, to achieve high level of improvement in adoption of improved agricultural technologies.

108

6. REFERENCES

Adam Bekele and Bedru Beshir, 2005.Adopting Improved Haricot Bean Varieties in the Central Rift Valley of Ethiopia. Ethiopian Agricultural Research Organization (EARO), Addis Ababa, Ethiopia Adams, M.E .1982.Agricultural Extension in Developing Countries, Longman Scientific and Technical Publications Xii, 108p. Adane Nabso Fufa.2002.Magnitude and determinants of rural household poverty in central Ethiopia.The case of Bereh-Aleltu district M.Sc Thesis in Agricultural Economics. Alemaya University, Ethiopia. Adane Nigus Mekonnen.2002.Economic analysis of adoption of soil conservation technologies:A case study of Kobo woreda in North Wollo, M.Sc. Thesis in Agricultural Economics. Alemaya University, Ethiopia. AdesinaA.AandZinnah, M.M.1993.Technology Characteristics, Farmers Perceptions and Adoption Decisions; A Tobit model application in Sierra Leone, Elssevier Science Publishers B.V., Amsterdam,U.S.A. Agricultural Economics.9(1993):297-311p. Adugna Haile, Workineh Negatu and Bisrat REta., 1991.Technology transfer for wheat production in Ethiopia, In Hilu Gebremariam, Tanner Doughlas G. and Mengistu Hulluka (eds). Wheat Research in Ethiopia: A Historical Perspective, Addis Ababa, and IAR/CIMMYT. Amemiya, T.1985.Advanced Econometrics, T.J.Press, Padstaw Ltd. Great Britain. Anderson, J.R. and Feeder, G.2002.The Rural Extension Services, World Bank, Agricultural Development Department, Washington D.C. Arnon, I., 1989. Agricultural Research and Technology Transfer, Elsevier Science Publishers ltd.,London and New York. Asfaw Negassa, Kisa Gunjal, Wilfred Mwangi and Beyene Seboka.1997.Factors Affecting the Adoption of Maize Production Technologies in Bako Area, Ethiopia, Ethiopian Journal of Agricultural of Economics 1(2) 52-72p. Bekele Hundie, Kotu hugo Verkuijl, Wilfred Mwangi and Douglas Tanner, 2000. Adoption of Improved Wheat Technologies in Adaba and Dodola Woredas of the Bale High Lands,

109

Ethiopia, Mexico D.F.:International Maize and Wheat Improvement Center (CIMMYT) and Ethiopian Agricultural Research Organization (EARO). Bisanda Shekania and Wilfred Mwangi.1996.Adoption of Recommended Maize Technologies, in Mbeya Region of the Southern Highlands of Tanzania, CIMMYT/ the United Republic of Tanzania, Ministry of Agriculture. Chandan J.S., 1998. Statistics for Business and Economics, Vikas Publishing house Pvtt Ltd., New Delhi. Chilot Yirga., 1994 .Factors Influencing Adoption of New Wheat Technologies in the Wolemera and Addis Alem Areas of Ethiopia, M. Sc. Thesis, Alemaya University. CIMMYT.1993.The Adoption of Agricultural Technologies: A Guide to Survey Design, Mexico, D.F.: CIMMYT. Curtis B.C. 2002.Wheat in the World. In Curtis Rajarm B.C.S. and Gomez H. McPherson, (Edits), Improvement and Production, FAO, Rome. Destaw Berhanu Nega.2003.Non-farm employment and farm production of smallholder farmers study, in Edja district of Ethiopia., M.Sc. Thesis, in Agricultural Economics, Alemaya University, Ethiopia. EARO.2004.Agricultural technology evaluation, adoption and marketing, Part 2.In: Tesfaye Zegeye,Legesse Dadi and Dawit Alemu .Proceeding of the Workshop held to discuss on The Socioeconomic Research Results of Farmers Participatory Research; Attempts and Achievement in the Central Highlands .1998-2002 August 6-8,2002,Addis Ababa, Ethiopia. Endrias Geta.2003.Adoption of improved sweet potato varieties in Boloso Sore woreda, southern Ethiopia, M.Sc. Thesis, Alemaya University, Ethiopia. ERSHA (Ethiopian Rural Self Help Association).2000.Evaluation of Bread Wheat Technologies on the Farmers Farm Condition Using Farmers Criteria in West Shoa Zone, Ambo Woreda Birbisa and Cherech Service- Cooperative, ERSHA Addis Ababa, Ethiopia. Feder, G. L. Just R.E. and Zilbernran D.1985.Adoption of Agricultural Innovation in Developing Countries; A Survey Economic Development and Cultural Change 32(2): 255 298p. FAO . 1993. Ethiopia. In: FAO:The State of Food and Agriculture .Rome. Franzel Steven. 1992. Features of smallholder farming systems. In Franzel and Helen van houten (edts). Research with Farmers: Lesson from Ethiopia. C.A.B International, UK.

110

Freund, John E.1967. Modern Elementary Statistics, third edition, prentice-hall, Inc.,Engle Wood Cliffs ,New Jersey. Getachew Agegnehu, Brhane Lakew and Kassa Getu.2002.On Farm evaluation of bread wheat varieties at Ginchi Watershed Site, Towards Farmers Participatory Research, at Central High lands of Ethiopia. Proceedings of Client Oriented Research Evaluation Workshop, 1618october2001, Holeta Ethiopia, EARO, Addis Ababa, Ethiopia. Gujarati D.N.1995. Basic Econometrics.3rd (Ed), Mc Graw-Hill.

Gujarati.D.N.1999.Essencial of Econometrics, Mc Graw Hill Company, Singapore, 220p Gujarati, D.N.2003.Basic Economics .4th(ed), McGraw Hill,New York. Hanson H., Noman E.B.and R.G. Anderson.1982.Wheat in the Third World. International West View Maize and Wheat Improvement Center, Press;Boulder, Colorado,USA. Harlan R.J. 1981.The early history of wheat earliest traces to the sack of Rome, In L.T. Evans and W.J. Feacu UK (edits), Wheat Science: Today and Tomorrow, Cambridge University press, Cambridge. Jha. Dayanatha, Behjat Hajjati, and Stephen Vosti.1991.The Use of improved agricultural technologies in eastern province of Zambia, In Celis Rafael, John T. Milimo and Sudlir wanmali, Adopting Improved Farm Technology, A Study of Small Holder Farmers in Eastern Province of Zambia, Rural Development Studies Bureau (University of Zambia), National Food and Nutrition Commission, Eastern Province Agricultural Development Project (Government of the Republic of Zambia), International Food Policy Research Institute (IFPRI), Washington, D.C. Kelejian, H..H..and W.Outes,1981.An introduction to economic analysis.2nd (ed) Horpeer and Row Publishers. Kiflu Bedane and Brhanu Kuma 2002.Farmer participatory research. An Overview .In Gemechu Keneni,Yohannes Gojjam,Kiflu Bedane,Chilot Yirga and Asgelil Dibabe (eds).Towards Farmers Participatory Research ,Central High lands of Ethiopia. Proceedings of Client Oriented Research Evaluation Workshop, 16-18october2001, Holeta Ethiopia, EARO, Addis Ababa, Ethiopia. Langer M.R.H. and Hill G.D.1982. Agricultural Plants, Cambridge University Press, Cambridge.

111

Legesse Dadi.1992.Analysis of Factors Influencing Adoption and the Impact of Wheat and Maize Technologies, In Arsi Negele, Ethiopia, M. Sc Thesis, Alemaya University, Ethiopia. Largesse Daddi .1998.Adoption and Diffusion of Agricultural Technologies: The Case of East and West Shoa Zones, Ethiopia, A Thesis Submitted to the University of Manchester for the degree of Doctor of Philosophy in the Faculty of Economics and Social Studies, School of Economic Studies. Largesse Daddi, Senait Regassa,Asnake Fikre and Demissie Mitiku .2005..Adoption of Chick Pea Varieties in the Central High Lands of Ethiopia, Ethiopian Agricultural Research Organization (EARO), Addis Ababa, Ethiopia. Lelissa Chalchissa and Mulat Demeke.2002.The Determinants of Adoption and Intensity of Fertilizer Use in Ejera District, West Shoa zone, Ethiopia , Institute of Development Research (IDR), Addis Ababa. Lucila,Ma.,A.Lapar and Sushil Pandey.1999.Adoption of soil Conservation :The case of the Philippine Uplands .The Journal of International Association of Agricultural Economics 21(3) 241-256. Maddala, G.S.1992.Introduction to Econometrics, second edition, Macmillan publishing company, New York. Maddala, G.S.1977. Econometrics, Singapore, Mcgrow-hill Book Company. Mergia Beyene .2002 Farmers participatory on farm research: An Alternative approach to agricultural technology promotion. In Gemechu Keneni,Yohannes Gojjam, Kiflu Bedane, Chilot Yirga and Asgelil Dibabe (eds.):Towards Farmers' Participatory Research: Attempts and Achievements in the Central Highlands of Ethiopia . Proceedings of Client-Oriented Research Evaluation Workshop, 16-18 Octobert2001, Holetta Agricultural Research Center, Holetta ,Ethiopia. MillionTadesse and Belay Kassa.2004.Determinants of Fertilizer use in Gununo area, Ethiopian Agricultural Research Organization (EARO), Addis Ababa, Ethiopia. Mullugetta Mekuria.1994.An Economic Analysis of The Smallholder Wheat Production and Technology Adoption in the Southeastern High lands of Ethiopia Ph.D Thesis, Department of Agricultural Economics Michigan State of University USA. Mussei,A., J. Mwanga, W.Mwangi, H. Verkuijl, R. Mongi, and A. Elanga. 2001.Adoption of Improved Wheat Technologies by Small Scale Farmers in Mbeya District, Southern Highlands of Tanzania, Mexico, D.F., International maize and wheat Improvement center (CIMMYTY) and the united Republic of Tanzania.

112

Nanyeenya,William Ntege,Mary Mugisa-Mutetikka, Wilfred Mwangi, and Hugo Verkuijl.1997.An Assessement of Factors Affecting Adoption of Maize Technologies in Iganga District ,Uganda.National Agricultural Reserch Organization/CIMMYT. Pearson C. Lorentz, .1967.Principles of Agronomy, Reinhold, New York. Rogers 1971.Rogers Everett M with F.Floyd Shoemaker.1971.Communication of Innovations; A Cross Cultural Approach, 2nd edition,The Free Press ,New York, USA. Rogers, Everett. M. 1983.Diffusion of Innovations: 3rd Edition .The Free Press, New York, USA. Sarantakos. 1998. Social Research, second edition. Macmillan Press Ltd. London Sherif Aliy Geda. 2001. The Social Nature of Agro-Technological Change: The Trajectory of the BBM Technology in the Joint Vertisol Project in Ethiopia, M. Sc Thesis, Wageningen University and Research Center. Solasya, B.D.S. Mwangi, H. Verkuijl, M.A. Odendo, and J.O. Odenya.1998.An Assessment of the Adoption of Seed and Fertilizer Packages and the Role of Credit in Small Holder Maize Production in Kakamaga and Viliga Districts, Kenya Tanner D. and R. Raemaekers.2001.Wheat: Triticum Spp., In Romain H.Raemalkers (edit.) Crop production in tropical Africa Directorate General for International Cooperation, Brussels, Belgium Techane Adugna Wakjira.2002.Determinants of Fertilizer Adoption in Ethiopia: The Case of major cereal production areas. M.Sc thesis, Alemaya University, Agricultural Economics Tesfaye Beshah .2003.Understanding Farmers: Explain soil and water conservation in Konso, Wollaita, and Wello, Ethiopia, Ph.D. Thesis, Wagneningen University and Research Center. Tesfaye Zegeye ,Bedassa Tadesse and Shiferaw Tesfaye.2001.Adoption of High Yielding Maize Technologies in Major Maize Growing Regions of Ethiopia ,research report no.41.,Ethiopian Agricultural Research Organization (EARO),Addis Ababa, Ethiopia. Tesfaye Zegeye and Alemu Haileye,. 2001. Adoption of Improved Maize Technologies and Inorganic Fertilizer in Northern Ethiopia: Research report no. 40. EARO, Addis Ababa, Ethiopia Tesfaye Zegeye, 2004.Adoption of Inorganic Fertilizer on Maize in Amhara, Oromiya and Southern Regions. In Abebe Kirub (edt.), Agricultural technology evaluation, adoption and marketing Part 2.EARO, Addis Ababa, Ethiopia

113

Vijayaragavan, K. and Singh, Y.P.1997. Managing Human Resources Within Extension .In Swanson, B.E.,Bentz,R.P. and Sofranko,A..J. (Eds.): Improving agricultural extension: A Reference manual, Rome: FAO, p.p.127-134. Yitayal Anley Mengistu.2004.Determinants of Use of Soil Conservation Measures by Small Holders Jimma zone: the case of Dedo District, M.Sc Thesis Alemaya University, Ethiopia.

7. APPENDICES

114

Appendix.1.Information on sample household demographic and socio-economic characteristics

Table 1.The distribution of sample respondents by age group Age (year) 19-30 31-45 46-64 Above or Equal to 65 Total N 11 41 34 13 99 Adopters % 11.11 41.45 34.34 13.13 100 N-adopters N 11 16 15 9 51 % 21.57 31.37 29.41 17.65 100 Total N 22 57 49 22 150 % 14.67 38 32.67 14.67 100

(Source: Computed from own survey data, 2005)

Table 2.Educational statuses of sample house hold head farmers Educational Status Illiterate Read & Write Grade1-6 Grade7-8 Grade9-12 Above grade 12 Total N 64 19 4 5 3 4 99 Adopters % 64.65 19.20 4.04 5.05 3.03 4.04 100 Non-adopters N % 33 64.71 14 27.45 4 7.84 51 100 Total N 97 33 8 5 3 4 150 % 64.67 22 5.33 3.33 2 2.67 100

(Source: computed from own survey data, 2005)

115

Table 3 .The sample household family size Family Size 1 2 3 4 5 6 7 8 9 10 11 Total Adopters N 2 2 8 17 17 18 13 11 5 2 4 99 % 2.02 2.02 8.08 17.17 17.17 18.18 13.13 11.11 5.05 2.02 4.04 100 N-Adopters N 4 11 10 5 6 5 4 5 1 51 % 7.84 21.60 19.61 9.80 11.76 9.80 7.84 9.80 1.95 100 N 6 2 19 27 22 24 18 15 10 3 4 150 Total % 4 1.33 12.67 18 14.66 16 12 10 6.67 2 2.67 100 Total Family Members N % 6 0.72 4 0.48 57 6.80 108 12.87 110 144 126 120 90 30 44 839 13.11 17.16 15.02 14.30 10.73 3.57 5.24 100

(Source: Computed from own survey data, 2005) Table 4.The sample household family size Family Size Maximum Minimum Range Average St.d. Adopters N 11 1 10 5.85 2.192 Non-Adopters N 10 1 9 5.10 2.385 Total N 11 1 10 5.6 2.27

(Source: Computed from own survey data, 2005)

116

Table 5.Total Family members of sample households in age group Age- group 0-14 age 15-64 age Above 64 N 361 452 26 % 43.03 53.87 3.10 100

Total 839 (Source: Computed from own survey data, 2005) Table 6.Respondents farming experience Age group (Year) 1-10 11-20 21-30 Above 30 Total Adopters N % 17 44 21 17 99 17.172 44.444 21.212 17.172 100 Non-adopters N % 9 25 12 5 51 17.65 49.02 23.53 9.80 100

Total N 26 69 33 22 150 % 17.33 46 22 14.67 100

(Source: Computed from own survey data, 2005) Table 7.Types of livestock and owners and the number of respondents Types of Livestock Oxen Cow Bull Heifer Calves Sheep Goat Horse Mule Donkey Poultry Bee-in-Hive Adopters N % 98 68 68.69 46 46.46 41 41.41 39 39.40 48 48.48 8 8.08 21 21.21 25 91 80 5 25.25 91.92 80.81 5.05 Non-adopters N % 42 33 64.71 19 37.25 14 27.45 20 39.22 29 56.86 4 7.84 6 38 29 1 11.76 74.51 56.86 1.96 Total N 137 101 65 55 59 77 8 20 31 129 109 6

% 91.33 67.30 43.30 36.70 37.70 51.33 5.33 16.67 20.67 86 72.67 4

(Source: Computed from Owen survey data, 2005)

117

Table 8.Sample house hold oxen ownership Number Of oxen No oxen One ox Two oxen Three oxen Four oxen Five& above Total Adopters N % 1 3 3.03 46 7 30 10 99 46.47 7.07 30.30 10.10 100 Non-adopters N % 9 3 5.86 25 3 6 4 51 49.02 5.88 11.77 7.84 100 Total N 10 6 71 10 36 14 150

% 4 47.33 6.67 24 9.33 100

(Source: computed from own survey data) Table 9.Sample house hold land ownership Types of Land ownership I. Own Land owners Cultivated Land Grazing land Home stead land Forest land Un-used land II .Shared/Rent land owners Cultivated land Grazing land III Growers of variety HAR-1685 variety HAR-1709 variety Paven-76 variety Adopters N 99 52 18 4 4 20 2 83 5 75 % 100 52.52 18.18 4.04 4.04 20.20 2.02 83.84 5.05 75.76 Non-adopters N 51 29 14 2 3 4 3 1 % 100 56.86 27.45 3.92 5.90 7.84 5.88 1.96 N 150 81 32 6 7 24 2 86 5 76 Total % 100 54 21.33 4 4.67 16 1.33 57.33 3.33 50.67

(Source: Computed from own survey data, 2005)

118

Appendix Table 10.Size of farmland holding of sample household

Farm Size In Ha. <1Ha 1-2Ha 2-3Ha 3-4Ha 4-5Ha 5-6Ha Total

N 12 32 30 18 5 2 99

Adopters % 12.12 32.32 30.30 18.20 5.05 2.01 100

Non-adopters N % 15 29.41 14 27.45 15 29.41 6 11.76 1 1.96 51 100

Total N 27 46 45 24 6 2 150 % 18 30.67 30 16 4 1.33 100

(Source: Computed from own survey data, 2005) Table 11.Respondents average land area and yield of wheat crops in 1996/97 E.C. cropping season. Item/List Area of land in Hectares Total farm Land Total wheat land Total improved B.W.L. Area of Paven -76 Area of HAR-1685 Area of HAR-1709 Area of Durum Wheat Area of Local Wheat Yield of wheat in Quintals Yield of Paven-76 -Yield of HAR-1685 Yield of HAR-1709 Yield of Durum wheat Yield of Local wheat Max 6.5 4.5 3.25 1.75 2.00 2.00 1.00 2.00 25 28 8 15 26 Min 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.50 3 .00 4.00 4.00 2.00 Range 6.25 4.25 3 .00 1.50 1.75 1.75 0.75 1.75 24.5 25 4 11 24. Average 2.44 1.125 1 .00 6.30 0.64 0.535 0.415 0.65 9.12 9.33 5.5 9.165 8.255 St.D 1.13 0.605 0.457 0.29 0.34 0.44 0.285 0.41 4.67 4.445 1.91 7.03 5.365

(Source: Computed from own survey data, 2005) N.B=Improved B.W.L. (Improved Bread Wheat Land)

119

Table 12.Respondents farm land ownership and crop type grown in 1996/97 E.C. cropping season Items/ list T.Farm Land Grazing Land Forest Land Improved Bread Wheat land (sum) - HAR-Paven - HAR-1685 -HAR-1709 -Durum -Local variety T.W.L. Other crop -Teff -Chick pea Lentils -Pea -Faba bean -Vegetables Adopters Ha. N % 309.26 99 100.00 20.45 54 54.55 101.00 101.75 44.75 53.50 3.5 5.5 53.8 161.05 80.25 42.01 2.63 20.38 2.25 0.69 4 99 75 83 5 4 86 99 92 98 9 59 6 6 4.04 100 .00 75.76 83.84 5.05 4.04 86.87 100 92.93 87.88 9.09 59.60 6.06 6.06 N-adopters Ha. N % 90.50 51 100.00 9.95 29 56.86 0.35 1.25 0.50 0.75 28.67 29.92 35.50 17.96 1.25 4.39 1.10 0.38 2 3 48 51 43 38 4 17 3 2 3.92 5.90 94.12 100 84.31 74.51 7.84 33.33 5.88 3.92 Total Ha. N % 399.76 150 100.00 30.40 83 55.33 1.36 100.00 45.25 54.25 3.50 5.50 64.55 190.97 115.75 59.97 3.88 24.77 3.35 1.07 6 99 4 104 150 147 136 76 9 8 5.33 4.00 100.00 2.67 69.33 100 98.00 90.67 13 8.67 50.67 6.00

(Source: Computed from own survey data, 2005)

120

Table 13.Respondents livestock ownership

Types of Livestock Oxen -Adop. - Nadop. Cow -Adop -N-Adop. Bull -Adop - N-Adop Heifer Adop - N-Adop Calf -Adop - N-Adop Sheep Adop - N-Adop Gaot Adop - N-Adop Horse Adop - N-Adop Mule Adop - N-Adop Donkey Adop - N-Adop Poultry Adop - N-Adop Bee-hive Adop - N-Adop

Livestock No. 301 110 85 42 64 22 52 18 42 29 223 106 25 21 4 26 7 178 61 510 131 18 1

Owners No owners 98 42 68 33 46 19 41 14 39 20 80 29 8 21 4 25 6 91 38 80 29 5 1

Owners (% ) 98.99 82.353 68.687 64.706 46.465 37.255 41.414 27.451 39.394 39.216 80.81 56.863 15.69 21.212 7.843 25.253 11.765 91.92 74.51 80.81 56.863 5.051 1.961

max

Min

Range

average per Owners 3.07 2.62 1.25 1.27 1.40 0.86 1.27 1.286 1.08 1.45 2.79 3.65 3.125 1.00 1.00 1.04 1.17 1 .96 1.60 6.37 4.572 3.60 1.00

10 6 3 3 3 2 3 2 3 3 18 13 5 1 1 2 2 5 4 45 12 10 1

1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1

9 5 2 2 2 1 2 1 2 2 17 12 3 1 1 4 3 44 11 9 -

(Source: Computed from own survey date, 2005)

121

Table 14.Respondents livestock ownership in Tropical Livestock Unit (TLU) Types of Livestock Oxen Cow Bull Heifer Calf Sheep Goat Horse Mule Donkey Poultry Total By- Adopters 301 85 48 39 10.50 29 3.25 23.10 28.60 124.60 6.63 698.68 Number of Livestock owned By- N- Adopters Total No. of livestock 110 411 42 16.50 13.50 7.25 13.78 4.40 7.70 42.70 1.703 259.533 127 64.50 52.50 17.75 42.78 3.25 27.50 36.30 167.30 8.333 958.213

(Source: Computed from own survey data, 2005) Table 15.Conversion factors used to estimate the households livestock ownership into tropical livestock units (TLU) Animals Calf Heifer & Bull Cows & Oxen Horse Donkey Ship & Goat Chicken/poultry Source: Strock et al., (1991) TLU-equivalent 0.25 0.75 1.00 1.10 0.70 0.13 0.013

122

Table 16.Discrete characteristics of respondents

Non Adopters (51) Characteristics Number Level of Education Illiterate Literate Read & Write Elementary School Junior Secondary High School House hold sex -Male -Female Health Status -Un-healthy -Healthy Access to credit -Yes -No Leadership /Social status -Yes -No Off-farm income -Yes -No Distance of credit institutions -Far -Close Distance of DA office -far -close Market Access -Far -Close Access to labor -No access -Employment and other Sources Access to Extension Service -Yes -No Education Level -Illiterate -Literate 33 4 14 11 40 7 44 43 8 48 3 42 9 43 8 48 3 43 8 48 3 51 18 33 Percent 64.71 7.84 27.45 21.57 78.43 13.73 86.27 84.31 15.69 94.12 5.88 82.35 17.65 84.31 15.69 94.12 5.88 84.31 15..69 94.12 5.88 100 35.29 64.71 Number 64 4 5 3 4 19 7 92 7 92 71 28 73 26 82 17 89 10 12 87 89 10 87 12 94 5 35 64 Percent 8.138 64.65 4..04 ..05 3..03 4..04 5..05 6.700*** 7.07 92..93 1.762 7.07 92.93 2.928* 71.72 28.28 8.965*** 73..74 26.26 0.005 82.83 17.17 0.994 89..90 10.10 1.456 12.12 87.89 0.994 89.90 10.10 1.456 87.88 12.12 2.665* 94.95 5.05 0.000 35.35 64..65 1 0.994 0.001 1 0.103 0.132 1 0.228 0.098 1 0.319 0.081 1 0.228 0.098 1 0.319 0.081 1 0.942 0.006 1 0.003 0.237 1 0.087 0.138 1 0.184 0.108 1 0.010 0.207 5 Adopters (99) Signifi cance (2sided) 0.149 Conting ency coeffici ent 0.227

X2

df

***, ** and* Significance at P<0.01, P<0.05 and p<0.10 respectively.

123

Table 17.Respondent farmers general information Adoption Category Adopters Summary of statistics Mean St.D Minimum Maximum Range Mean St.D Minimum Maximum Range Mean St.D Minimum Maximum Range House hold head Age 46.1010 13.2560 19.0000 80.0000 61.0000 46.4706 14.5305 20.0000 80.0000 60.0000 46.2300 13.6550 19.0000 80.0000 61.0000 House hold family size 5.85 2.19 1.00 11.00 10.00 5.10 2.39 1.00 10.00 9.00 5.59 2.28 1.00 11.00 10.00 Farmers extension experience (Ys) 7.8687 4.7866 2.0000 20.0000 18.0000 3.7647 1.7842 1.0000 9.0000 8.0000 6.4730 4.4850 1.0000 20.0000 19.0000 Farming experience 21.8990 11.0809 2.0000 55.0000 53.0000 20.7647 10.9829 4.0000 60.0000 56.0000 21.5130 11.0230 2.0000 60.0000 58.0000

Non-adopters

Total

(Source: Computed from own survey data)

124

Table 18.Factors affecting Intensity of adoption of improved bread wheat varieties (Maximum Likelihood Tobit Model Estimation) Variables HHHSEX HHHAGE EDUHHH HEALSTAT FAMILYSI HHOFFINC DISDAOF1 TOTLIVUN SUMOWRE CRINMFF1 OTSOLA1 OXTLU YEXPEXTS GEXSERVE
Sigma

Coefficient 0.1854701150 0.3863329500 0.1697049462 0.38138220125 0.1544308922 0.1409765567 0.1685119034 -0.8908541184 -0.3802017439 0.1719842596 -0.2466522041 -0.3024284552 0.4835498401 0.2590100298

Standard Error 0.82006662 0.15596286 0.56398485 0.72107639 0.11637647 0.68473026 0.88888172 0.10884712 0.22997223 0.96902492 0.80402014 0.22093233 0.54842929 0.11097534

b/St.Er. 2.262 2.477 3.009 5.289 1.327 2.059 1.896 -0.818 -1.653 0.018 -0.307 -0.137 0.882 2.334

P (/Z/>z) 0.0237** 0.0132** 0.0026*** 0.0000*** 0.1845 0.0395** 0.0580* 0.4131 0.0983* 0.9858 0.7590 0.8911 0.3779 0.0196**

0.2268484851

0.15882571

14.283

0.0000

Log likelihood function= 6.582514 *, **And*** indicate the level of significance at 10%, 5% and 1% respectively.

125

Appendix.2. Interview Schedule for data collection from. Farmers

The objective of this Interview Schedule is to collect information from farmer respondents on improved bread wheat production in Akaki area, rural part of Akaki Kaliti sub-city of Addis Abeba administration from December/ 2004 to March/ 2005. The study is conducted for academic purpose. Hence, we request your honest & fair responses to fill up this interview schedule. 1. General & personal information of the respondent 1. Respondents name.. 2. Sex; 0 = female 1 = Male

3. Age..years 4. Marital statuses; 1.Married, 2.Single or unmarried, 3.Divorced, 4.Widow/Widower. 5. Rural Kebelie Administration/ Peasant Association Village. 6. Previous or current leadership status; 0 = No, 1 = Yes 7. Educational Status: 0 = Illiterate, 1 = Literate 8. Educational level: 1. Read & Write, 2.Grade 1- 6, 3.Grade 7- 8, 4. Grade 9- 12, 5.above grade 12 9. Household Characteristics Information Table 19.Household characteristics No Name of house hold members Sex Age Educational status

126

10. Land holding and farm characteristics of the sample households Table 20.Land holding & Farm Characteristics of the sample households No 1 2 3 4 5 6 Types of land use Cultivated (farm) land Grazing land Homestead land Forest land Unused land Total land holding Own (ha) Rent (ha) Total (ha)

11. Livestock ownership Appendix Table 21.Livestock ownership

No 1 2 3 4 5 6 7 8 9 10 11 12

Types of Livestock Ox Cow Calf Bull Heifer Horse Mules Donkey Goats Sheep Chicken Bee in Hive

Number

12. Types of crop grown in the survey year

127

Table 22.Types of crop grown in the survey year No Types of crops Land in (ha)

13. Involvement in irrigation production; 0 = No, 1 = Yes 14. Land size for irrigation production---------- ha. 15. Involvement in improved bread wheat production: 1.only this year 2.This year & in the previous year 3.In the previous year but not this year 4. Never involve. 16. Reasons for involvement: 1.High yield 2.High market demand quality 3.Pest/disease resistance 4.Frost resistance 5.Short maturity date 6.High food quality 7. Good storage quality 8. Good quality of cook ability 9. Good straw quality 10.Seed availability 11.Seed availability 12. Good information service 13. Fertilizer availability 17: Reasons for un-involvement: 1. Low yield 2.Low market demand 3.Low pest/ disease resistance 4.Low frost resistance 5.Long maturity date 6.Poor food quality7. Poor storage quality 8.Poor cooks ability 9.Poor straw quality 10. High seed price 11.Shortage of seed 12.Shortage of fertilizer 13.Lack of information 14.Lack of money and credit 15.Late arrival of seed 16.Late arrival of fertilizer 17.High interest rate of credit 18. Reasons for discontinuity: 1.Poor yield performance 2.Poor pest/ disease resistance3. Poor market demand 4. Poor frost resistance 5.Poor storage quality 6.Long maturity date 7. Poor cook ability 8.Poor straw quality 9.Poor food quality 10. High seed price 11. Seed shortage, 12.Fertilizer shortage 13.Poor extension supports 14.Late arrival seed 15.Late arrival of fertilizer, 16.Lack of money & credit 17.High interest rate of credit. 19. Total farm land/ cultivated land ----- ha.

128

20. Total wheat land---- ha 21. Land for improved bread wheatha. 22. Do you know paven-76? 0=no; 1=yes 23. Do you know HAR- 1685 (Kubsa)? 0=no; 1=yes 24. Do you know HAR- 1709 (Mitike)? 0=no; 1=yes

25: Use of disease and pest control chemical: 0 = No, 1 = Yes

26. Your future plan of involvement in improved bread wheat production 0 = Discontinue, 1 = Continue 27. Presence of problems related to fertilizer: 0 = No, 1 = Yes 28. Problems related to fertilizer: 0 = No, 1 = Yes 29.If yes, 1.High fertilizer price 2.Lack of credit to purchase fertilizer 3.High interest rate of credit to use credit to purchase fertilizer 4.Far distance of distribution center 5.Poor quality (mixed with impurities and caked) 6.Shortage 7.Lately arrival 8.Lengthy process & complicated format 9. Poor distribution processes 30. Presence of problems related to improved bread wheat seed: 0 = No, 1 = Yes 31. Types of problems: 1.Shortage 2.Poor seed quality 3.Late availability 4.Far distance of distribution center 5 Impurity problems and 6.Poor germination problem.

32. Extension support: 0 = No, 1 = Yes 33. Extent of extension support: 1. Poor 2. Medium 3.Good

34. Improved wheat seed rate application: 1.The recommended rate 2. Below the recommended rate 3.Above the recommended rate

129

35. Fertilizer rate of application 1: Apply the recommended rate, 2: Below the recommended rate 3. Above recommended rate 36. Chemical application: 1. Apply the recommended rate 2. Below the recommended rate, 3.Above recommended rate 37. Reasons for Below & Above recommendation use of agricultural inputs 1.Low quantity of input availability 2.High price of inputs 3.High interest rate of credit 4.Lack of credit & money 38. Frequency of weeding: 1.One 2.Two 3.Thrice 4. Four & above 39. Frequency of plowing: 1. One 2.Two 3.Three 4. Four & above

40. Characteristics of improved bread wheat varieties: (1 = High, 2 = Medium, 3 = Low) Appendix Table 23.Improved bread wheat varieties characteristics No 1 2 3 4 5 6 7 8 9 Characteristics Frost resistance Pest/Disease resistance Seed size Cocking time Storage quality Yield performance Market demand Food quality Color quality Varieties HAR-1685 KAR-1709

Fovon-76

41. Seed selection criteria: 1.Pest/Disease resistance 2.Frost resistance 3.High yield performance 4.high market demand 5.Attractive color 6.Short maturity data 7. Good food quality 8.Low time taking 9. Good straw out put and good quality 10. Good storage quality 11. Good Germination and till ring capacity 42. Improved bread wheat seed source: 1.Purchase from market, 2.Exchange from other owners, 3.Own seed from previous product, and 4. Borrow from owner formers, 5.Cooperative, 6.MOA 7. Seed enterprise, 8.Research organization

130

43. Other Agricultural input sources: 1 Cooperatives, 2.MOA, 3.others 44. Access to credit service: 0= No 1: yes 45. Credit sources: 1.Cooperatives, 2.Ethiopian 3. MOA, 4.Other- Credit Institution, 5 Individual/ private lenders 46. Presence of credit problems: 0=no 1=Yes 47. Types of credit problems: 1.Shortage 2.Long and complex process, 3.high interest rate, and 4. Far distance 48. Support from relatives and other colleagues to solve financial constraints to purchase inputs: 0=no 1=yes 49. Distance of credit providers Institutions=far 50. Do you have Access to market? 0=no 51. Market distance=far 1=close 52. Do you have Access to extension Service? 0=no 1=yes 53. Distance of Development Agent Office: 0=far 54. Have you attended training? 0=no 1=yes 1=yes 1=close 1=yes 1= close

55. Have you attended demonstration and field day programs? 0=no

56. Can the DA call the farmers for extension meeting with out the permission of government authorities? 0=no 1=yes 57. What did you feel when called for extension meeting? 0=un-happy 1=happy 58. What are your Sources of Agricultural and input information sources? 1. DA 2. Radio, 3.Television, 4.Written materials, 5.Training, 6.Field, day and demonstration, 7.Posters, 8.PAleaders, 9.Community leaders, 10.neighbours and colleague farmers,

131

59. From the following, to which one assign your- self? 1. Mode farmer, 2.follower farmer, 3.Neither of them 60. How many years of Experience do you have in agricultural extension? Years 61. Total farming experience in years? ...................... Years 62. Have you got training and sufficient information on improved bread wheat? 0=no 1= yes 63. If you did not get training how did you perform production operations? 1. Using try and error methods, 2. By copying from other experienced farmers, 3. By asking support from DA 64. Purpose and use of off-farm income: 1.for house hold food consumption and other costs, 2.for input purchase, 3.for labor hiring, 4.for health cost covering, 5.for all 65. Do have access to labor outside the household labor? 0=no 1=yes

66. If yes, your sources of labor out side the household labors: 1. Hired labor 2. Cooperation labor from colleague and relative framers, 3. Exchange labor 67. Who make the decision on off-farm income? 1. Family head, 2. Husband 3. Wife, 4.Husband and Wife, 5.The household members 68. Which type of agricultural operation is critical to you and need higher labor? 1. Plowing, 2.Sowing, 3.Weeding, 4.Harvesting, 5. Threshing 69. Do you have plowing oxen? 0=no 1=yes

70. If no, how do you plow your farmland? 1. Using oxen plowing, 2.Through labor exchanges 3. By asking cooperation

132

Table 24.Cramers V and Pearsons R values for Discrete and Continuous variables

Hypothesized independent variables and their values Continuous variables Pearsons R value for Discrete variables Cramers V-value for Continuous variables discrete variables Age 0.257 Sex 0.211 Family size 0.167 Education 0.001 Farm land Livestock ownership in (TLU) Oxen ownership 0.292 0.257 0.247 Health status Leadership-position Off-farm income Distance of DA-office Extension service Other labor source Market access Credit service 0.108 0.244 0.006 0.099 0.133 0.099 0.081 0.140

(Source: Own computation) *Notice: 0 value=no association, 0-0.4 value= weak, 0.4-0.7= moderate and >0.7= strong association ((Sarantakos, 1998).

Table 25.Respondents leadership position Have leadership position Yes No Total (Source: own computation) ADs 48 (94.12%) 3 (5.88%) 51(100) NADs 73 (73.74%) 26 (26.26%) 99 (100) X2-test Total

121 29 8.965*** 150

133

ASSESSMENT OF FARMERS EVALUATION CRITERIA AND ADOPTION OF IMPROVED BREAD WHEAT VARIETIES

M. Sc. Thesis

DEREJE HAMZA MUSSA

December 2005 Alemaya University

ASSESSMENT OF FARMERS EVALUATION CRITERIA AND ADOPTION OF IMPROVED BREAD WHEAT VARIETIES

A Thesis Submitted to the Department of Rural Development and Agricultural Extension, School of Graduate Studies ALEMAYA UNIVERSITY

In Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE IN AGRICULTURE (AGRICULTURAL EXTENSION)

By Dereje Hamza Mussa

December 2005 Alemaya University

APPROVAL SHEET OF THESIS

SCHOOL OF GRADUATE STUDIES ALEMAYA UNIVERSITY

As members of the Examining Board of the Final M. Sc. Open Defense, We certify that we have read and evaluated the thesis prepared by DEREJE HAMZA MUSSA and recommend that it be accepted as fulfilling the thesis requirement for the degree of MASTER OF SCIENCE IN AGRICULTURE (AGRICULTURAL EXTENSION) . Name of Chairman Signature Date

. Name of Internal Examiner

Signature

Date

. Name of External Examiner

Signature

Date

Final approval and acceptance of the thesis is contingent up on the submission of the final copy of the thesis to the council of the Graduate Studies (CGS) through the Departmental Graduate Committee (DGC) of the candidates Major Department. I here by certify that I have read this thesis prepared under my direction and recommend that it be accepted as fulfilling the thesis requirement. Ranjan S. Karippai (Ph.D ) Name of Thesis Advisor .. Signature . Date

ii

DEDICATION

I dedicate this thesis manuscript to my wife, TIRUWORK ABATE and my son, SOLOMON DEREJE, for their love and untold-enormous partnership effort in my academic success.

iii

STATEMENT OF AUTHOR

First, I declare that this thesis is my bona fide work and that all sources of materials used for this thesis have been duly acknowledged. This thesis has been submitted in partial fulfillment of the requirements for an advanced M. Sc. degree at Alemaya University and is deposited at the University Library to be made available to borrowers under rules of the Library. I solemnly declare that this thesis is not submitted to any other institution anywhere for the award of any academic degree, diploma, or certificate.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the School of Graduate Studies when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

Name:

DEREJE HAMZA MUSSA

Signature:

Place: Alemaya University, Alemaya Date of Submission: December 2005.

iv

BIOGRAPHICAL SKETCH
Dereje Hamza Mussa was born in Jamma District (Sora-Micha village), South Wollo Zone, and Amhara region on August 6, 1964. He attended his elementary and junior education at Boren -Teklehaimanot and Jamma-Degollo elementary and junior schools (both found in my district) respectively. He also attended his High-school education at Woreilu secondary Senior high school. After completion of his high school education, he joined Awassa Junior Agricultural College (under Addis Ababa University) (now Debub University) to attend a two years Diploma program in Animal Science and Technology. After graduation he was employed in Ministry of Agriculture and has worked for more than 15 years. After this much time service he got an opportunity to join at Alemaya University to attend his degree program education in Agricultural Extension in Mid-career program from 1999 to 2002. After completion and graduation his education he turned back and joins at Alemaya University to attend his M. Sc. degree education in Agricultural Extension Since 2004.

ACKNOWLEDGEMENT

First and for most, I am greatly indebted to Ranjan S. Karippai (Ph.D) my major advisor and Senait Regassa (Ph.D) my co-advisor for their unreserved help, advice, directing, insight guidance, support on the field, critical review of my thesis manuscript, invaluable support and suggestions as without their professional help it was difficult to be successful in my research work and Thesis write up; in addition, my acknowledgement should forwarded to Dr. Ranjan for his professional and critical review and Dr. Senait for her help in SPSS and Limdep computer soft wares as well as Logit, Probit and Tobit, econometrics models, t-test and x2-test statistics uses and application. My sincere thanks should also go to Tesfaye Lema (Ph.D) and Tesfaye Beshah (Ph.D) for their unlimited review of my thesis manuscript help, guide and continues encouragement to be successful in my study and research.

Above all, I am greatly indebted to Ato Zewdu Teferi and his children (Eyu, David and Dani), Seid Ahmed (Ph.D), Solomon Asseffa, (PhD) for their greatest financial and material contribution as well as moral encouragement and all sided help.

My thanks and appreciation should also extend to many individuals, to Belayneh Leggesse (Ph.D), Prof. Panjabin, Asegdew Gashaw with his wife, Mehadi Egi, Ato Walelign and his wife Abaye, Bizuhayehu Asfaw and Amare Berhanu and Tewodros Alemayhu from Alemaya university; to Ato Yishak Berado, Amsalu Bedaso, Admassu Terefe and his wife Belay with their children Eyuti and Mitisha from Alage technical agricultural college; to Ato Chane Gebeyhu, Ato Gebyhu, Alemayehu and Tekle from Akaki-Kality subcity and agricultural unit; to woreilu wereda agricultural office staff members, and Lulseged Bekele, Mohammod Yimer , Abebaw gidelew and (Ato Eshetu Woraei and w/o Toyiba -through their every day pray) from woreilu woreda ; to Ato kasye Afre and his wife Turye Getye with their family, particularly Mamush, Zelalem and Sisaye Kasye-they always accompanied me to and from bus station of Addis Ababa in my every travel to or arrival from Alemaya University for academic and research purposes; to Wondye Kasye, Tesafa Belay, Alemye Argaw, Esubneh

vi

Checolle, Kasye Mohammed and to my brothers Endale and Eshetu Hamza should deserve acknowledgement for their moral, financial and material as well as all sided helps, wishes and encouragements to accomplish my study successfully.

My sincere and special thanks should go to Jifar Tarekegn and Yodit Fekadu for their free charge and complete computer, office provision and all sided co-operations; particularly, Jifar Tarekegn for his additional and unlimited helps in computer and statistical manipulation through out my thesis write up.

My heart felt and deepest thanks should go to Tiruwork Abate and Solomon Dereje my wife and my son respectively who received and paid all suffers and scarifications but the greatest contributors and partnership in my research and academic success.

I would like to extend my thanks to my mother-Ayelech Sebsibe, and my wife's mothers Alganesh Afre, who are always with me in help and wish for my success through their everyday pray.

Several organizations, Alemaya University, School of Graguate Studies, Department of Rural Development and Agricultural Extension of Alemaya University, Debre-Zeit Research Center and Agricultural Unit of Akaki-Kality sub-city, Alage Agricultural College should deserve acknowledgement for their contributions to my study. At last but not the least, I would like to extend my acknowledgement to IFAD (International Fund for Agricultural Development) that offered a budget support for this research through EARO and Debrezeit research center.

vii

LIST OF ABBREVIATIONS

AU B.B.M. DA EA EARO EARI IR PA RKA

Alemaya University Broad Bed Molder Development Agent Extension Agent Ethiopian Agricultural Research Organization Ethiopian Agricultural Research Institute Institution of Research Peasant Association Rural Kebele Administration

viii

TABLE OF CONTENTS

APPROVAL SHEET OF THESIS DEDICATION STATEMENT OF AUTHOR BIOGRAPHICAL SKETCH ACKNOWLEDGEMENT LIST OF ABBREVIATIONS TABLE OF CONTENTS LIST OF TABLES LIST OF TABLES IN THE APPENDIX LIST OF TABLES IN THE APPENDIX ABSTRACT 1. INTRODUCTION
1.1. Background 1.2. Statement of the Problem 1.3. Objectives of the Study 1.4. Significance of the Study 1.5. The Scope and Limitations of the Study 1.6. Organization of the Thesis

ii iii iv v vi viii ix xi xiii xiii xiv 1

1 2 4 4 5 6

2. LITERATURE REVIEW
2.1. Concept and Theoretical Framework of Adoption 2.2. Empirical Studies on Adoption 2.3. Farmers Participation in Agricultural Technologies Development and Evaluations 2.4. Conceptual Framework of the Study

7
7 12 19 23

3. RESEARCH METHODOLOGY
3.1. Description of the Study Area (Akaki) 3.1.1. Location, relief and climate 3.1.2. Agriculture and demographic characteristics of the study area 3.1.3. Institutional services of the study area 3.2. Description of Data Collection and Data Analysis Methods and Procedures 3.2.1. Sources and types of data 3.2.2. Sample size and sampling techniques 3.2. 3. Data collection methods

27
27 27 30 31 34 34 35 36

ix

TABLE OF CONTENTS (Continued)

3.2.3.1. Quantitative data collection methods 3.2.3.2. Qualitative data collection method 3.3. Analytical Models 3.3.1. Logit model 3.3.2. Tobit model 3.3.3. Other Quantitative data analysis methods 3.3.4. Qualitative data analysis method 3.4. Hypotheses Testing and Definitions of Variables 3.4.1. The Dependent variables of logit and tobit models 3.4.1.1. The Dependent variable of logit model 3.4.1.2. The Dependent variable of tobit model 3.4.2. The Independent variables and their definitions used in logit and tobit models

36 37 37 37 40 42 43 43 44 44 44 44

4. RESULTS AND DISCUSSION

50

4.1. Analysis through descriptive statistics 50 4.1.1. Sample Households Demographic Characteristics 50 4.1.2. Respondents` livestock and land ownership 60 4.1.3. Accessibility of respondents to different institutional services 63 4.1.4. Agricultural information sources of the study area 72 4.1.5. Farmers selection and evaluation criteria of improved bread wheat varieties 76 4.2. Analytical results and discussion 79 4.2.1. Analysis of determinants influencing probability of adoption of improved bread wheat varieties and their marginal effect 85 4.2.2. Analysis of determinants influencing intensity of adoption of improved bread wheat varieties and their marginal effects 90

5. SUMMARY AND CONCLUSION

5.1. Summary 5.2. Conclusion and Recommendations

101
101 105

6. REFERENCES 7. APPENDICES
Appendix.1.Information on sample household demographic and socio-economic characteristics Appendix.2. Interview Schedule for data collection from. Farmers

109 114
115 126

LIST OF TABLES Tables Pages

1.The Livestock and crop types in the study area 2.The land use of farmers in the study area 3.The summary of oxen ownership 4.Improved agricultural input distribution of the study area in different years 5. Improved bread wheat seed distribution of the study area in different years 6.Credit Distribution of the study area in various years 7.Sample household heads distribution by Sex, Kebele and adoption category 8.Marital status of respondents 9.Association of adoption of improved bread wheat and sex of sample household head 10. Respondent farmers demographics characteristics 11.Adopters and non-adopters demographic characteristics 12.Reasons given for not using improved bread wheat varieties 13.Level of awareness of improved bread wheat varieties 14.Sample Farmers perception on benefit of fertilizer 15.Beginning time of cultivation of improved bread wheat varieties of sample farmers 16.Health status and adoption of improved bread wheat varieties 17.Sample household educational status 18.Livestock and land ownership of respondents farmers 19.Respondents land ownership in 1996/97 Ethiopian major cropping season 20.Respondents opinion on extension service of the study area 21.Extension support on improved bread wheat varieties and distance of DAs office 22.Summary of respondents opinion on credit 23.Association between credit and market service 24.Summary of households accessibility of off-farm job 25.Respondent farmers reasons for not involvement of their family in off-farm job 26.Rrespondents opinion on decision of off-farm and other household resources 27 Pattern of off-farm income utilization of respondent farmers

29 30 31 32 33 34 51 52 53 53 54 56 56 57 58 58 59 61 62 63 64 65 66 67 68 68 69

xi

28.Family labor utilization of respondent farmers 29.Types of activities and family labor utilization of respondents 30.Respondents accessibility to non-family labor and to off-farm income 31.Respondent farmers labor sources outside their family members 32.Respondents participation in training, field day and demonstration 33.Respondent farmers sources of information 34.Farmers evaluation and selection criteria of improved bread wheat varieties 35.Farmers preference (selection and evaluation criteria) of improved bread wheat varieties disseminated in the study area 36.Variable Inflation Factor for the continuous explanatory variables

70 70 71 72 74 75 77

78 84

37.Contingency Coefficients for Dummy Variables of Multiple Linear Regressions Model 84 38.Factors affecting Probability of adoption of improved bread wheat varieties and the 39.The effects of changes (marginal effect) in the significant explanatory variables on the intensity of adoption of improved bread wheat varieties 92 88

xii

LIST OF TABLES IN THE APPENDIX

Appendix Tables 1.The distribution of sample respondents by age gro 2 Educational statuses of sample house hold head farmer 3 .The sample household family size 4.The sample household family size 5. Total Family members of sample households in age group 6.Respondents farming experience 7.Types of livestock and owners and the number of respondents 8 .Sample house hold oxen ownership 9. Sample house hold land ownership 10.Size of farmland holding of sample household

Pages 115 115 116 116 117 117 117 118 118 119

11.Respondents average land area and yield of wheat crops in 1996/97E.C.cropping season119 12 Respondents farm land ownership and crop type grown in 1996/97 E.C.cropping season120 13.Respondents livestock ownership 14 Respondents livestock ownership in Tropical Livestock Unit (TLU) 15. Conversion factors used to estimate the households livestock ownership in tropical livestock units (TLU) 16. Discrete characteristics of respondents 17. Respondent farmers general information 18. Factors affecting Intensity of adoption of improved bread wheat varieties (Maximum Likelihood Tobit Model Estimation) 19. Household characteristics 20. Land holding & Farm Characteristics of the sample households 21. Livestock ownership 22. Types of crop grown in the survey year 23. Improved bread wheat varieties characteristics 24. Cramers V and Pearsons R values for Discrete and Continuous variables 25.Resondents leadership position 125 126 127 127 128 130 133 133 122 123 124 121 122

xiii

ASSESSMENT OF FARMERS EVALUATION CRITERIA AND ADOPTION OF IMPROVED BREAD WHEAT VARIETIES

Major advisor: Ranjan S. Karippai (Ph.D) Co-advisor: Senait Regassa Bedadda (Ph.D)

ABSTRACT

Wheat is beneficial to man long before the dawn of recorded history. Ethiopia is one of the largest wheat producers in sub-Saharan African next to South Africa. Wheat is one of the most important cereal crops grown in the study area, Akaki. It contributes to the major share of daily consumption and cash source. The objectives of this study were: to identify farmers evaluation and selection criteria of improved bread wheat varieties disseminated in the study area; to assess probability and intensity of adoption of farmers in the study area; and to know and analyze determinants of probability and intensity of adoption of improved bread wheat varieties in the study area. In this study, data were collected and analyzed qualitatively and quantitatively. Quantitative data analysis methods employed in this study were (percentage, tabulation, t-test and X2, Logit and Tobit models) using SPSS and Limdep computer soft ware programs and qualitatively through group discussion and observations.. In farmers evaluation and selection criteria of improved bread wheat varieties disseminated in the study area HAR1685 ranks first, Paven-76 second and HAR-1709 third. White color, large grain size, market demand, straw quality were the most important, germination capacity, cooking quality, better yield performance were the second important, water logging resistance, tillering capacity, good food quality, short maturity date the third important, disease and pest resistance and frost resistance the fourth, storage and harvesting quality were the fifth important quality were identified as a selection and evaluation criteria of improved bread wheat varieties in the study area. Out of the total 150 samples, adopters were 99(66%) and non-adopters were 51(34%). In determining factors influencing probability of adoption through logit analysis, distance of DA-

xiv

office from farmers home, leadership status of respondent farmer, market accessibility and sample farmers experience in extension were identified and (b) intensity of adoption through tobit analysis, house hold sex, age, education, health status, off-farm income, distance of DA office, size of farm land holding and extension service were identified. To enhance probability and intensity of adoption, closer placement of DAs, encouragement of those farmers having less education, female farmers, popularization of improved varieties, improving varieties qualities of characteristics and farmers environmental, economic and social situation should get a serious consideration.

xv

1. INTRODUCTION

1.1. Background

Grain cultivation and the intensive utilization of wild grains in the horn of Africa probably began by or even before 1300 B.C. However, modern agricultural technologies and crop improvement activities to increase grain production have been introduced to the region very recently. Wheat was made beneficial to man long before the dawn of recorded history. Archeological findings and discoveries have indicated that wheat domestication and use as a human food has a long history, for at least 6000 years (Pearson, 1967); as early as 7500 B.C. (Langer and Hill, 1982); that took place between 17,000 and 12,000 B.C. (Tanner and Raemaekers 2001); and for 8000 years (Curtis, 2002).

Wheat is today, one of the most important of all cultivated plants, more nutritious of cereals and continue to be most important food grain source to human nutrition (Pearson 1967; Harlan, 1981; and Curtis, 2002) and its contribution to the human diet puts it clearly in the first rank of plants that feed the world (Harlan, 1981). World wide, wheat is used as human food, seed, livestock feed, and as an industrial raw material (Tanner and Raemaekers, 2001).

Ethiopia is the largest wheat producer in sub-Saharan Africa. Wheat is an important food crop and it is one of the major cereal crops in Ethiopia. Wheat in Ethiopia is ranking fifth in area and production after teff, maize, barley and sorghum and fourth in productivity. Ethiopia endowed with a wealth of genetic diversity, particularly for tetraploid-wheats. Nevertheless, the productivity of wheat has remained very low mainly because, improved production technologies have not been adopted by the farming community (Adugna et.al, 1991). It is grown in the highlands at altitudes ranging from 1500 to 3000 masl. However, the most suitable agro ecological zones for wheat production fall between 1900 and 2700 masl. Major wheat production areas are located in the Arsi, Bale, Shewa, Illubabor Western Harerghe, Sidamo,Tigray, Northern Gonder, and Gojam regions (Bekele et. al , 2000).

Ethiopia began to use improved varieties of bread wheat on a commercial level in 1968. Most of the early improved bread wheat varieties released were developed in Kenya. The first varieties of Mexican origin were released in 1974. The first improved durum wheat was released in 1976, and the first bread wheat varieties developed in Ethiopia were released in 1980 (Adugna et.al, 1991).

Wheat technology demonstrations have been conducted by MOA (Ministry of Agriculture), AUA (Alemaya University of Agriculture, now Alemaya University) and IAR (Institution of Agricultural Research now renamed - EARI (Ethiopian Agricultural Research Institute), since 1958. Through these demonstrations, many wheat technologies have been transferred to farmers, particularly improved wheat varieties. (Getachew et al, 2002).

In the past, variety development and recommendation was made based on on-station trial with testing and selecting of promising genotypes under high external input and optimum crop management practices with low participation of farmers. In most of the cases, varieties developed under such conditions were poor and failed to prove their superiority under on-farm conditions and farmers management practices. This could be due to differences in management levels practiced by researchers and farmers and due to the lack of farmers participation and interaction in the variety evaluation and selection processes. To fill the gap of low adoption of technologies by farmers and increase farmers participation in technology evaluation and recommendation, a participatory research approach through client-oriented research should be employed widely (Getachew et al, 2002).

1.2. Statement of the Problem

Agriculture is the main economic sector in Ethiopia, providing employment for about 85% of the population, and accounting for around 50% GDP. Despite the importance of agriculture in its economy, Ethiopia has been a food deficit country for several decades (Tesfaye, 2004). Available evidence indicates that peasant agriculture in Ethiopia is characterized by

inadequate resource endowment and traditional methods of cultivation and husbandry practices. The majority of small holders in Ethiopia have limited access to land saving agricultural innovations such as high yielding varieties, inorganic fertilizers and chemicals (FAO, 1993).

Wheat is one of the most important cereal crops grown in Akaki, the study area and in the country. It contributes to the major share of daily consumption demand of rural households. In addition, it is used as cash source for a household. Wheat is one of the major products marketed. In the area, farmers grow both the improved and local varieties. Even though there is a tremendous and continuous effort made by agricultural development workers and researchers adoption and the yield increment of improved bread wheat varieties have not reached to the required level. There fore, assessing the level of adoption and the related problems by involving and participating farmers in the study can help to get reliable information that can be useful to facilitate and fasten t the production of improved bread wheat varieties.

Most of the times, in the country as well as in the study area, development, introduction and promotion of improved bread wheat varieties and other agricultural technologies are conducted without due consideration of farmers circumstances, constraints, local environment and their participation. As a result, less achievement in adoption of improved bread wheat varieties as well as in other improved agricultural technologies has been resulted. Therefore, the information revealed in this study on the probability and intensity of adoption and on farmers evaluation and selection criteria of improved bread wheat varieties by involving farmers is believed more reliable to use as an input in promotion of improved bread wheat varieties in the study area and in other areas having similar socio-economic and geographical conditions.

1.3. Objectives of the Study

In general, the objective of this study was to know the status of adoption of improved bread wheat varieties in the study area. However, the study was focused on the following specific objectives:

1.to identify farmers evaluation criteria of improved bread wheat varieties distributed in the study area;

2 .to assess the adoption and intensity of farmers improved bread wheat varieties use in the study area; and

3 .to identify determinants of adoption and intensity of improved bread wheat varieties use in the study area.

1.4. Significance of the Study

The findings and the results of this study could help to strengthen the promotion and production of improved bread wheat in the study area and in other areas having similar geographical and socio-economic characteristics with the study area. Therefore, based on the knowledge generated from the study, policy makers, government officials, NGOs, extension personnel, researchers and other development organizations can use as an input in policy, decision-making, in their development programs and efforts, in order to accelerate, the diffusion, dissemination and yield performance of improved bread wheat varieties as well as to make the quality improvement of the varieties characteristics through their professional efforts. The findings of this study might be also helpful and serve as a springboard for further investigation and research activities. More over, research organizations and extension providers can also use as an input in their activities.

1.5. The Scope and Limitations of the Study

The most important reasons to select and to conduct this research in Akaki area were the interest of the funding agencies EARO and Debrezeit research center. The Extension Division of Debrezeit research center also recommended to be conducted this research in this area and the recommendation got acceptance by EARO head office crop research department. More over in this area there is a wide wheat production practices going on; the area is also one of the sixth on farm research and demonstration sites of Debre Zeit research center; and there is also one research station to conduct on farm verification and adaptation trial in this area. From those six research sites of Debre Zeit research center, only Akaki is the area where wheat and other highland cereal crop research activities are conducted. Since wheat is the first most important crop in this area, the farming population uses this crop as major food and income source. These are some of the basic reasons why this research was conducted in Akaki.

The other limitations of this study were the budget or financial scarcities to cover the payments requested by enumerators for data collection and by respondent farmers for the information they intended to give and for the time they spent during the interview. But to maintain the quality of data, efforts were made and some payment arrangements for respondents per interview and for enumerators per interview schedule were made.

Due to the above-mentioned problems, the study was not conducted in other wheat growing area of the country and was also constrained to cover wider areas and larger sample size even in this area. As a result the study was limited to be conducted in Akaki area and cover only two PAs (Peasant Associations) or RKA (Rural Kebeles Administration) and only 150 respondents randomly selected from the two-selected sample PAs of this area, Akaki. The other limitation of this study might arise due to its closer location to the capital city of the country Addis Ababa since some farmers in this area may not spend their full time on their farm. As a result there is a doubt that they may not provide the real information.

On the other hand, the scope of this study was limited to cover and analyze only those factors influencing adoption and intensity of adoption behavior of farmers such as farmers age, education, health, gender (sex), leadership, extension service, distance of DA office and credit providers institutions from the farmers village, market and credit accessibility farmers farming and extension experience, family size, resource endowment like land, livestock, oxen and labor source. Other factors like farmers perception, knowledge, needs and attitude towards the various characteristics of improved bread wheat varieties were not covered in this study; hence, it is required to conduct further investigations.

1.6. Organization of the Thesis

This thesis is organized into seven major parts. Part one constituted the introduction, which focuses mainly on the background, statement of the problem, objectives, significance, the scope and limitation of the study as well as the organization of the thesis. Part two deals with review of different literatures on adoption of improved technologies and factors affecting adoption and intensity of adoption of improved bread wheat varieties. Part three describes the materials and methods including a brief description of the study area, data collection procedures and analytical techniques. Part four contains result and discussion. Part five constitutes summary and conclusion of the study. The remaining parts of this thesis are reference and appendices, which are covered under part six and seven respectively.

2. LITERATURE REVIEW

2.1. Concept and Theoretical Framework of Adoption

Adoption was defined as the degree of use of a new technology in long-run equilibrium when a farmer has all the information about the new technology and its potential. Adoption refers to the decision to use a new technology, method, practice, etc. by a firm, farmer or consumer. Adoption of the farm level (individual adoption) reflects the farmers decisions to incorporate a new technology into the production process. On the other hand, aggregate adoption is the process of spread or diffusion of a new technology within a region or population. Therefore, a distinction exists between adoption at the individual farm level and aggregate adoption, within a targeted region or within a given geographical area (Feder et. al., 1985).

If an innovation is modified periodically, the adoption level may not reach equilibrium. This situation requires the use of economic procedures that can capture both the rate and the process of adoption. The rate of adoption is defined as the proportion of farmers who have adopted new technology overtime. The incidence of adoption is the percentage of farmers using a technology at a specific point in time (e.g. the percentage of farmers using fertilizer). The intensity of adoption is defined as the aggregate level of adoption of a given technology, e.g., the number of hectares planted with improved seed. Aggregate adoption is measured by the aggregate level of use of a given technology with in a geographical area (Feder et. al., 1985).

Diffusions scholars have long recognized that an individuals decision about an innovation is not an instantaneous act. Rather, it is a process that occurs over a period of time and consists of a series of actions (Rogers and Shoemaker, 1971). Adoption is not a sudden event, but a process. Farmers do not accept innovations immediately; they need time to think things over before making a decision. There are several well-known schemes for explaining the adoption process. A popular one involves awareness, interest, evaluation, trial and adoption; and

another; knowledge, persuasion, decision and confirmation (Adams, 1982; and Rogers and shoemaker, 1971).

They elaborated these four stages of adoption process as follows (1) Knowledge: - when the individual learns of the existence of the innovation and again some understanding of its function. (2) Persuasion: - when the individual forms a favorable or unfavorable opinion of the innovation (3) Decision: - when the individual engages in activities that lead to a choice between adoption and rejection, (4) Confirmation: - when the individual makes a final decision to accept or abandon the innovation. According to their expression, it is well known that some people are more innovative (responsive to new ideas) than others. Therefore, adopters have been subdivided in to categories on the basis of the relative time they take to adopt innovations (innovators, early adopters, early majority, late majority and laggards). Innovativeness generally can be related to other personal characteristics; background, social status, affiliations, attitudes, etc. Research has shown that adoption of innovations often follows a bell shaped or normal curve when plotted against time.

Innovations are new methods, ideas, practices or techniques, which provide the means of achieving sustained increases in farm productivity and income. It is the extension workers job to encourage farmers to adopt innovations of proven value. It is an idea or object

perceived as a new by an individual. The innovation may not be new to people in general but, if an individual has not yet accepted it, to that person it is an innovation. Some Innovations originate from agricultural research stations, others from farmers. Innovations relate to objects social acts and abstract ideas. Generally, innovations may be classified in to technical and social innovations (Adams, 1982).

Innovations are also classified into process and product innovation (Adams, 1982). A process innovation is an idea input to a production process, while product innovation is a material input to the production process. The term innovation and technology are used interchangeably. Adoption and diffusion are distinct but inter-related concepts. Adoption refers to the decision to use a new technology, method, practice, etc. by a firm, farmer or consumer. The concept of

diffusion refers to the temporal (of time) and spatial (of area) spread of the new technology among different economic units (firms, farmers, and consumers). These two concepts defined by many researchers belongings to different academic disciplines (Legesse, 1998).

Among the many definitions that suggested by Rogers (1983) has been used in several adoption and diffusion studies. He defined aggregate adoption (i.e. diffusion) behavior as the process by which a technology is communicated through certain channels over time among the members of a social system. This definition encompasses at least four elements:(1) Technology, which represents the new idea, practices or objects being diffused (2) Channel of communication, which represents the way information about the technology flows from change agents (such as extension workers or technology suppliers to final users or adopters, (3) Time which represents the period over which a social system adopts a technology and (4) Social system, which is comprised of individuals, organizations or agencies and their adoption strategies (Kundson, 1991, in Legesse, 1998). Rogers defined adoption as use or non-use of new technology by a farmer at a given period of time. This definition can be extended to any economic units in the social system (Legesse, 1998).

With regard to the measurement of intensity of adoption, a distinction should be made between technologies that are divisible and technologies that are not divisible. The intensity or extent of adoption of divisible technologies can be measured at the individual level in a given period of time by the share of farm area under the new technology or by the per hectare quantity of input used in relation to the research recommendation (Legesse, 1998).

Feder et al., (1985) suggested that this measure might also be applied at the aggregate level for a region. In the case of non-divisible agricultural technologies such as tractors and combine harvesters, the extent of adoption at the farm level at a given period of time is dichotomous (adoption or non-adoption) and the aggregate measure becomes continuous. Thus, aggregate adoption of lumpy technology can be measured by calculating the percentage of farmers using the new technology within a given area.

There is also a great difference between the agricultural sectors of developing and developed countries. Agriculture in developing countries is heavily dependent on natural phenomena, while the effects of natural factors are, to some extent, mitigated by the application of modern technology and improved weather forecasting systems in developed countries. Moreover, farmers in developing and developed countries do not face the same types of constraints and opportunities. Therefore, conclusions concerning technology adoption cannot be drawn for agriculture in developing countries based on experience of the agriculture in developed countries (Legesse, 1998).

All individuals in a social system do not adopt a technology at the same time. Rather they adopt in ordered time sequence. Based on the time when farmers first begin using a new technology five possible adopter categories can be identified in any social system: innovators, early adopters, early majority, late majority, and laggards (Rogers, 1962 and Rogers and Shoemaker, 1971, in:Legesse, 1998).

In describing the characteristics of these groups, Rogers (1962, cited in: Largesse, 1998)) suggested that the majority of early adopters have expected to be more educated, venturesome, and willing to take risks. In contrary to this group the late adopters are expected to be less educated, conservative, and not willing to take risks. A practical aspect of the classification of adopters into adopter categories has been in the field of deliberate or planned introduction of innovation. Nevertheless, the usefulness of this categorization is restricted as there is evidence indicating possible movement from one category to another, depending on the technology introduced (Runguist, 1984, in: Legesse, 1998).

Attention has also been given to explain the mode (or approach) and sequence of agricultural technology adoption. Two approaches seem to appear in agricultural technology adoption literatures. The first approach emphasizes the adoption of the whole package and the second one stresses the sequential or stepwise adoption of components of a package. The technical scientists often advocated the former approach while the latter has advocated by the field practitioners, especially by farming system and participatory research groups. There is a great

10

tendency in agricultural extension programmers to promote technologies in a package form whereby farmers are expected to adopt the whole package. Experiences of integrated agricultural development projects such as CADU, in Ethiopia, however, show that this approach has not brought needed technical change because of resource limitation (Legesse, 1998).

The adoption of agricultural innovations in developing countries attracts considerable attention because it can provide the basis for increased production and income. That means farmers will adopt only technologies that suited their needs and circumstances (Nanyeenya et. al., 1997). In efforts to increase agricultural productivity, researchers and extension staffs in Ethiopia have typically promoted a technological package consisting of a number of components. However, because of capital scarcity and risk considerations, farmers are rarely adopting complete packages (Million and Belay, 2004).

Agricultural development implies the shift from traditional methods of production to new, science-based methods of production that include new technological components and/or even new farming systems. For farmers to adopt these new production technologies successfully, they must first learn about them and how to use them correctly in their farming system (Swanson and Claar, 1984).

The transfer of technology approach grounded on the diffusion model focuses on technology generation by scientists then handed over to extension to pass on to farmers. In this model, farmers considered as passive receivers and extension as technical delivery conveyer belts. The new roles of farmers, the new participatory approaches and methods and the new learning environments all imply new roles for agricultural scientists and extension (Kiflu and Berhanu, 2004). They have to learn from farmers and develop technologies that can serve the diverse and complex farmers situations. Farmers show great interest to evaluate the promising varieties that could be suitable to their local situation (Mergia, 2002).

11

Adoption of a new technology must be preceded by technology diffusion, e.g., the act of making new technology known to the potential adopters. Diffusion is therefore the link between research and development and adoption. Effective diffusion is an essential but not sufficient condition for adoption. The farmers of a given target category must not only be made aware of an available technology, they must also be convinced that adoption is in their best interests and above all, they must be able to adopt the proposed technology (Andersen, 2002 and Arnon, 1989). Adoption studies in developing countries started two to three decades ago following the green revolution in Asian countries. Since then, several studies have been undertaken to assess the rate, intensity and determinants of adoption. Most of these studies focused on the Asian countries where the green revolution took place and was successful. In Africa, new agricultural technologies have only been introduced recently (Roy, 1990 and Rukuni, 1994, cited in: Legesse, 1998).

The effectiveness of agricultural extension work highly depends on the availability of extension professionals who are qualified, motivated, committed and responsive to the everchanging social, economic and political environment. Adoption of technology by farmers can be influenced by educating farmers about improved varieties, cropping techniques, optimal inputs use, price and market conditions more efficient methods of production management storage, nutrition, etc. (Anderson and Feder, 2002).

2.2. Empirical Studies on Adoption

Feder, et al (1985) estimated the relationships among technologies already adopted by maize growing farmers in Swaziland by using factors analysis. They found farmers adopted the technologies investigated in three independent packages (1) improved maize verity, basal fertilizers and tractor ploughing (2) top dressing fertilizers, and chemical (3) planting date, and plant population (density). These empirical findings do not support a sequential or stepwise adoption process. They reported that farmers in Swaziland tend to adopt a package of technologies and the social system adopts a technology, which is comprised of individuals, organizations, or agencies with their adopting constraints.

12

Jha et al (1991) made a study in eastern province of Zambia to evaluate how small holders respond to interventions that promote the use of improved biochemical (seeds and fertilizers) and mechanical (animal traction) technologies. In eastern province, farmers were adopting both labor saving and yield increasing techniques. Agro ecological factors play a critical role even with in a relatively small region. Farm size affects the use of fertilizer in eastern province of Zambia. The age and gender of head of the household significantly influence the adoption of hybrid maize. Heads of household that are older and females are not likely to adopt hybrid maize. Extension makes only a small contribution to the process of adopting and diffusing technology. It contributes to specialized commodity-oriented programs but not to maize, the main crop.

A study conducted in Sierra Leone by Adesina and Zinnah (1992) on technology characteristics, farmers perceptions and adoption decision using tobit model analysis. The result of tobit analysis demonstrated that apart from age, farm size, extension service and experience were positively related to adoption decisions.

The study done by Legesse (1992) in Aris Negelle area on adoption of new wheat technologies indicated that experience, credit, expected yield, expected profit, cash availability for down payment, participation in farm organization as a leader, and close exposure to technology were the factors which significantly influenced the probability of adoption of improved varieties and intensity of adoption of fertilizer and herbicide. He found that the probability of adoption of improved varieties increases with an increase in farming experience. Farmers with higher experience appear to have often full information and better knowledge and were able to evaluate the advantage of the technology is considered. The study also revealed that credit is a crucial factor affecting the probability of adoption of improved varieties. And the quantity of fertilizer farmers applied was found to be sensitive to access of credit. The coefficient of the variable expected yield was significant and shows the intensity; of fertilizer application on wheat and maize is related to its expected profitability. In his study, farm size was not found to be important factor affecting probability of adoption of improved varieties and intensity of fertilizer application. However, the variable farm size per person significantly and negatively

13

influenced the intensity of herbicide adoption for weed control in wheat. The role of direct extension visits (as represented by frequency of visit by extension agent) was not found to be a significant factor affecting adoption. This can be attributed to the limited frequency of direct extension agent visits to non-contact farmers. On the other hand, the variable close exposure to technology was found to significantly affect the probability of adoption of improved varieties.

Chilot (1994) conducted a study using Probit and Tobit analytical models to identify factors influencing the three dependent variables such as rate of adoption of new wheat varieties, intensity of fertilizer adoption and intensity of adoption of 2.4.D weed control chemical under the title of adoption of new wheat technologies, by hypothesizing of eleven independent variables and this research result showed that several factors were affecting adoption of new wheat technologies in two extension centers, Wolemera and Addis Alem areas. As the study revealed, access to timely availability of fertilizer, perceived related profitability of the improved variety, number of extension contacts and wealth position were positively and significantly related to new improved wheat variety adoption. None of the household characteristics were significantly related to variety adoption. With regard to the intensity of fertilizer use, timely availability of fertilizer, number of livestock owned and perceived profitability were positively and significantly related to the intensity of fertilizer use. Literacy, wealth position of the farmers, exposure to improved technology and timely availability of 2, 4 D were significantly and positively related to the intensity of 2, 4, - D use. The result showed that only one variable, distance of extension agent office from farmers home was the common influencing independent factor affecting inversely and significantly all the three dependent variables namely the rate of adoption of new wheat varieties, intensity of fertilizer adoption and intensity of adoption of 2.4.D weed control chemical.

Another study made by Bisadua and Mwangi (1996) in southern high lands of Tanzania Mbeya region showed that farmers were at various components of the recommended package of improved maize production. Besides, farmers have adopted these components gradually. The four major factors that contributed this gradual adoption were cost of technologies, environmental factors, timely availability of inputs, and source of information of new

14

technologies. Technologies which require little cash out lay such as row planting are easily taken up by farmers because it was less costly and had an added advantage of simplifying weeding. Environmental stress affected the adoption of some of the recommendations especially where maize is planted during the day season, which utilizes residual moisture in the soil. Farmers who dry planted their maize did not apply basal fertilizer. This might be because of the fear of scorching their maize seed due to low soil moisture. Others did not perform the second weeding, apparently because rigorous weed germination will be suppressed by the moisture conditions. Lack of timely availability of inputs was widely cited as constraint to use them. In availability of improved maize seed was considered bottleneck to its use.

Other study in Sudan highlands also suggested unavailability of inputs as major constraint to their use (Lyimon and Temu, 1992, In: Bisauda and Mwagi, 1996). Giving the extension service is charged with the responsibility of extending information on new technologies; their low rates of contact with farmers may be acting as a constraint to the use of these technologies (Bisauda and Mwangi, 1996).

A study conducted on factors affecting adoption of maize production technologies in Bako area, Ethiopia, by Asfaw et al., (1997) using logit analytical model by hypothesizing seven independent variables to influence three dependent variables namely adoption of fertilizer, improved variety and row planting. The result of the model analysis showed that only one variable, namely extension activities was significantly influence adoption of improved varieties. Among the seven proposed independent variables, only two independent variables affected fertilizer adoption and two independent variables influenced the adoption of row planting significantly. In addition, only one independent variable namely extension activity, was influence all the three dependent variables in common among the seven proposed independent variables. Though, all of the hypothesized independent variables were expected to influence significantly the identified three dependent variables. Among hypothesized variables, one independent variable namely credit was excluded from further model analysis due to its less important to affect adoption of row planting as mentioned as a possible reason

15

in the report of the study but included in model analyses of fertilizer and improved variety adoption.

An assessment of the adoption of seed and fertilizer packages and the role of credit in small holder maize production in Kakamega and Vihiga districts, Kenya by Salsya et al (1998) showed that the age of household head, primary education, cash crop area, farm size, and credit were not significantly correlated with adoption. Secondary education, cattle ownership, use of hired labor, and access to extension significantly influenced the adoption of improved maize varieties. The use of hired labor and manure, cattle ownership and membership in an organization were significant factors affecting the adoption of fertilizer. Livestock ownership serves as a source of wealth to purchase inputs that affect significantly and positively in adoption of fertilizer. Farmers who use manure had lower probabilities of adopting fertilizer. Membership in an organization increased the likelihood of adopting fertilizer. Farmers who belong to an organization are likely to benefit from better access to input and to information on improved farm practices.

Farmers' participation in leadership of farmers' organizations seems to be the best prediction of adoption behavior of the farmer characteristic variables. The relationship between technologies may be independent, sequential or simultaneous and the patterns of adoption follow the domicile Period of time by the share of farm area under the new technology or by the per hectare quantity of input used in relation to the research recommendation (Rauniyer and Goode 1996, in:Legesse ,1998).

Another study conducted on adoption of soil conservation technologies in philppines uplands of two areas namely Cebu and Claveria by Lucila, et al. (1999) using probit model by hypothesizing nine independent variables to influence adoption of soil conservation practices in these two areas. The result showed that in Cebu and Claveria in each area only three independent variables were significant to influence the dependent variable adoption of soil conservation practices. But the common significant independent variable among those nine

16

hypothesized independent variables to affect the mentioned dependent variable in the two study areas was only one variable, which was the percent of land slope.

A study conducted on adoption of wheat technologies by Bekele et al (2000) in Adaba and Dodola woredas of Bale highlands of Ethiopia using tobit analysis model demonstrated that adopters of improved wheat technologies were younger, more educated, those who had larger families and farm, hired more labor and owned more livestock.

Another study conducted by Tesfaye and Alemu (2001) on adoption of maize technologies in northern Ethiopia shows that applying chemical fertilizer, access to credit, access to extension information, distance from development center, distance from market center and family size were factors affecting adoption of improved maize positively and significantly. The level of education was found to have no significant influence on the adoption decision of farmers for improved maize. Attendance of field day and access to extension information were negatively and significantly related to the adoption of decision of chemical fertilizer use. In this study, farm size, though positive, was not found to have a significant influence on the adoption of chemical fertilizer. Access to credit and use of improved maize are the most important factors found to positively and significantly influence the adoption decision of chemical fertilizer.

A study conducted by Tesfaye et al, (2001) on adoption of high yielding maize variety in maize growing regions of Ethiopia indicated that use of chemical fertilizer, access to credit, attendance of formal training on maize production and other agricultural techniques, access to extension information, distance to the nearest market center, family size and tropical live stock unit had significant and positive influence. On the side of adoption decision of chemical fertilizer, access to credit, level of education, farm experience, total farm size, use of improved maize, use of community labor were found to have a significant and positive influence.

Another study conducted on adoption of improved wheat technologies by small scale farmers in Mbeya district, southern highlands of Tanzania by Mussei et al, (2001) clearly indicated

17

that farm size, family size, and the use of hired labor significantly influenced the probability of land allocation to improved wheat varieties. Farm size, family size, the use of hired labor and credit significantly influenced the probability of fertilizer use. A unit increased in farm size among adopters decreased the probability of adopting fertilizer by 2.4% family size by 9.7%, use of hired labor by 4.7% credit by 5.9% has increased the probability of adopting fertilizer among adopters. Credit enables farmers to purchase inputs and increased the probability of adopting fertilizes among adopters by 5.1 %.

Another study conducted by Lelissa and Mullat (2002) on determinants of adoption and intensity of fertilizer use in Ejera district West Shoa Zone, Ethiopia, using probit and tobit analytical models and the result of probit model analysis indicated owning of draught power, credit access, owning of large farm size, access to extension service affect adoption of fertilizer positively. But age affects negatively and education has no significant influence on fertilizer adoption. Regarding the result of Tobit model analysis on the determinants that influence the adoption and intensity of fertilizer use; family size, education, draught power, access to credit and extension service have influenced positively.

A study conducted by Tesfaye (2004) on adoption of in organic fertilizer on maize in Amhara, Oromia, and southern regions, shows that on the adoption of chemical fertilizer, farm experience, access to credit, use of improved crop varieties, use of farm yard manure, family size, level of education, total farm size were considered significant. The larger the farm size the greater the probability of adopting of chemical fertilizer. In this study, family size was found to have a positive and significant impact on the adoption decision of chemical fertilizers. Access to credit can relax the financial constraints of farmers and allows farmers to buy inputs. The result of the study revealed that credit availability has significantly and positively impacted up on chemical fertilizer adoption. Educational level has increased the probably of adoption of chemical fertilizer. Use of improved variety of crops also influenced the decision of farmers to use chemical fertilizer positively and significantly.

18

A study conducted in Gumuno area of southern Ethiopian by Million and Belay (2004) to identify determinants of fertilizer use (adoption decision) shows that age of household head, access to credit, frequency of development agent visit, livestock holding and off- farm in-come influenced the adoption of fertilizer positively and significantly.

A study conducted by Adam and Bedru (2005) on adoption of improved haricot bean varieties in the central Rift Valley of Ethiopia, using logistic analytical model found that sex, total livestock unit, credit, and participation in extension service affect adoption of haricot bean varieties but dependent family members and land size affect negatively and significantly.

Another study was conducted in the central highlands of Ethiopia, on adoption of chickpea varieties by Legesse, et al, (2005) using logistic analytical model. The result of analysis demonstrated that the level of education of household head, farm size, access to extension service proportion of chick pea area and access to seed affect positively and significantly the adoption of chick pea varieties.

2.3. Farmers Participation in Agricultural Technologies Development and Evaluations

According to Hanson (1982), farmers, millers, bakers, and consumers differ in their concepts of desirable qualities in wheat. To farmers, a variety of wheat has quality if it resists diseases, matures at the proper time, doesnt topple over before harvest, and gives a good yield of plump grains without shattering (grain falling, to the ground before harvest). The miller is concerned with the grain. The Kernels should be uniform, the grain should be free of foreign matter, the moisture content should be low and the protein content high and the yield of flour per 100 kg of wheat should be high. The baker who produces leavened bread looks for flour that produces dough with desirable characteristics, the dough should be able to hold gas bubbles and yield a large loaf with good internal texture and color. The consumer does not see what grain before it is milled, but he or she has strong preferences regarding the appearance, texture, aroma and flavor of the breads, biscuit, cakes and other products that trace their

19

character partly to the wheat Kernels. These different viewpoints of farmers, millers, bakers, and consumers must all be considered to raise the wheat production.

It is useful to examine several features of small holders farming system in Ethiopia, and in the third world in general, and their implications for agricultural researchers. First, farmers are economically rational, that is, they adopt new practices that are in their interests and reject those that are not. When farmers resist a new technology, it is probably because it is not compatible with their objectives, resources or environment and not because of their backward ness, irrationality or management mistakes. Moreover, the small holders farming system is complex; small holders allocate their limited resources of land, labor and capital among many enterprises in a manner determined by their agro-ecological and socio- economic environment.

Farmers need to compromise enterprises to increase productivity. Farmers consider both technical and socio- economic aspects when deciding whether to use a new technology. Researchers are need to obtain an accurate and balanced assessment of the performance of the varieties, using both scientific and farmers own criteria. Farmers rarely adopt complete packages all at once, that is, complete set of recommended technological components concerning how to mange an enterprise. Instead, farmers usually use a step-by step approach, testing components individually and incorporate the successful ones into their system. Therefore, researchers need to evaluate new technologies individually or in simple combinations under farmers own management conditions. The greater the farmers participation in the designing and testing of a technology, the greater is the chance that they adopt it (Franzel, 1992).

A study conducted in West Shoa Zone, Ambo Woreda Birbisa and Cherech servicecooperative by Ethiopian Rural Self Help Association (ERSHA) in (2000) to evaluate bread wheat technologies on the farmers farm condition using farmers criteria. According to the study, farmers have formulated the criteria to evaluate the bread wheat varieties at different growing stages. The criteria formulated by farmers were crop stand (uniform germination, strong and healthy, deep green and many tillers), flowering (uniform flowering), heading

20

(panicle size, number of spike lets per head, resistance to lodging, frost and disease), yield (superior in yield, easy to thresh, stored for long time), Grain quality, (size, color, full body), baking (dough quality, baking quality and taste). Farmers evaluation criteria judging varieties during vegetative growth stage in order of importance were: tillering capacity (many tillers per plant), head size (panicle size, head length), frost and disease tolerant (healthy leaf and shoot, uniform germination and crop stand, resistance to lodging and shattering. Farmers evaluation criteria for judging grain quality characteristics were: yield per unit area, grain size (fill body, no shrink seeds and deformed seeds), baking quality (dough quality and being good bread), and color (important for market) and easy to thresh.

These days, it is well known that farmers participation in agricultural research and development processes are increasingly improved by realizing that the socio- economic and agricultural conditions of small-scale farmers are too complex, diverse and risk prone. Conventional approaches, which are well known by station-based researches followed by topdown technology transfer system, are not often adopted in a sustainable manner. Hence, building a partnership and management with farmers is needed throughout the cycles of diagnosis, experimentation and technology dissemination. This increases the understanding of the opportunities and constraints faced by farmers on top of their technical knowledge. This in turn enhances the prospects of technological development and its adoption rates (Mergia, 2002).

It was realized that farmers have their own priorities in their production strategy and often accepts those technologies, which they consider as most advantageous to their production system. Close engagement with farmers through the cycle of diagnosis, experimentation and dissemination increases understanding of conditions, of the opportunities and constraints farmers face, and of their own technical knowledge. The package-testing program also helped to get the assessments and evaluations of the technologies from the beneficiaries themselves. The approach has considerably contributed in increasing the understanding of the biological researches towards the farmers complex and linked circumstances and constraints. It has also

21

contributed in improving the linkage between research, extension and farmers as compared to previous approaches (Mergia, 2002).

A study conducted on the use of B.B.M (Broad Bed Maker) technology on vertisol, Sheriff (2002) shows that farmers in the study area got an opportunity to identify and select different crop varieties and grow the crop they preferred that can best meet their needs, interests and the corresponding agronomic practices of their specific agro-ecological conditions. Farmers do not operate according to the assumption of policy makers and scientists. It, moreover teaches us that agricultural knowledge varies and is accorded different social meanings depending on how it is applied in the running farms. This leads to differential patterns of farm management style, cropping patterns and levels of production. Farmers are heterogeneous and they are indeed knowledgeable and capable actors who consciously pursue various objectives.

Technological patterns of development should refer back to various resources and farmers capacity. The achievement of these objectives is influenced by the images they have of various aspects involving institutions. These call for the negotiation of values and resulting unintended consequences that could be referred to as counter development. We therefore, need to learn that technology transfer has to address different farmers needs, perceptions and strategies. We need to intervene with redesigning their use of technologies (Sherif, 2002).

Technology is not always a product of scientific institutions. Human beings are inherently capable of modifying their environment in the process of adaptation, where by technology is created and subsequently utilized. The struggle between the environment and people never stops, though under some circumstances, a long time may pass before intended changes are achieved. For various reasons, some societies adhere to certain technologies for centuries where as others pass comparable level of technology in a relatively shorter period of time. For instance, the revolution of farming tools for different operations in developed countries and the stagnation of the same in a developing country such as Ethiopians explain this observation (Tesfaye, 2003).

22

2.4. Conceptual Framework of the Study

In general, it could be inferred that agricultural technology adoption and diffusion patterns are often different from area to area. The differences in adoption patterns were attributed to variations in agro-climatic, information, infrastructures, as well as environmental, institutional and social factors between areas. Moreover farmers adoption behavior, especially and in lowincome countries, is influenced by a complex set of socio- economic, demographic, technical, institutional and biophysical factors (Feder et al, 1985).

Understanding and considering these factors when analyzing and interpreting farmers response to agricultural innovations has, there fore, become important both theoretically and empirically. Adoption rates were also noted to vary between different groups of farmers due to differences in access to resources (land, labor, and capital) credit, and information and differences in farmers perceptions of risks and profits associated with new technology. The direction and degree of impact of adoption determinants are not uniform; the impact varies depending on type of technology and the conditions of areas where the technology is to be introduced (Legesse, 1998).

Farmers decision to adopt or reject new technologies can also be influenced by factors related to their objectives and constraints. These factors include farmers resource endowments as measured by (1) size of family labors, farm size and oxen ownership, (2) farmers socio economic circumstance (age, and formal education, etc) and (3) institutional support system (available of inputs) (CIMMIYT, 1993).

In many developing countries it has become apparent that the generating new technology alone has not provided solution to help poor farmers to increase agricultural productivity and achieve higher standards of living. In spite of the efforts of National and International development organizations, the problem of technology adoption and hence low agricultural productivity is still a major concern (CIMMIYT, 1993).

23

The inability of farmers to achieve high yield levels has been blamed on many different sources on extension services side, for not properly disseminating the research stations technologies, on input supply systems side, for failing to make the new technologies available, on policy decision makers side, for making the new technologies unprofitable to use due to policy distortion and on farmers themselves, who are alleged (assumed) to be too conservative. However, many studies point to another cause of low adoption rate the research center recommendations that are irrelevant to the small farmers priorities, resource constraints, and the physical, cultural and economic environment (Winkelmann, 1977, in: Mulugetta et al, 1994).

On agricultural technology adoption and diffusion determinant factors in different countries across the world, Africa including Ethiopia, several and various studies have been conducted and many researchers have obtained various findings. The researchers lack of understanding of the farmers problem and the conditions under which they operate may result in the development of inappropriate technologies and low rates of technology adoption (Fresco, 1984, in: Mulugeta et al, 1994).

In this study efforts were made to revealed factors affecting adoption and intensity of adoption, the pattern and direction of adoption of improved bread wheat varieties (part of agricultural technologies) that varied according to farmers resource endowment, environmental situations, technological development, personal characteristics, accessibilities to different services such as credit, extension, information market and the importance, suitability, management and cost of the technologies.

Moreover literatures, practical experiences and field observations have confirmed that technologies adoption by farmers can be fasten, enhanced and make sustainable by understanding those factors influencing the pattern, degree and direction of adoption and by designing and establishing technologies diffusion and adoption pattern strategies through

24

farmers empowering, making farmers access to infrastructure, information, technologies, credit, field support how to utilize new technologies.

Other factors should be also included in agricultural technologies disseminations and adoption. Farmers participation in technologies development, selection and dissemination strategies as well as result evaluation should be considered, because farmers have a long year of farming and environmental experience. The need and interest of farmers towards agricultural innovations also varies depending on farmers farming environment, their belief, experience, economic status and their personal background and characteristic. Therefore, disseminating improved agricultural technologies without consultation of farmers most probably ends with failure.

Several literatures, practical experiences and observations of the reality have been showed that one factor may enhance adoption of one technology in one area at one time and may hinder it in another situation, area and time. Therefore it is difficult to develop a one and unified adoption model in technology adoption process because of the socio economic and ecological variations of the different sites, and the various natures of the determinant factors. Hence, the analytical framework presented in the below figure shows the most importance variables expected to influence the adoption and intensity of adoption of improved bread wheat varieties in the study area, Akaki.

25

Asset endowment and other income source - Livestock -Farm land - Off-farm

Institutional variables - Extension service - Credit access -Market access - Distance of extension office -Distance of credit provider Institutions

Decision to adopt improved bread wheat and to increase size of farm land for improved bread wheat production

Household socioeconomic characteristics - Sex - Age -Health - Education -Experience in extension

Labor Sources -Oxen -Labor source -Family size

Note = the above Figure shows the chart of conceptual frame work of the study

26

3. RESEARCH METHODOLOGY

It is well known, that, there are two research methodologies classified under the broad headings: the qualitative and quantitative research methodologies.. Methods are the tools of data generation and analysis techniques practically; methods are the tools of the trade (job) for social scientists and are chosen on the basis of criteria related to or even dictated by the major elements of the methodology in which they are embedded, such as perception of reality, definition of science, perception of human beings, purpose of research, type of research, type of research units and so on.

As many people described the basic objective of a sample is to draw inferences about the population from which such sample is drawn. This means that sampling is a technique, which helps us in understanding the parameters or characteristics of the universe or population by examining only a small part of it. Therefore, it is necessary that the sampling technique be reliable. (In general, a study on relatively small number of units, are the sample, should be representative of the whole target population. Sampling is, thus, the process of choosing the units that could be included in the study, determine the sample size and the sample selection procedures. A sample design is a definite plan, completely determined before any data are collected for obtaining a sample from a given population. In this study under this chapter the study area description and sample farmers demographic, resource ownership and institutional services has conducted.

3.1. Description of the Study Area (Akaki)

3.1.1. Location, relief and climate

The socio-economic and environmental factors of the area play a great role for better performance of any activity done in that particular area. There fore it is highly valuable to

27

describe the area where the activity is planed to be under taken. . In addition, its accessibility and the budget constraint of the research were some of the factors to fix and conduct this research in this area.

This research activity was decided to undertaken at Akaki area, which was selected by its wide growing and demonstration of improved wheat crop varieties and wide utilization of other improved agricultural technologies in this area The reasons to conduct this research, in Akaki area due to wide wheat production practices and high-improved agricultural inputs utilization as well as wide demonstration practices on agricultural inputs applications and utilizations in this area.

Akaki, located at South East of Addis Ababa and it is the rural part of Addis Ababa and Akakikaliti sub-city, one of the sub-cities of Addis Ababa. It is bounded by Oromia region to the east and southern part. The study area, Akaki, constituted 9 Peasant Associations (PAs) or Rural Kebele Administrations (RKAs) in the mean time when this research survey was carried out. But at the end of this research survey, these Rural kebele Administrations /Peasant Associations were reorganized into four reduced number of Rural Kebele Administrations as a result of a new restructuring program of Addis Ababa administration.

The agro-ecological zone of the study area, Akaki, is 100% high land and its altitude ranges from 2100 2300 masl. In the study area there are different soil types. The most important and dominant soil type in area coverage is heavy ver ti sol or black clay soil. Except some few hilly landscapes of the study area, Akaki, virtually is plain.

Therefore, soil erosion by water is not a problem in the study area. But water logging is a very serious problem resulted from its flat landscape.

28

Table 1.The Livestock and crop types in the study area No. Types of crop grown in the study area Land Coverage in (Ha) 3580 1930 1560 90 604 80 80 201 44 199 30 104 42 4360 0.70 2.38 0.96 100 Poultry Bee-in-Hive 17807 No data Land coverage in percent 82.1113.85 Types of Livestock reared in the study area Cattle Oxen (Cow and Others) Sheep Goat Pack Animals (Horse, Mule And Donkey) 13211 10380 3064 Number of Livestock 17269 4058

Cereal crops Wheat Teff

Others Pulse crops Faba bean peas Chick pea lentils

5012

3 4 5

others Oil crops Vegetables Others (Fenu greek) Total

(Source: Akaki Agricultural Unit Office, 2005)

The study area, Akaki, is characterized by the high land climate. It has the main and small rainy seasons. Farmers in the study area rely on the main rainy season known as kiremt or Meher rainy season for their agricultural production activities. There is no a practice and experience of crop production using the rain of small rainy season known as Belg rainy season and irrigation production. The small rainy season extends some times starting from January or February and ends some times in May or June. But the main rainy season is similar like otherparts of Ethiopia that extends from end of or mid of June to most of the time mid of September, according to Akaki agricultural unit office. Some of the major crops grown in the study area are Wheat, Teff, Faba bean, Chickpea and Lentils.

29

3.1.2. Agriculture and demographic characteristics of the study area

The total farming population of the study area is 14519. Of which, 7626 were male, and 6893 female. The number of household heads of the farming population is 2490 male and 265 female with a total of 2755 household heads. The average family size of the study area is 5.27 per household. In Akaki, agriculture, which includes crop and livestock production, is the main stay of the farming community like other parts of Ethiopia. The types of crops, the farm land coverage by each crop type and the types of livestock and the size of livestock population reared by farming community in the study area are summarized in Table 1.

Table 2.The land use of farmers in the study area No 1 2 3 4 5 Types of land use Cultivated land Grazing land Forest land Village and construction Others Total Coverage in (Ha) 4360 151 80 2496 22 7109 Percentage 6 1.33 2.12 1.13 35.11 0.31 100

(Source: Akaki Agricultural Unit Office)

As indicated in table 2, the larger proportion of land in the study area is used for cultivation, which is 4360 (61.33 %.). Of which the major proportion goes to cereal crop production particularly for wheat production followed by Teff production. In the study area, there is a very serious grazing land scarcity, greatly affecting the livestock production, resulted from high population pressure and extended farming practice that shrinks grazing land and compete with livestock production. The farming society in the study area used crop by product for animal feed though it is poor nutritionally.

30

Table 3.The summary of oxen ownership No. 1 2 3 4 5 6 Oxen Ownership With no oxen With one oxen With two oxen With three oxen With four oxen With five % above Total (Source: kaki Agricultural Unit Office) Number of owners 1272 243 555 139 442 104 2755 Percentage 46.17 8.82 20 .15 5.05 16.04 3.77 100

In the study area as summarized in table 3, around 46.17% farmers are with out ploughing oxen, according to the Akaki Agricultural Unit office. In the study area Akaki, there is a fuel wood scarcity resulted from unwise practice of deforestation for long time. There fore the farming population has forced to use cow dung as their source of energy for heat and food preparation. Using of cow dung for fuel can affect the utilization of compost to improve soil fertility for better crop yield.

3.1.3. Institutional services of the study area

Effective Agricultural Extension services have paramount importance to farmers to get timely advices and information on the availability, use and application of new, improved and modern agricultural inputs, technologies and practices. The Akaki Agricultural Unit office is responsible to offer agricultural extension service in the study area. Under this unit office different expert with different professions has assigned at the Unit Office level and Development Agent (DA) center level. The 9 Extension Agents / Development Agents at the center level were assigned and responsible to give extension service to the farming community. They were accountable to Akaki Agricultural Unit. But at the end of this research survey and data collection process due to the new restructuring program of Addis Ababa Administration some DAs from their center and other professionals from Akaki Agricultural

31

unit office has transferred to other unrelated duties with their professions that can affect negatively the extension services and rural development efforts of the study area.

Table 4.Improved agricultural input distribution of the study area in different years Crop Season 1997 1998 1999 2000 2001 2002 2003 2004 Total DAP Quit. 5224.5 4695.5 3759 4172 3866 3868 3994 167.5 29746.50 Urea Quit. 3000.5 2506.5 2026.5 2235.5 2034.5 2182 2340 167.5 16493 Teff Quit. 4.5 5.5 11.85 12 13.05 19.5 66.40 Pesticide K.g. 10 5 5 5 4 4 4 37 Pesticide Lit 30 30 20 10 10 4 7 111
Weedi-cide

Lit 50 76 48 33 35 5 50 297

(Source: Akaki Agricultural Unit Office)

Availability of improved agricultural inputs to use and credit service to purchase agricultural inputs is very vital for technology adoption. In the study area different agricultural inputs and credit were distributed in different time for farmers. Table 4 showed the agricultural input distribution of the study area in different years.

Table 5 showed the improved bread wheat varieties seed distribution and table 6 showed the improved credit distributions in different years to the farming community of the study area. According to the Akaki Agricultural unit office information the major inputs distributed in the study area were fertilizer (DAP and Urea) and improved bread wheat varieties as well as improved teff varieties. The major input distribution of the study area from 1995 to 2004 were, fertilizer in quintal, (DAP=29746.50 and 16493), improved bread wheat varieties 732.75 and improved Teff 66.40 quintals.

32

Table 5.Improved bread wheat seed distribution of the study area in different years No 1 2 3 4 5 6 7 8 Crop Season (Years G.C.) 1997 1998 1999 2000 2001 2002 2003 2004 Total Improved Land covered Bread Wheat (Quit.) (Ha.) 43.5 29 74.25 49.5 48 32 72 82.5 120 145.5 147 732.75 50 55 80 97 98 490.50 No.of Participant Farmers 58 99 64 100 110 160 194 196 981

(Source: Office of Akaki Agricultural Unit office)

As it is presented in table 4 and table 5, the average annual DAP fertilizer distribution were 3718.3 quintals, urea 2061.63, improved bread wheat seed 8.3 quintals and that of improved Teff was 4.625 quintals. The fertilizer distribution was almost satisfactory. But the improved bread wheat and teff seed distribution was very low quantity.

Regarding the credit distribution of the study area as presented in Table 6, the larger credit for the last seven years a total of 376478.06 Ethiopian birr for fertilizer and improved grain crop seed purchase was distributed .The larger proportion of the credit were used for fertilizer purchase. This is because farmers in the study area got credit in the form of fertilizer.

33

Table 6.Credit Distribution of the study area in various years Credit for Fertilizer and Improved seed (ET.Birr) 27,858.50 54,076.60 16,275.75 60,552.70 25,446.61 94,218.75 98,049.15 376478.06 Livestock Production (ET.Birr) 47,800 118,885 166685

No 1 2 3 4 5 6 7

Crop season (Year G.C.) 1997 1998 1999 2000 2001 2002 2003 Total

(Sources: Office of Akaki Agricultural unit office).

In the study area credit were also distributed for livestock production as showed in Table 6.For this sector of the economy credit was distributed before five years ago. The credit distribution was covered only for two consecutive years. The credit service limitations in amount, type and facilities can affect negative the adoption of improved agricultural technologies, agricultural development and the over all rural community lively hood living situation improvement.

3.2. Description of Data Collection and Data Analysis Methods and Procedures

3.2.1. Sources and types of data

It is very helpful for researchers to anticipate and think over in advance about the sources and types of data that are relevant to the research and, therefore, need to be gathered. This help to avoid confusion and unnecessary time, labor, finance and other resources wastages. The types of data, primary and secondary, were collected to answer and fulfill this research questions and objectives. All information about determinants of adoption and intensity of adoption and

34

farmers evaluation and selection criteria of improved bread wheat varieties, demographic, socio-economic, environmental situations, wheat production, credit facilities, extension service and others relevant data to the study were gathered from primary sources quantitatively through interview schedule and qualitatively through group discussion and observation.

Data also were gathered by examining secondary sources such as documents, reports and records of DAs (Development Agents), and other related agricultural offices and research centers. All these data in the process of the study were gathered using different methods and techniques based on the nature, types and characteristics of the data. Both quantitative and qualitative data were gathered through different data collection methods from primary and secondary sources.

3.2.2. Sample size and sampling techniques

This study was determined to conduct in Akaki area, which is the rural part of Addis Ababa Administration. In this study sample size was determined by taking different factors such as research cost, time, human resource, accessibility, availability of transport facility, and other physical resource accessibilities. By taking these factors into account, it was fixed to cover two Peasant Associations out of 9 PAs and 150 household head respondents from the total 2755 household head population of the study area, Akaki. Through out sample selection processes simple random selection method has employed. The two stage sampling techniques were applied in sample selection processes. First, the two Peasant Associations (PAs) or Rural Kebele Administrations (RKAs) namely Koye and Gelan-Edero involved in improved bread wheat production were selected out of nine PAs using simple random selection method.

Second 150 sample household head farmers were selected from total wheat growers of the two samples Pas. About 65 (43%) Sample farmers from Gelan-Edero PA and from Koye sample PA 85 (57%) sample farmers were selected proportionally. From Gelan-Edero PA 36 were adopters and 29 were non-adopters and from Koye PA 63 were adopters and 22 were non-

35

adopters. Out of 150 respondents (132) 88% were male and the remaining (12%) was female. From total respondents (99) 66% were adopters and the rest 51 (34%) were non-adopters. From total adopters 93% were male adopters and the remaining 7% were female adopters. Concerning non-adopters 78% were male and 22%were female as presented in Table7. All sample selection processes were carried out in pursuing of statistical procedures and with consultation of DAs, Akaki Agricultural Unit Office professionals and PA leaders of the study area.

3.2. 3. Data collection methods

Data for this study were gathered from sample household head farmer respondents through interview, group discussion and observations. Both qualitative and quantitative data were gathered using the mentioned methods. Secondary data that were relevant to this study were also gathered through examining of published and unpublished data that was gathered and organized by other bodies for other purposes .In this process care was taking in taking and selection of the relevant data suitable and relevant for this study.

3.2.3.1. Quantitative data collection methods

In this regard, primary data were collected through personal and face-to-face interview using structured and pre-tested interview schedule that were filled up by recruited and trained enumerators under the close supervision of the researcher. Totally, 150 randomly selected samples household head farmer respondents were covered under the survey. Also, secondary data were gathered by examining secondary sources such as records, reports, and research results and other documents and publications from office of agriculture, research centers and other respective offices.

36

3.2.3.2. Qualitative data collection method

In the study area primary qualitative data on improved bread wheat varieties selection and evaluation criteria of farmers were gathered through group discussion and individual discussions conducted with farmers and professionals. Researchers personal observation and transect walk were also used in this data gathering processing. Data gathering through these methods were continued to the point of saturation, to crosscheck, triangulate, elaborate and enrich the information on both qualitative and quantitative data to increase the reliability and trustworthiness of the information. The group members and individuals were familiarized to the discussion points and encouraged to forward their opinion they felt with out any reservation. In this process, recording, coding, reorganizing and arrangements, refining expanding of information was conducted.

3.3. Analytical Models

3.3.1. Logit model

Several models are available to analyze factors affecting technology adoption. The choice of one may depend up on several factors. Some of these alternative models are the discrete regression models in which the dependent variable assumes discrete values. The simplest of these models is that in, which the dependent valuable Y is binary (it can assume only two values denoted by 0 and 1).

The three most commonly used approaches to estimate such models are the linear probability model (LPM), the logit model and the probit model .The linear probability model has an obvious defect in that the estimated probability values can lies outside the normal range 0-1 range. The fundamental problem with the LPM is that it is not logically a very attractive model because it assumes that the marginal or incremental effects of explanatory variables remain constant, that is Pi=E (Y=1/x) increases linearly with x (Maddala, 1997 and Gujaratti, 1998).

37

The authors suggested that the sigmoid or S-shaped curve were very much resembles the cumulative distribution function (CDF) of random variable is used to model regressions where the response variable is dichotomous, taking 0-1values.

The cumulative distribution functions (CDFs), which are commonly chosen to represent the 01response models, are the logit (logistic CDF) model and the probit (normal CDF) model. Logit and probit models are the convenient functional forms for models with binary endogenous variables (Tohnston and Dianardo, 1997 cited in Techane, 2002).

These two models are commonly used in studies involving qualitative choices. To explain the behavior of dichotomous dependent variables we will have to use a suitably chosen cumulative Distribution Function (CDF). The logit model uses the cumulative logistic function. But this is not the only CDF that one can use .In some applications the normal CDF has been found useful. The estimating model that emerges from normal CDF is popularly known as the probit model (Gujarati, 1999).

The logistic and probit formulations are quite comparable the chief difference being that the logistic has slightly flatter tails that is the normal curve approaches the axes more quickly than the logistic curve. There fore, the choice between the two is one of mathematical convenience and ready availability of computer programs (Gujrati, 1988 cited in Techane, 2002). A relevant model offers better explanation on the underlying relation ship between adoption decision and factors influencing it. The most widely used qualitative response models are the logit and the probit models (Amemiya, 1985).

Both the probit and logit models yield similar parameter estimates and it is difficult to distinguish them statistically (Aldrich and Nelson, 1984). How ever, because of the fact that binomial logit model is easier to estimate and simpler to interpret. Therefore a logit model is used in this study to determine the relation ship between adoption decision and factors affecting the adoption of improved bread wheat varieties in the study area, Akaki.

38

The specification of the logit model is as follow: = ( = 1) = exp (Zi) 1+exp(Zi)

Where denotes the probability that the i

th

farmer will fall in the adopters class (yi=1) and

exp (Zi) stands for the irrational number e to the power of Zi. The un observable stimulus index Zi, assumes any value.

However, the Logistics transformation guarantees each corresponding value of Pi to fall inside the 0-intervals. The stimulus index Zi, also called the Log of the odds ratio, in favor of improved bread wheat varieties adoption, is actually a linear function of factors influencing adoption decision as specified hereunder: =0+11+22i + .+ i+e Z= ln [
pi 1-p

] i

Where: X1, X2, X3,...........Xp=explanatory variables 0, 1, 2, 3, 4, ,p = Logit Parameters to be estimated, ei = the error term
= 0+11i+22i+33i +.ppi +

In reality, the significant explanatory variables do not all have the same level of impact on the adoption decision of the farmers. Therefore, the impact of each significant variable on the probability of adoption was calculated by keeping the continuous variables at their mean values and the dummy variables at their most frequent values (zero or one).

The estimated coefficients of the Logit model of improved bread wheat adoption are listed in the table 20. The likelihood ratio statistics is significant at 10 percent probability level and implies that the independent factors taken together influenced improved bread wheat varieties adoption. The model correctly predicted 81.33 percent of the adopters and non-adopters.

39

3.3.2. Tobit model

Adoption studies based up on dichotomous regression model have attempted to explain only the probability of adoption versus non-adoption rather than the extent and intensity of adoption Knowledge that a farmer is using high yielding variety may not provide much information about farmer behavior because he /she may be using some percent or 100 percent of his /her farm for the new technology. Similarly, with respect to adoption of fertilizer is, a farmer may be using a small amount or a large amount per hectare area. A strictly dichotomous variable often is not sufficient for examining the extent and intensity of adoption for some problems such as fertilizer (Feder et al., 1985).

There is also a broad class of models that have both discrete and continuous parts. One important model in this category is the Tobit. Tobit is an extension of the probit model and it is really one approach to deal with the problem of censored data (Johnston and Dinardo, 1997 cited in Techane, 2002).

When examining the empirical studies in the literatures, many researchers have employed the Tobit model to identify factors influencing the adoption and intensity of technology use. For example Nykonya et al (1997) ; Lelissa (1998) ; Bezabih (2000); Croppenstedt et al. (1999) as cited in techane (2002) used the Tobit model to estimate the probability and the intensity of fertilizer use ( Adesina and zinnah ,1993; in Techane ,2002).

The econometric model applied for analyzing factors influencing the intensity of adoption of improved bread wheat varieties is the Tobit model shown in equation (1). This model was chosen because, it has an advantage over other adoption model (LPM, Logistic and Probit Models) in that, and it can reveal both the probability of adoption of improved bread wheat varieties and the intensity of use of the varieties. The Tobit model can be defined as:
Yi = X i + U i , Yi = Yi if Y* > 0 = 0 if Yi * 0 (1)
* *

i = 1,2,...n

40

Where, Yi= the observed dependent variable, in our case the land size in hectare covered with improved bread wheat variety. Yi*= the latent variable which is not observable
Xi=

vector of factors affecting adoption and intensity of fertilizer use

i= vector of unknown Parameters Ui= residuals that are independently and normally distributed with mean zero and a common variance 2. Note that the threshold value in the above model is zero. This is not a very restrictive assumption, because the threshold value can be set to zero or assumed to be only known or unknown value (Amemiya, 1985). The Tobit model shown above is also called a censored regression model because it is possible to view the problem as one where observations of Y* at or below zero are censored

The model parameters are estimated by maximizing the Tobit likelihood function of the following formula (Maddala, 1997 and Amemiya, 1985).

1
L = C

Yi >0

f(

Yi

i X i

) Y * 0 F ( i

i X i

(2)

Where, f and F are respectively the density function and cumulative distribution function of Yi*. IIYi*< 0 means the product over those i for which Yi* < 0, and IIYi*<0, and IIYi>0 means the product over those i for which Yi*>0.

It may not be sensible to interpret the coefficients of Tobit in the same way as one interprets coefficients in an uncensored linear model. Hence one has to compute the derivatives to predict the effects of changes in the exogenous variables.

1. The marginal effect of an explanatory variable on the expected value of the dependent variable is:
(Yi ) = F (Z ) i X i (3)

41

Where

i X i is denoted by z,

2. The change in the probability of adopting a technology as independent variable Xi change is:
F ( Z ) = f(Z) i X i
(4)

3. The change in intensity of adoption with respect to a change in an explanatory variable among adopters is:
(Yi / Yi > 0) f (Z ) f(Z) 2 = i [1 Z ( ) ] X i F (Z ) F(Z)
*

(5)

Where, F (z) is the cumulative normal distribution of z, f (z) is the value of the derivative of the normal curve at a given point (i.e., unit normal density). Z is the z score for the area under normal curve, B is a vector of Tobit maximum likelihood estimates and 0* the standard error of the error term.

Parameter estimates of the Tobit model for the intensity of adoption of improved bread wheat varieties (measured in terms of size of land in hectare used for growing of improved bread wheat varieties over the total wheat land in hectare) as shown in Table 22. And the results are discussed under section (5.1.2.). The Tobit model was used or applied to analyze the factors that determine the intensity of adoption of improved bread wheat varieties because the mean proportion of land allocated to improved bread wheat varieties is a continuous variable but truncated between zero and one. This model is relevant to predict the intensity of adoption of improved bread wheat varieties by farmers when the dependent variable is continuous.

3.3.3. Other Quantitative data analysis methods

In this study, in addition to econometrics models, Logit and Tobit models described in the above, descriptive statistics such as percentage, tabulation, mean, standard deviation, t-test and 2 - test data analysis methods were employed in quantitative data analysis of the study.

42

3.3.4. Qualitative data analysis method

The qualitative data analysis has conducted to strengthen the evidences obtained through quantitative data collection methods or survey method. In this study, the qualitative data obtained from the group, individual formal and informal discussions and through the researchers personal observation regarding farmers selection and evaluation criteria, their priorities of improved bread wheat varieties disseminated in the study area were summarized using the criteria established by the group members by analyzing the characteristics of these varieties to what extent these varieties satisfy and fit their needs, interests and to their environmental situations. The qualitative data were analyzed through explanation of idea, opinion, and concept explanation method. Researchers personal observations and transect walk watching were analyzed through, further explanation of the real world under observation.

3.4. Hypotheses Testing and Definitions of Variables

In this study the variables were selected and hypothesized using literatures, by considering farmers production practices, area situations and objectives of the study. In this study it was decide to concentrate the research effort and limited resources on socio economic and environmental conditions and constraints that was expected to influence probability and intensity of adoption because as the Ethiopian extension history shows that in this area extension service was provided for long years using different methods such as demonstration farmers field day, on farm field visit and support. More over, farmers in this area have better access to different information sources .As a result, were expected to have better understanding, knowledge, and attitude towards improved agricultural technologies. As a result, more emphasis was given to exogenous socio economic variables than internal variables to hypothesize, test and analyze using the Logit and Tobit Analytical models.

43

3.4.1. The Dependent variables of logit and tobit models

3.4.1.1. The Dependent variable of logit model

The dependent variable of the binomial logit models log-odds ratio is the probability of adopting or not adopting the improved bread wheat varieties which can used to identify factors determining adoption of improved bread wheat varieties is the natural logarithm of the ratio of the probability that a farmer adopts the improved varieties (Pi) to the probability that he /she will not (1-Pi). The log-odds ratio is a linear function of the explanatory variables.

3.4.1.2. The Dependent variable of tobit model

The dependent variable of Tobit model has continuous value, which should be the intensity, the use and application of the technology. As observed in different empirical studies this variable can be expressed in terms of ratio, actual figure and log form depending on the purpose of the study. For example in their study of factors influencing adoption of fertilizer, Nkonya et.al, (1997) as cited in Techane (2002) considered fertilizer applied per hectare as the dependent variable of the tobit model. Likewise Shiyani et al., (2000) as cited in Techane considered the proportion of area under chickpea varieties in their study of adoption of improved chickpea varieties. Consequently, in this study the ratio of actual land size under improved bread wheat varieties to total wheat land size was taken as a dependent variable of the tobit model.

3.4.2. The Independent variables and their definitions used in logit and tobit models

Adoption literatures provide a long list of factors that may influence the adoption of agricultural technologies. Generally these factors can be grouped into demographic personal, socio economic, physical and institutional factors (Million and Belay, 2004). There fore,

44

farmers decision to use improved bread wheat varieties and the intensity of the use in a given period of time is hypothesized to be influenced by a combined effect of various factors such as household characteristics, socio-economic and physical environments in which farmers operate. Based on Feder et.al, (1985) that extensively reviewed factors affecting adoption of agricultural technologies in low income countries, and on the brief literature review in this study the variables mentioned below are hypothesized to explain improved bread wheat varieties adoption and the intensity of the use of these varieties by the sample households.

In the course of identifying factors influencing farmer's decision to use improved bread wheat varieties, the main task is to analyze which factor influence how and by how much. It is hypothesized that adoption and intensity of adoption are influenced by the combined effect of various determinants. There fore, in the following section potential variables that are supposed to influence adoption and intensity of adoption of improved bread wheat varieties in the study area will be explained. More specifically, the following potential explanatory variables hypothesized to influence the adoption and intensity of adoption of improved bread wheat varieties in the study area, on a priori grounds is indicated below:

1. Farmers age (HHHAGE) As the farmers age increases it was expected that farmer become conservative. Then it is hypothesized that the farmers age, adoption, and intensity of adoption of improved bread wheat varieties are inversely correlated. Therefore, in this study it was assumed that the lesser age group could adopt improved bread wheat varieties more than the older age group farmers. Then, in this study farmers age and adoption are expected to relate negatively. As farmers age increase probability of adoption/intensity of adoption is expected to decrease.

2. Gender/Sex (HHHSEX) - It is hypothesized that male household headed farmers are expected to adopt improved bread wheat varieties more than female headed ones. Because it is expected that male-headed farmers have a better opportunity to access to credit and extension service. In this study gender/sex was coded if the household is male 1 and 0, otherwise.

45

Adoption/Intensity of adoption was expected to increase and correlate positively as the farmer being male.

3. House hold head Education (EDUHHH) This represents the level of reading and writing and formal schooling attended by the household headed. It is expected that educated farmhouse hold head can make better decision to adopt improved bread wheat varieties than non-educated ones. Here, education extends from read and write to attending regular school education. In this study this variable was treated as a dummy variable and has coded if the house hold head can read and write as well as attending the regular school education as 1 and 0, otherwise. Adoption was expected to correlate positively as education increases.

4. Family size (FAMILYSI) - household heads with large family size are less likely to adopt improved bread wheat varieties due to risk aversion. In this study it had expected that the family size, adoption, and intensity of adoption would have related inversely. As family size increase adoption/intensity of adoption has expected to decrease.

5. Extension-services (GEXSERVE) - the more frequent DA visit, using different extension teaching methods and training, attending demonstrations and field day can help the farmers to adopt a new technology and can also increase the intensity of adoption. If the farmers get better extension services are expected to adopt better-improved bread wheat varieties than others. In this study this variable had treated as a dummy variable in that if the farmer get extension service is coded as 1 and 0, otherwise. As extension service increase adoption/Intensity of adoption was expected to increase and correlate positively.

6. Off- farm income (HHOFFINC) - the household head that have off farm income are expected to adopt improved wheat varieties better than who have not off farm income. This variable also was treated as a dummy variable that if the farmer has off-farm income coded 1, otherwise, 0. As the number of farmers number increases to involve in off-farm work it was expected to increases adoption positively.

46

7. Access to credit (GECRSERV) - It is expected that those who have better access to credit can adopt improved bread wheat varieties than other who do not have access. Because it is expected that credit can solve the financial limitation of farmers. The variable in this study was treated as dummy variable in that, if the farmer gets credit service coded as 1 and 0, otherwise. As credit service increase adoption/Intensity of adoption was expected to increase and correlate positively.

8. Livestock ownership (TOTLIVUN) house holds that have large number of livestock are likely to adopt improved bread wheat varieties better than others who have less number of livestock Because those who have better number of livestock can have better opportunity to get credit. In this study it was assumed that livestock ownership and adoption would be related positively. As livestock ownership increases adoption/intensity of adoption was expected to increase and correlate positively.

9. Labor accessibility (OTSOLA1) those farmers who have access to labor are expected to adopt improved bread wheat varieties than those who lack labor accessibility since improved varieties required more labor. The variable has been treated as a dummy variable in that if the farmer has an access to labor coded 1, otherwise 0. As labor accessibility increases adoption/Intensity of adoption was expected to increase and correlate positively.

10. Social/leadership status of the respondent (PRTILEDE) - those farmers who have experience of leadership and better social status previously or currently are likely to adopt wheat technologies than others. Because, it is expected that they have an opportunity to get and interpret the information they get about improved bread wheat variety. The variable was coded as 1 and 0, otherwise. As the number of farmers increase to involve in leadership position adoption was expected to increase and to correlate positively.

11. Distances from extension agent office (DISDAOF1) those who are closer to extension agent are expected to adopt improved bread wheat varieties than others as a result of

47

accessibility. The variable was coded as 1 if the farmer is close to the DAs office and 0, otherwise/far. As distance of DA office increase adoption/intensity of adoption was expected to decrease and correlate negatively. As distance of DA office decrease the correlation will be vise versa.

12.Distance from input and credit supply institutions (CRINFAR1) those farmers closer is likely to adopt improved bread wheat varieties than those who are not close since they can easily facilitate and follow up the credit process. This variable was treated as dummy variable and had coded as 1 if the farmer is close, 0 if not close or if far. The far distance might affect negatively and the close distance affects adoption and Intensity of adoption positively.

13. Oxen ownership (OXTLU) those who have oxen for ploughing is likely to adopt improved bread wheat varieties since they could solve ploughing power problem. Then, oxen ownership and adoption were expected to relate positively. As the number of oxen owned by farmers increased, adoption/Intensity of adoption was expected to increase.

14. Farmers experience in any extension activities (YEXPEXTS)- Farmers who have long involvement in any agricultural extension activities is expected to use improved bread wheat varieties than with less experience. Then, this variable was hypothesized to correlate and influence positively improved bread wheat varieties adoption and the intensity of adoption.

15. Access to market (CRINMFF1) - Access to market was hypothesized to be positively related to the probability of adoption of improved bread wheat varieties in that if the house holds near to market tend to buy improved agricultural inputs including improved bread wheat seed and they can have easy access to dispose of and sell their production in the market. Therefore, the variable was treated as a dummy variable in that if the house hold has an access to market has coded as 1and 0, otherwise. As market distance increases adoption and intensity of adoption was expected to decrease.

48

16. Health status of the household head (HEALSTAT)-this is a dummy variable, which takes a value 1 if the household head was healthy and 0, otherwise. As farmers health statuses improve adoption and intensity of adoption of improved bread wheat varieties was expected to increase and correlate positively.

17. Farm land holding (SUMOWRE)-farmer who has large farm size is likely to adopt improved varieties than those who have lesser farm size. Because farmers with large farm size can distribute the yield loss risk and better land ownership serve as insurance to get credit which can use to purchase improved agricultural inputs. As farmers farmland holding increases adoption/intensity of adoption might increase and correlate positively.

49

4. RESULTS AND DISCUSSION

4.1. Analysis through descriptive statistics

This study was intended to examine the farmers evaluation and selection criteria of improved bread wheat varieties and to identify factors affecting adoption and intensity of adoption of improved bread wheat varieties in the study area Akaki as well as to know the effect of hypothesized independent variables on the dependent variables. In this section of analyses descriptive statistics such as mean standard deviation, percentage, frequency tabulation, t-test and chi-square test will be employed using SPSS- computer soft ware program.

In this study, adoption of a technology refers to a continued use of the technology on an area of land, which is large enough to contribute to the economy of the household. Here, the respondents who have cultivated improved bread wheat varieties and continued growing at least one of the distributed improved bread wheat varieties in the study area during the survey year and in any one of the years before the survey year of this study are considered as adopters. Farmers who never adopted and those who discontinued from growing of improved bread wheat varieties are categorized as non-adopters.

4.1.1. Sample Households Demographic Characteristics

In order to understand the sample households, it is very important to describe their demographic characteristics. The number of household head respondents was from two selected Rural Kebele Administrations or Peasant Associations namely Koye and GelanEdero. The sample house hold heads covered in this study from Koye PA/RKA were (76) 89.41% male and (9) 10.59 % female with a total of 85 constituted 56.67% which of the total sample and (56) 86.15% male and (9) 13.85% female with a total of 65 were from GelanEdero PA/RKA which constituted (43.33%) of the total sample as presented in Table 7.

50

Out of the total 150 respondents in the sample, adopters were 99 (66%) and non-adopters were 51 (34%). From 150 sample respondents, 132 (88%) were male and 18 (12%) were female respondents. From 132 male respondents 92 (69.70%) were adopters and 40 ((30.30%) were non-adopters. From total 18 female sample respondents 7 (38.89%) were adopters and 11 (61.11%) were non-adopters as presented in Table 7.

Table 7.Sample household heads distribution by Sex, Kebele and adoption category Adopters (99) Samples Male-sample Female-Sample Total Koye-Kebele Gelan-Edero Total Koye Kebele Male Female Total Gelan-Edero Kebele Male Female Total 32 4 36 88.89 11.11 100.00 24 5 29 82.76 17.24 100.00 56 9 65 86.15 13.85 100.00 58 5 63 92.06 7.94 100.00 17 5 22 77.27 22.73 100.00 76 9 85 89.41 10.59 100.00 N 92 7 99 63 36 99 Percent 92.93 7.07 100.00 63.64 36.36 100.00 Non-Adopters (51) N 40 11 51 22 29 51 Percent 78.48 21.57 100.00 43.14 56.86 100.00 Total (150) N 132 18 150 85 65 150 Percent 88 12 100.00 56.67 43.33 100.00

(Source: Computed from own survey data, 2005)

Out of the total 99 adopters 92 (92.93%) were male and 7 (7.07%) were female and from the total 51 non-adopters 40 (78.43%) were male and 11 (21.57%) were female. The numbers of sample household heads from Koye PA were 63 adopters and 22 non-adopters, where as from

51

Gelan-Edero PA, the number of sample household head adopters was 36 and that of nonadopters were 29 as indicated in Table 7.

In similar studies in the past, the major reasons for farmers to adopt improved and new technologies are technical, institutional, social, and economical reasons. Farmers do not adopt a technology if they are not convinced of its benefits, costs and risk associated with it. By seeing their fellow farmers, by attracting of high yield performances of improved varieties, market demand as well as DAs information and extension support farmers might urged and motivated to use and adopt improved varieties. On the other hand, the major reasons for those non-adopters might shortage of improved varieties, credit problem, or cash, land, labor or other farm resource constraints (Legesse et al., 2005).

Table 8.Marital status of respondents Marital status Married Un-married Divorced Widow/er Adopters (99) N 85 1 2 11 Percent 85.86 1.01 2.02 11.11 Non-Adopters (51) N 36 2.00 4.00 9.00 Percent 70.59 3.92 7.84 17.65 Total (150) N 121 3 6 20 Percent 80.67 2.00 4.00 13.33

(Source: Computed from own survey data, 2005)

The marital statuses of respondents are summarized in Table 8 as (121) 80.67% married, (3) 2% unmarried or single, (6) 4% divorced and 20 (13.33%) widow/er. The proportion of married respondents was much larger than the remaining marriage categories. As indicated in Table 8, the married adopters were 85.86 percent and that of non-adopters were 70.59 percent. The remaining categories of respondents constituted fewer proportions of respondents both in adopters and non- adopters.

52

Table 9.Association between adoption of improved bread wheat varieties and sex of sample household head

House hold head sex Female Male Total

Adopters 7 (7.1%) 92 ((92.9%) 99 (100%)

Non-adopters 11 (21.6%) 40 (78.4%) 51 (100%)

Chisquare

C.Coef

df

Sig.

Total 18 (12%) 132 (88%)

6.700***

0.207

0.01

150 (100%)

(Source: Computed from own survey data, 2005),

*** significant at 1%level

Table 10. Respondent farmers demographics characteristics Adoption Category Adopters Summary of statistics Minimum Maximum Range Minimum Maximum Range Minimum Maximum Range House hold head Age 19 80 61 20 80 60 19 80 61 House hold family size 1 11 10 1 10 9 1 11 10 Farming House hold head experience in extension experience 2 2 20 18 1 9 8 1 20 19 55 53 4 60 56 2 60 58

Nonadopters

Total

(Source: Computed from own survey data)

In adoption of new agricultural technologies, farmers age has an influential effect as it was observed in many adoption studies. The maximum and the minimum ages of total respondents were 80 and 19 years respectively. The adopters maximum age was 80, which was equal to the non-adopters maximum age. The minimum age of adopters was 19 years and that of nonadopters was 20 years. The age variation between maximum and minimum age of adopters, non-adopters and that of total respondents were 61, 60 and 61 respectively as presented in Table 10.

53

The respondents' maximum years of experience in extension were 20 for adopters 9 for nonadopters and 20 years for total samples. The minimum years of experience in extension for adopters was 2 years, for non-adopters 1 year and for total respondents was 1 year. The variations between maximum and minimum years of experience in extension were 18 years for adopters, 8 years for non-adopters and 19 years for total respondents as presented in Table 10. The maximum and minimum farming experience for adopters were 55 and 2 years, for nonadopters 60 and 4 years and for that of total respondents were 60 and 2 years. The variation between maximum and minimum total farming experience of adopters was 53, non-adopters were 56 and that of total respondents was 58 years as indicated in Table 10.

The maximum and minimum family size of adopters respectively were 11 and 1 for adopters, 10 and 1 for non-adopters and 11and 1 for total samples. The variation between maximum and minimum family size was 10 for adopters, 9 for non- adopters and 10 for that of total respondents as indicated in Table10.

Table 11.Adopters and non-adopters demographic characteristics Adopters (99) Characteristic Age Family size Experience in Extension (years) Farming Experience (years) Mean 46.10 5.85 7.87 21.90 SD 13.256 2.192 4.787 11.08 Non-Adopters (51) Mean 46.47 5.10 3.765 20.80 SD 14.53 2.385 1.784 10.98 T-test 0.157 1.927* -0.907*** -0.596 Significance Level (2-tailed) 0.876 0.056 0.000 0.552

***and* significant at 0.01and 0.10 p-value respectively.

The average age of adopters, non-adopters and total respondents were, 46.10, 46.47 and 46.23 years respectively. The S.D (Standard Deviation) of adopters, non-adopters and total respondents ages were 13.256, 14.53 and 13.655 respectively as indicated in Table 11.T-test

54

statistics was run to check whether there is a significant mean difference in age between adopters and non-adopters. The result of t-test showed that there was no statistically significant mean age difference.

The respondents average/mean and S.D (Standard Deviation) family size of adopters, nonadopters and total respondents were 5.85, 5.10 and 5.59 respectively as indicated in Table 11. T-test statistics was run to know whether there is statistically significant variation in average family size between adopters and non-adopters. The result of t-test analysis showed that there was statistically significant difference in average family size at 10 percent probability level as indicated in Table 11.

The respondents average (mean) and S.D (Standard Deviation) of experience in extension and total farming experience in years is presented in Table 11. The average years of experience in agricultural extension as well as the total farming experience of adopters were 7.87 and 21.9 that of non-adopters were 3.765 and 20.8 and the total respondents experience in extension were 4.485 and 11.023 years respectively. T-test was conducted to see the variation in average years of experience in agricultural extension and in total farming experience between adopters and non-adopters. The result of t-test analysis showed that there is a significant difference in average years of experience in agricultural extension participation involvement at 1 percent probability level as indicated in Table 11.

But the total farming experience was not significant in t-test analysis. Because those farmers who have better experience in extension could got better extension service that help them to adopt better improved bread wheat varieties.

From the total sample respondents 103 (68.67%) were involved in improved bread wheat production, while the remaining 47 (31.33%) respondents were not involved in improved bread wheat production during the survey year. Their reasons why they were not involved had summarized and presented in Table 12 from the response they gave during the interview.

55

Table 12.Reasons given for not using improved bread wheat varieties No 1 2 3 4 5 6 7 8 Reasons limiting to involvement in improved bread wheat production Farm land shortage Lack of information Fertilizer shortage High price of fertilizer Lack of extension support Labor problem Seed scarcity Lack of ploughing oxen N (47) 15 9 6 6 4 3 2 2 Percent Rank 32.00 19.15 12.77 12.77 8.51 6.38 4.25 4.25 1st 2nd 3rd ,, 4th 5th 6th ,,

(Source: Computed from own survey data, 2005)

Farmland shortage and lack of information were the two most important reasons that limit farmers in the study area. The remaining reasons were less important for farmers to adopt improved agricultural practices as indicated in Table 12.

Table 13.Level of awareness of improved bread wheat varieties Improved bread wheat Aware varieties N Percent HAR-1685 (Kubsa) 123 82 HAR-1709 (Mitike) Paven-76 54 115 36 76.67 Not aware N Percent 27 18 96 35 64 23.33 Total N Percent 150 100 150 150 100 100

(Source: Computed from own survey data, 2005)

56

Concerning respondents awareness of improved bread wheat varieties, interviews were conducted. About (123) 82% respondents knew HAR-1685, (115) 76.67% knew Paven-76

and 54 (36%) knew HAR-1709 variety. As indicated in Table 13, HAR-1685 was known by larger proportion of respondents variety by respondents. followed by Paven-76 and HAR-1709 was the least known

Table 14.Sample Farmers perception on benefit of fertilizer 1 1.1 1.2 1.3 1.4 1.5 2 Perception on fertilizer benefit Low profit No loss or no profit High profit Very high profit Encountered loss Total Perception on fertilizer Problems 2.1 High price (high interest rate) 2.2 Un-timely and lately arrival 2.3 Credit scarcity and credit service related problems to purchase fertilizer Total N 98 26 14 9 3 150 N 75 56 19 150 Percent 65.33 17.33 9.33 6.00 2.00 100 Percent 50 37.33 12.67 100

(Source: computed from own survey data, 2005)

Respondents were interviewed to know their opinion based on their experience about the benefits obtained from fertilizer use. About 65.33% low profit, for 17.33% reported no loss or no profit for 9.33% high profit, for 6% very high profit could be obtained and 2% said encountered loss. In this study the larger proportion of farmers reported the low profit from fertilizer as indicated in Table 14.

Respondents

were also interviewed to get their idea on problems related to fertilizer in their

area. About 50% of respondents have reported high price, 37.33% reported un-timely and late arrival and about 12.67 % reported credit scarcity and credit service related problems to

57

purchase fertilizer as indicated in Table15. The problems arisen due to weakness of the credit provider institutions and less attention of the government as observed during data collection time.

Table 15.Beginning time of cultivation of improved bread wheat varieties of sample farmers

No 1 2 3 4 5 6

Starting Years Before During ,, ,, ,, ,, 2000 2000 2001 2002 2003 2004 Total

N 9 10 27 28 27 2 103

Percent 8.74 9.71 26.26 27.30 26.26 2.02 100

(Source: computed from own survey data, 2005)

In this study to know their commencement or beginning time of using improved bread wheat production, respondents were interviewed and their responses were summarized in Table 16.

Table 16.Health status and adoption of improved bread wheat varieties Health status Non- adopters Un-healthy Healthy Total 7 (13.73%) 44 (86.27%) 51(100%) Adopters 7(7.07%) 92 (92.93%) 99(100%) 2-test df Co coef Sig. Total 14(9.33%) 136(90.67%) 150(100%)

1.762

0.108

0.184

(Source: Computed from own survey data, 2005); Co coef =contingency coefficient

To accomplish the agricultural activities as required, the farmers need to be healthy. In this study, it was tried to assess the household head respondents health situation. The respondents were grouped into healthy and un- healthy farmers (those who face a health problem) to

58

accomplish their day-to-day agricultural activities. From total adopters the healthy farmers were 92.93% and that of unhealthy were 7.07%. In the case of non-adopters 86.27% were healthy and 13.73% were unhealthy. Out of 150 respondents, (136) 90.67% were fully healthy and the remaining (14) 9.33 % had health problem. To check the relationship of the health situation of the respondents and adoption of improved bread wheat varieties, a chi-square test was conducted and the result showed that the relationship between health status and adoption of improved bread wheat varieties was not statistically supported and insignificant as indicated in Table 16.

Table 17.Sample household educational status Education Illiterates Literate Total NA 33(64.70%) 18(35.30%) 51(100%) Ad 64(64.65%) 35(35.35%) 99(100%) 0.000 1 0.001 0.994 X2 df Co. coef Sig. Total 97(64.67%) 53(35.33%) 150(100%)

(Source: Computed from own survey data, 2005); * NA=Non-adopters, Ad=Adopters; *Co coef =contingency coefficient.

Education is very important for the farmers to understand and interpret the information coming from any direction to them. Farmers education is also pivotal for the effective work of extension personnel because if the farmer has better education status he/she can has a capability to understand and interpret easily the information transferred to them from Extension Agent (EA). From total non-adopters 35.30% were literates and 64.70% were illiterates. In the case of adopters 35.35% were literates and that of 64.65 % were illiterate. The proportion (percentage) of illiterate adopters and non-adopters as well as that of literate adopters and non-adopters was almost equal as indicated in the Table 17.

59

In this study the literacy was extended from read & write to attending regular school education. To see the relationship and the intensity of relationship, the chi-square- test was conducted. But the result of chi-square- test was not statistically significant as indicated in Table 17. This means there is no any significant difference in adoption between adopters and non-adopters due to education.

4.1.2. Respondents` livestock and land ownership

In the study area mixed farming is practiced with crop and livestock production. Each household owns at least one or more types of livestock and a piece of land for crop and livestock production.

Livestock in the study area provides traction and manure and also serves as a source of income through sale of livestock and livestock products. Livestock also serves as a source of fuel in the study area. Crop residue and by-products serve as livestock feed source.

As it confirmed in many studies farmers who have better livestock ownership status are likely to adopt improved agricultural technologies like improved bread wheat varieties; because, livestock can provide cash through sale of them and their products and draught power for agricultural operations. In this study, it was revealed that the average livestock ownership of adopters and non-adopters in TLU were 6.834 and 5.02 respectively.

To know whether there is a variation in average livestock ownership between adopters and non- adopters and as a result if there is any significant difference due to the resource position, t-test was conducted. The result of t-test showed that there is a significant variation in average livestock ownership between adopters and non-adopters at one percent probability level as indicated in Table 18 and the average oxen ownership of adopters was also significantly larger (3.03) than non-adopters (2.157) at 5 percent probability level as indicated in Table 18. As ttest indicated, adopters had larger livestock and oxen ownership as compared to non-adopters.

60

This implied that large ownership of oxen and livestock can help farmers to adopt agricultural innovations by solving the power force need for improved wheat production practices and cash constraint by providing income from sale of livestock and their by products to purchase agricultural inputs.

Table 18.Livestock and land ownership of respondents farmers

Characteristics Livestock ownership (TLU) Oxen owner ship Improved Bread wheat land (ha) Total wheat land Total farm land (ha) Total land (ha)

Ads Mean 6.8340 3.0300 0.9621 1.2400 2.7141 3.01

SD 3.2724 1.6317 0.4579 0.6358 1.08 1.25

NAs Mean 5.0200 2.1570 0.0400 0.8530 2.0147 2.28

T-test SD 3.2120 1.6294 0.1759 0.3846 1.121 1.25 -3.236*** -3.107** -13.859*** -3.983*** -3.710*** -3.382***

Significance (2-tailed) 0.001 0.002 0.000 0.000 0.000 0.001

*** and ** Significant at 1 and 10 percent probability level

Sample farmers vary in their adoption of improved bread wheat varieties by their livestock and oxen ownership. Adopters average livestock ownership was significantly larger than nonadopters. This indicate that livestock ownership help farmers to adopt improved bread wheat varieties since the income from livestock obtained through selling of the animals or their by products can help to solve their financial limitation s to purchase inputs.

Land is the main asset of farmers in the study area. Farmers in the study area use both their own land and rent farm land for crop production and grazing land for livestock production .All 150 sample households have their own land and only (24) 16% and (2) 1.33% respondents rented cultivated and grazing land respectively. The average land holding of adopters was 3.01 hectares total average land holding, 2.7141 hectares total average farmland, 1.24 hectares total average wheat land and 0.9621 hectares was average farm land used for improved bread wheat

61

production. In the case of non- adopters the total average land holding was 2.28 hectares, 2.0147 hectares was average farmland, 0.8530 hectares was average wheat land and only among improved bread wheat growers of non-adopters was 0.0400 hectares in average was used for improved bread wheat production. To know whether there is the mean land holding variation, between adopters and non-adopters, t-test analysis was carried out and the result showed that there were the significance differences in all types of land holding at one percent probability level as indicated in Table 18.The result showed that farmers who have better land ownership can adopt improved bread wheat varieties better than non-adopters.

Table 19.Respondents land ownership in 1996/97 Ethiopian major cropping season Area of land ownership in hectares Total farm Land Total wheat land Total improved B.W.L. Max 6.5 4.5 3.25 Min 0.25 0.25 0.25 Range 6.25 4.25 3 .00 Average 2.44 1.125 1 .00 St.D 1.13 0.605 0.457

(Source: Computed from own survey data, 2005) * N.B=Improved B.W.L. (Improved Bread Wheat Land)

The total farmland ownership of respondents ranges from 0.25 hectares to 6.5 hectares. The total wheat land ownership of sample households ranges from 0.25 hectares to 4.5 hectares. The improved bread wheat land holding of sample household ranges from 3.25hectares. 0.25 hectares to

On the average sample households owned 2.44 hectares total farmland, 1.13 hectares total wheat lands and 0.96 hectares used for improved bread wheat production as presented in Table 19. In the study area respondents farmers allocated most of their farmland for wheat production as presented in table one.

62

4.1.3. Accessibility of respondents to different institutional services

In this study respondents were interviewed to get their opinion about the importance of extension services based on their experiences. About 84.67, 3.33% and 12% respondents have reported important, not important and do not have any opinion respectively.

The respondents have also been interviewed to give their opinion about the extension support they obtained. About 46.67% reported extremely weak due to un-availability of development agent, about 27.33% reported very weak even though the Development Agents are available around. The remaining 26% responded that the extension service they got in their area becomes extremely weak due to unknown reasons for them as indicated in Table 20.

Table 20.Respondents opinion on extension service of the study area

No 1. 1.1. 1.2. 1.3. 2. 2.1. 2.2. 2.3.

Respondents opinion on extension service About importance of extension service Important Do not have any idea Not important Total Status of extension service of the study area Extremely weak due to un-availability of DA Very weak even though the DA available Extremely weak due to un-known reasons for them Total (Source: Computed from own survey data, 2005)

N 127 18 5 150 70 41 39 150

Percent 84.67 12 3.33 100 46.67 27.33 26 100

Data were collected regarding the type extension service obtained by the respondents as indicated in Table.20. The whole non-adopters and 94.95% of adopters did not get extension service during the survey year on improved bread wheat variety. As indicated in Table 21 only

63

five respondent farmers reported that they got extension support. The respondents

were also

interviewed to get their opinion on the distance of DAs office from their home. As indicated in table 22 about 90% reported far and the remaining 10% reported close to their home.

Table 21.Extension support on improved bread wheat varieties and distance of DAs office Service accessibility Extension service on bread wheat Responses No Yes Total Far Close Total NAs 51 51 48 3 51 Ads 94 5 99 87 12 99 X2 df Sig. C.coef Total . 145 5 0.138 150 135 0.098 15 150

2.665*

0.103

Distance of DA office

1.456

0.228

*Significant at 10 % probability level. ; Ccoef = Contingency coefficient *NAs =Non-adopters, Ads = Adopters, df =Degree of freedom.

To know the association of extension service and distance of DA office with adoption of improved bread wheat variety, chi-square analysis was conducted. The result of chi-square analysis (2.665) showed that there is a significant association between extension service and adoption of improved bread wheat varieties at 10 percent probability level. But the chi-square test result of distance of DA office and adoption of improved bread wheat varieties was not significant as indicated in Table 21.

As it has indicated in many literatures, credit is considered as one of the favorable factors for improved agricultural technologies adoption because it can solve financial constraints of farmers to purchase and use improved agricultural inputs. Respondent farmers have reported about credit institution services and related problems in their area based on their experience. Of that, 83 (55.33%) have reported that there is scarcity, 26 (17.33%) reported that there is a complex and boring procedures and the remaining 41(27.34%) reported that there is a high interest rate problems as indicated in Table 22.

64

Table 22.Summary of respondents opinion on credit

No 1 2 3

Responses on credit related problems Scarcity High interest rate Complexity of procedures Total Responses on the importance of credit

N 83 41 26 150

Percent 55.33 27.34 17.33 100.00

1 2

Credit is important Credit is not important Total Suggestions for better credit services Easy and Reduced procedures Low Interest rate Total

101 49 150 59 42 101

67.33 32.67 100 58.42 41.58 100

1 2

(Source: Computed from own survey data, 2005)

Respondent farmers were also interviewed to provide their opinion about the importance of credit and their suggestions for better credit service in the future. The respondents responses on these issues were summarized and presented in Table 22. About 67.33% respondent farmers reported that credit is important and the remaining 32.67% reported that credit is not important.

From those 101 respondent farmers who supported the credit service as important also provided their opinion for better credit service. About (59) 58.42% reported reduced processes and procedures and about (42) 41.58% suggested to reduce the interest rate as indicated in Table 22.

65

As presented in Table 22 from the total 150 sample respondents, only 26 adopters and 8 nonadopters with a total of 36, which constituted 24% of the total respondents, were credit users particularly from individual credit sources in the mean time when this study conducted. In credit utilization, adopters were larger in proportion than non-adopters. Almost all of the credit users of sample respondents have reported that the major credit sources for them were informal and private lenders.

Table 23.Association between credit and market service Accessibility Credit Service Response NA No Yes Total Far Close Total 43(84.31) 8(15.69) 51(100) 43(84.31) 8(15.69) 51(100) Ad 71(71.72) 28(28.28) 99(100) 89(89.90) 10(10.10) 99(100) 0.994 1 0.319 0.081 X2 df Sig. Co.coe Total 114 36 150 132 18 150

2.928*

0.087

0.138

Market Access

*Significant at 10% probability level; Numbers in brackets are in percentage

To see the association between adoption of improved bread wheat varieties and credit service, a chi-square test was carried out. The result showed that there is a significant relationship between adoption of improved bread wheat varieties and credit services at 10 percent probability level as indicated in Table 23. Market accessibility is also another important factor for farmers to adopt improved agricultural inputs. If farmers are closer and having access to credit services they can easily purchase improved agricultural inputs and sell their agricultural outputs without moving long distances. Farmers also motivated to use improved agricultural inputs if they have access to attractive market for their output to sell in good price. In this study respondent farmers were interviewed to provide their idea regarding the market accessibility. About 84.3% non- adopters and about 89.90% have reported far from market and the remaining 15.69% non-adopters and 10.10% adopters reported close to market access as

66

indicated in Table 23. A chi-square-test analysis was carried out to check the association between market access and adoption of improved bread wheat varieties. The result showed that the relationship was not statistically significant as indicated in the Table 23.

Table 24.Summary of households accessibility of off-farm job

No 1

Respondent farmers access to off-farm job Have access to off-farm job Adopters Non-adopters

N 26 17 9 124 82 42 150

Percent 17.33 65.38 34.62 82.67 66.13 33.87 100.00

Have not access to off-farm job Adopters Non-adopters Total (Source: computed from own survey data, 2005)

Income from off-farm job can play a great role in adoption of improved agricultural technologies. Because, it has hypothesized that the income from off-farm can solve farmers financial constraints to purchase and use improved agricultural inputs. In this study about 17.33% sample households reported that one of their family members has off-farm job and the remaining 82.67% do not have family members who have off-farm job. From the total 26 sample households that have off-farm job 65.38% were adopters and the remaining 34.62% were non-adopters as indicated in Table 24.

67

Table 25.Respondent farmers reasons for not involvement of their family in off-farm job No 1 2 3 4 5 Reasons Under and over aged Students Work on the house hold farm Do not have family members Less income from off-farm job Total (Source: Computed from own survey data, 2005) N 42 12 52 16 2 124 Percent 33.87 9.68 41.94 12.90 1.61 100

From the total 124 sample respondents whose family members were not involved in off-farm activities had reported their reasons during the interview why the house hold members did not involve in off-farm job. As indicated in Table 25, about 33.87% have reported that their family members couldnt involve in off-farm job due to the under and over age, 9.68% reported that their family members are students, 41.94% described the time constraint since the household members would work on the house hold farm, 12.9% reported that they do not have family members and 1.61% reported less income from off-farm job.

Table 26.Rrespondents opinion on decision of off-farm and other household resources No 1 2 3 4 Total : (Source: Computed from own survey data, 2005) Respondents opinion on house holds resources decision maker Husband Husband and wife Wife House hold members together N 121 27 1 1 150 Percent 80.66 18.00 0.67 0.67 100.00

68

Concerning the decision on the off-farm and other agricultural income, the household head respondents were interviewed. All of them were reported that the house hold member who involved directly in the off-farm job make a decision for what purpose the income from offfarm job need to be used. But concerning the other agricultural incomes, from the total 150 respondents about 80.66%, 18%, 0.67% and 0.67% reported that the decision were made by husband, by husband and wife, by wife and by the house hold members together, respectively as indicated in the Table 26. In this study, almost all decision on the agricultural resources of the farming household made by the husband.

Table 27 Pattern of off-farm income utilization of respondent farmers No 1 2 3 4 5 Total Use of off-farm income Household food consumption Cloth purchase Health treatment Input purchase Labor hiring N 7 10 5 3 1 26 Percent 26.92 38.46 19.23 11.54 3.85 100.00

(Sources Computed from own survey data, 2005)

On the use of off-farm income, the Total 24 household respondent farmers who themselves and their family members had off-farm job reported about their and their family members offfarm income utilization. About 26.92 %, 38.46 %, 19.23%, 11.54% and 3.85% reported for household food consumption, cloth purchase, health treatment, input purchase, and labor hiring purposes respectively as indicated in Table 27. In this study the order of importance in off-farm income utilization from higher to the lower were, for cloth purchase, food consumption, health treatment, labor hiring and input purchase. Allocation of off-farm income to agricultural input purchase took the least proportion.

69

Table 28.Family labor utilization of respondent farmers No 1 2 Family labor utilization Utilized family labor Not utilized family labor Total Ad NA Total Percent 91.33 8.67 100

93(93.94) 44(86.27) 137 6(6.06) 99(100) 7(13.73) 51(100) 13 150

(Source: Computed from own survey data, 2005)

In this study the respondent respondents` labor source and their family labor utilization were revealed through interview. From the total 150 respondents, about (137) 91.33% have reported that they used their family labor on their farm activities for weeding, harvesting, threshing, plowing and sowing as indicated in Table 28.

Table 29.Types of activities and family labor utilization of respondents No 1.1 1.2 1.3 1.4 1.5 No 2.1 2.2 2.3 1.Activities on which family labor used Ploughing Sowing Weeding Harvesting, Threshing 2.Critically labor required activities Weeding Sowing Ploughing N 5 2 120 7 3 Total 137 N 109 28 13 Total 150 Percent 3.65 1.46 87.59 5.11 2.19 100 Percent 72.66 18.67 8.67 100

(Source: Computed from own survey data, 2005)

70

Out of these sample households who used their family members labor on the household farm, about 67.9% were adopters and the remaining 32.10% were non-adopters. Respondents were also interviewed to describe the type of agricultural activities they used their family labor. From the total 137 respondents who used their family labor on their farm, about 87.59%, 5.11%, 2.19%, 3.65%, and 1.46% reported for weeding, harvesting, threshing, plowing and sowing respectively. Concerning the critical labor requirements of the respondents labor

requirement were about 8.67% were for plowing and sowing, 18.67% for weeding, and 72.66% were reported for harvesting and threshing as indicated in Table 29.

Table 30.Respondents accessibility to non-family labor and to off-farm income Access to Labor outside the house hold labor Access to off-farm income Response No Yes Total No Yes Total NA 48(94.12) 3(5.88) 51(100) 42(82.35) 9(17.65) 51(100) Ad 87(87.88) 12((12.12) 99(100) 82(82.83) 17(17.17) 99(100) 1.456 1 0.228 0.098 X2 df Sig. C.coef Total 135 15 150 124 26 150

0.005

0.942

0.006

(Source: computed from own survey data, 2005) * Numbers in brackets are in percentage

To check the association between off-farm income of the sample household and adoption of improved bread wheat varieties chi-square analysis was carried out and the result showed that there is no a systematic association statistically supported between adoption of improved bread wheat varieties and off-farm income as revealed in this study as the result presented in Table 30. To see the labor source and accessibility of respondents to labor outside the house hold

labor and its relation ship with adoption of improved bread wheat varieties, chi-square test was conducted and the result showed that the two variables, adoption and the utilization of labor outside the household labor is not statistically significant as indicated in Table 30.

71

Table 31.Respondent farmers labor sources outside their family members

No 1 2 3 4

Types of labor sources Employed labor Exchange labor Relatives and colleagues support Not used labor outside their family labor Total

N 80 41 15 14 150

Percent 53.333 27.333 10 9.333 100

(Source: Computed from own survey data, 2005

Respondents were also interviewed for their labor source other than their family labor. About 53.33%, were reported that they used hire or employed labor, about 27.33% reported as they used exchange labor, 10% were report that they used support labor from relatives and colleagues and 9.33% were reported that they do not used labor outside their family labor source. The respondents labor source outside the family members labor, employed and exchange labor was very important. In the study area the agricultural activities required the higher labor are harvesting, threshing and weeding as indicated in Table 31.

4.1.4. Agricultural information sources of the study area

Access to information or extension messages as well as various extension services was one of the institutional characteristics hypothesized to influence farmers decision to adopt a new technology. One can gain access to information about new technologies through various means such as attending field days, visiting demonstration fields, participating training, listening to agricultural programs on radio, through contact with Extension or Development Agents, and through various forms of communication with neighbors, relatives, other colleague farmers and leaders of community, religious and PA (Peasant Associations) and through other means (Tesfaye et al, 2001).

72

As shown in various literature, different extension methods such as training, demonstration, farmers field day, farmers meetings (mass-meeting), group discussions, posters, mass communication methods and other extension methods have described that can be employed to transfer extension messages to the farmers. Using and practicing these extension methods properly to transfer extension messages can facilitate diffusion and adoption of improved agricultural inputs.

In this study the improved bread wheat grower respondents were interviewed to give their opinion how they got extension messages regarding the utilization and application of improved bread wheat varieties production and management. Out of 103 respondent farmers who grew improved bread wheat during the survey year, only (4) 3.88% were non-adopters and the remaining (99) 96.12 were adopters. From 99 total adopters only (3) 3.03% reported that they got training on improved bread wheat varieties production and management as presented in Table 32.

As shown in the Table 35 about 100 of improved bread wheat growers (adopters and non adopters) who do not got training were interviewed how they precede the improved bread wheat varieties production. About 89% of improved bread wheat growers reported by seeing other grower farmers, (copying mechanism or farmer to farmer extension), about 8% reported by trial and error method and the remaining growers reported by asking the help of Development Agent and other educated people living in their area.

To know the field day and demonstration program participation of Improved bread wheat growers, their interview responses had summarized as indicated in Table 32., only 11.77% respondents told that they got an opportunity to attend field day and demonstration program. But the remaining (91) 89.23% had not have it. Out of these 12 respondents only 8.33% were non-adopters and 91.67% were adopters.

73

Table 32.Respondents participation in training, field day and demonstration No 1. 1.1. 1.2. 2. 2.1. 2.2. 3. 3.1. 3.2. 3.3. Training, field day and demonstration participation Adopters (99) Non-adopters (4) Total (103) Total (103) 3 (2.91) 100 (97.09%) Total 103 12 (11.77%) 91 (88.23 %) Total 150 17 (11.33%) 67 (44.67%) 66 (44 %)

Training participation Attend training 3 (3.03%) Not attending training 96 (96.97%) Field day and Demonstration program participation Attending training 11 (91.67%) Not attending training 88 (96.7) Attending extension meeting called by DA Feel happy to attend Un-happy No feeling (Source: Computed from own survey data, 2005)

4 (100%)

1 (8.33%) 3 (3.3%)

The respondent farmers were also interviewed to know their feeling when called by DA for extension meeting Regarding their feeling they were show when they receive DAs call for extension meeting; about 11.33%, 44.67% and 44% were reported that they feel happy, unhappy and did not have any feeling on this issue respectively as indicated in Table 32.

In the study area respondent farmers were interviewed to provide their idea regarding their agricultural information sources. As it is presented in Table 33, neighbors and colleague farmers, DA, community leaders, farmers field day, PA leaders, demonstrations, radio, newspaper/news letters, publications, posters, training programs, TV and religious leaders have served as sources of general and agricultural information sources for them.

74

Table 33.Respondent farmers sources of information

Information Sources Neighbors and colleague farmers DA Community leaders Farmers field day PA leaders Demonstration Radio News paper/News letter Other publication Poster Training TV Religious leaders (Source: Computed from own survey data, 2005)

N 138 136 133 126 123 123 123 122 122 107 105 104 101

Percent 92 90.67 88.67 84 82 82 82 81.33 81.33 71.33 70 69.33 67.33

Rank 1st ,, 2nd 3rd 4th ,, ,, 5th ,, 6th 7th 8th 9th

As in Table 33 presented farmers neighbor and colleagues are the major and the firs important farmers source of information. This survey result is similar with the result of group discussion conducted in this study. According to this study DA serve as the second information source. The survey result showed that the third and fourth sources of information are community leaders and farmers field day respectively. As showed in the Table 33 PA leaders, demonstration and radio serve as fifth source of information. Newspaper/news letter and other publications serve as sixth information source. The remaining, poster, training TV and religious leaders serve as seventh, eighth, ninth and tenth sources of information respectively as indicated in Table 33.

75

In this study, it was also tried to summarize the agricultural information sources of the farmers in the study area through group discussion. During the time of group discussion the group members were familiarized to the discussion point and were expected to identify and prioritize the agricultural information sources of farmers in their area .The group members took care in listing of all alternative sources of information available in their area using brain storming method and tried to refined, summarized and prioritized the listed alternative information sources listed through brain storming method.

The result of the group discussion showed that; a neighbor stands first and the most important and TV stands the last and least important. The result of the group discussion findings showed that farmers got more information easily from their neighbors than other sources available in their area. The second most important information sources of farmers in the study area were religious and community leaders. PA leaders and DAs serve as third and fourth respectively as sources information. Demonstration and field day training and posters serve as fifth, sixth and seventh sources of information respectively. The remaining, publications, radio and TV serve as eighth, ninth and tenth sources of information respectively for the farmers in the study area.

4.1.5. Farmers selection and evaluation criteria of improved bread wheat varieties

Varieties characteristics play a vital role in adoption of improved varieties if their characteristics satisfied the need, interest and in line with the environmental situations of the farmers. The information on evaluation and selection criteria of improved bread wheat varieties of the farmers in the study area was analyzed through personal interviews and group discussion. The procedure to analyze the information through group discussion was conducted as; first make familiar farmers to the discussion agenda, and let them to establish and set evaluation and selection criteria of improved bread wheat varieties disseminated in their area. In the process of setting and establishing the criteria the group were applied the method of brain storming and list down all the ideas provided and forwarded by the group members .The group continued to refined the ideas forwarded by the group members and set or established

76

the evaluation and selection criteria with a common agreement. In this study, the result of the group discussion showed that the best improved bread wheat variety should constituted the white grain color, large seed size, and high disease, pest and frost resistance, good food quality, good straw quality as animal feed and attractive market demand characteristics. This study is in line with the study of (Ethiopian Rural Self Help Association /ERSHA, 2000).

Table 34.Farmers evaluation and selection criteria of improved bread wheat varieties

No Variety Characteristics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 White grain color Large grain Size Straw quality Market demand Germination capacity Cooking quality Better yield performance Water lodging resistance Tillering capacity Food quality Short maturity date Disease resistance and pest resistance Frost resistance Harvesting quality Storage quality

N 140 140 140 140 139 139 139 138 138 138 137 135 133 97 97

Percent 93.33 93.33 93.33 93.33 92.67 92.67 92.67 92.00 92.00 92.00 91.33 90.00 88.67 64.67 64.67

Rank 1st ,, ,, ,, 2nd ,, ,, 3rd ,, ,, 4th 5th 6th 7th ,,

(Source: Computed from own survey data, 2005)

The results in evaluation and selection of improved bread wheat varieties disseminated in the study area has showed in Table 34 that white grain color, large grain size, market demand and straw quality were the first and most important criteria. The traits such as better yield performance, cooking quality and germination capacity got the second rank. Food quality, tillering capacity and water lodging resistance got the third rank Short maturity date, pest and

77

disease resistance and frost resistance were got the fourth, fifth and sixth ranks respectively. Harvesting and storage qualities were got the seventh rank by farmers judgment.

Table 35.Farmers preference (selection and evaluation criteria) of improved bread wheat varieties disseminated in the study area No Variety Characteristics 1 Market demand 2 Cooking quality 3 Water logging rsistance 4 Straw quality 5 Storage quality 6 Frost resistance 7 Seed size 8 Yield performance 9 Rain shortage and Drought 10 resistance Weed resistance HAR-1685 (123) HAR-1709 (54) N Percent N Percent 97 86 86 84 83 80 79 77 77 76 78.8.6 69.92 69.92 68.29 67.48 65.04 64.23 62.60 62.60 61.80 57.73 56.91 53.66 12 23 27 36 33 27 26 38 32 42 37 32 40 405 31.154 3rd 22.22 42.60 50 66.67 61.11 50 48.15 70.37 59.26 77.78 68.52 59.26 74.04 Paven-76 (115) N Percent 41 41 37 30 34 43 45 35 41 32 42 45 44 510 39.231 2nd 35..65 35.65 37.12 26.10 29.57 37.40 39.13 30.43 35.65 27.82 36.52 39.13 38.26 -

11 Grain color 71 12 Food quality 70 13 Disease and Pest resistance 66 Sum 1032 Average 79.385 Rank 1st

(Source: Computed from own survey data, 2005)

Respondent farmers were interviewed to get their idea on evaluation and selection criteria of improved bread wheat varieties and their responses has summarized in Table 35.The respondent farmers responses showed that the improved bread wheat varieties should constitute the characteristics mentioned in Table 35. These evaluation and selection criteria are the most important criteria for the farmers in the study area. The respondent farmers have given their preference of improved bread wheat varieties distributed in their area. The larger

78

proportion of respondent farmers selected HAR-1685 improved bread wheat variety. The remaining varieties Paven 76 and HAR, 1709 ranked second and third respectively. This survey result was also supported and similar result was obtained from group discussion conducted in this study.

4.2. Analytical results and discussion

The purpose of this section is to identify the most important hypothesized independent variables that influence the dependent variables namely the probability of adoption for nonadopters using logit model and the intensity of adoption for adopters using tobit model analysis of improved bread wheat varieties in the study area, Akaki. Before conducting the model analysis selection, screening and verification of hypothesized variables were conducted by considering various situation to get best variables, those can fit with the analytical models, describe the sample groups, environmental and practical situation of the study area. This was done in consultation of professionals and experienced people, based on literatures, practical situations, observation and experience of the researcher and the relevance as well as the importance of the variables. As a result, the variable, distance of credit provider institutions was dropped because the major credit source for the farmer in the study area were the private individual credit providers in time when this research was conducted. These individual credit providers do not have a specific place and including this variable in model analysis was not relevant.

In the case of significant level of hypothesized independent variables, independent sample test between the groups using t-statistics or t-test for continuous variables to describe the pattern of sample data and to test the significance of a given independent variable on adopters and nonadopters groups as well as to check the mean values differences of continuous variables in the two groups and the chi-square test also to test the differences between the two groups for discrete variables in relation to dependent variables (Lind and Mason, 1994 as cited in Adane, 2002). In the analysis some independent variables might show significant and others might show insignificant relationship with dependent variables. The insignificant association doesnt

79

guarantee about the strength or direction of relationship between those insignificant hypothesized independent variables and the dependent variables. The reason for the insignificant relationship of some of the independent variables is mainly because of the fact that there is a drawback with any univariate approach in that it ignores, but there could be a possibility in the collection of variables analysis, each of which is weakly associated with the univariate outcomes can become an important predictor of out come when taken together. Therefore, we should consider them as candidates to be indicated in the multivariable models analyses along with all known important variables (Hosmer and Lemeshow, 1989 as cited in Adane, 2002).

Moreover, there are several literatures and previous research works (Chilot, 1994; Bekele, 2000; Adane N.F., 2002; Adane N.M., 2002; Techane 2002; Endrias, 2003; Yitayal, 2004; Adam and Bedru, 2005) conducted in a similar way which can substantiate this study. These previous research results showed that those hypothesized independent variables were included in econometrics model analyses regardless of the significant or insignificant results of these hypothesized independent variables in descriptive analysis. The model analyses results might show significant or insignificant results differently or similarly to descriptive statistics results. Regarding multicollinearity or high degree of association problem among selected and screened hypothesize independent variables were primarily checked before including in the models as well as before running the models analyses.

Secondly, prior to running the Logit and Tobit models, the presence or absence of correlations or associations between hypothesized independent and dependent variables were checked. The presence or absence of correlation or association, that is, whether or not there is a correlation between the variables in question (Sarantakos, 1998).

Existence, direction and strength of correlation are demonstrated in the coefficient of correlation. A zero correlation indicates that there is no correlation between the variables. The sign in front of the coefficient indicates whether the variables change in the same direction (positive correlation) or in opposite direction (negative correlation), except for nominal

80

measures, where the sign has no meaning, in which case coefficient describe only the strength of the relationship (a high or a low association) between the variables of the study. The value of the coefficient shows the strength of the association with values close to zero meaning a weak correlation and those close to 1 a strong correlation. A correlation of +1is just as strong as one of -1; it is the direction that is different (Sarantakos, 1998). Therefore, in this study, the presence or absence of association or correlations of hypothesized independent variables with the dependent variable, adoption of improved bread wheat varieties were assessed to identify and drop from model estimation if hypothesized independent variables do not have any relationship with dependent variable, adoption of improved bread wheat varieties. The Cramers v coefficient for discrete variables and Pearsons correlation coefficient for continuous variables was calculated using SPSS computer program.

In addition, the direction, range, intensity or degrees of strength of association of each hypothesized variables with dependent variables were assessed. The result showed that there was no total absence of association between hypothesized independent variables and dependent variables. Rather, association between hypothesized independent and dependent variables exist with various degrees of association ranging from moderate to weak. As a result, it was decided to include all selected, verified, screened hypothesized independent variables, those have various degrees of relationship with dependent variables, in models analyses to see their combined effect they have on dependent variables namely probability and intensity of adoption. From these total selected independent variables, only farmland showed moderate correlation with the dependent variable. But the rest showed weak association as indicated in Appendix table 24.

Thirdly, before including the hypothesized variables and running the model analyses the existence of a serious of multicollinearity or high degree of association problem among independent variables for all continuous and discrete variable were checked. There are two measures that are often suggested to test the existence of multicollinearity or association problems among independent variables. These are: Variance Inflation Factor (VIF) for multicollinearity problem among continuous independent variables and contingency

81

coefficients for existence of high degree of association among independent dummy variables. The technique of variance inflation factor (VIF) was employed to detect the problem of multicollinearity for continuous variables. VIF shows how the variance of an estimator is inflated by the presence of multicollinearity (Gujarati, 2003).

It is obvious that multicollinearity problems might arise when at least one of the independent variables shows a linear combination of the others; with the rest that we have too few independent normal equations and, hence, cannot derive estimators for all our coefficients. More formally, the problem is that a high degree of multicollinearity results in larger variances for the estimators of the coefficients. A larger variance implies that a given percentage (eg.95%) confidence interval for the corresponding parameter will be relatively wide; a large range of values of the parameter, perhaps including the value zero, will be consistent with our interval. This suggests that, even if the corresponding independent variable problem may make it quite difficult for us to estimate accurately the effect of that variable. Consequently, we may have little confidence in any policy prescriptions and biased on these estimates (Kelejian and Outes, 1981).

Very often the data we use in regression analysis cannot give decisive answers to the questions we pose. This is because the standard errors are very high or the t-ratios are very low. This sort of situation occurs when the explanatory variables display little variation and/or high inter-correlations. The situation where the explanatory variables are highly inter -correlated is referred to as multicollinearity (Maddala, 1992).

According to Maddala (1992), VIF can be defined as: VIF (xi) =

1 1 Ri2

Where Ri2 is the square of multiple correlation coefficients that results when one explanatory variable (Xi) is regressed against all other explanatory variables .A statistical package known as SPSS was employed to compute the VIF values. Once VIF values were obtained the R2 values can be computed using the formula. The larger the value of VIF, the more will be trouble-some or the collinear of variable Xi. As a rule of thumb, if the VIF of a variable

82

exceeds 10, there is multicollinearity. If Ri2 exceeds 0.90, that variable is said be highly collinear (Gujarati, 2003). The VIF values displayed in Table 36 have shown that all the continuous independent variables have no multicollinearity problem.

Similarly, contingency coefficients were computed from survey data to check the existence of high degree of association problem among discrete independent variables. Contingency coefficient is a chi-square based measure of association .A value of 0.75 or more indicates a stronger relationship (Healy, 1984 as cited in Destaw, 2003).

The contingency coefficients are computed as: C=

2 N + 2

Where, C= Coefficient of contingency

2 = Chi-square random variable and

N = total sample size.

Which assumes a value between 0 and 1 to indicate the degree of association between the discrete variables as indicated in Table 37.The decision rule for contingency coefficients says that when its value approaches 1, there is a problem of association between independent discrete variables. As indicated in Table 37 that there is no a problem of high degree of association among independent discrete variables.

83

Table 36.Variable Inflation Factor for the continuous explanatory variables Variables Age TLU Farm land holding Oxen ownership Experience in extension Family size R2i 0.047 0.637 0.342 0.638 0.036 0.231 Variance Inflation Factors (VIF) 1.049 2.754 1.520 2.759 1.036 1.301

(Source : Own Computation)

Table 37.Contingency Coefficients for Dummy Variables of Multiple Linear Regressions Model HHH SEX HHHSEX EDUHHH HEALSTAT PRTILEDE HHOFFINC DISDAOF1 CRINMFF1 OTSOLA GEXSERVE GECRSERV Source: Own computation 1 EDU HHH 0.184 1 HEAL STAT 0.023 0.003 1 PRTI HHOF LEDE 0.128 0.044 0.155 1 FINC 0.154 0.030 0.035 0.088 1 DIS DAOF1 0.055 0.033 0.046 0.172 0.082 1 CRIN MFF1 0.073 0.058 0.23 0.130 0.048 0.442 1 OTS OLA 0.14 0.196 0.046 0.106 0.094 0.110 0.055 1 GEXS GECR ERVE SERV 0.068 0.018 0.068 0.097 0.085 0.062 0.046 0.183 1 0.063 0.024 0.034 0.080 0.406 0.021 0.174 0.031 0.155 1

As it has indicated in many studies and literatures, if there will be serious multicollinearity or a high degrees of association problems among independent variables, these situations can

84

create difficulties to differentiate the separate effects of independent variables on dependent variables and also seriously affect the parameter estimate because of strong relationship among them. Hence, should not be included in the model analysis (Maddala, 1983; Kathari, 1990 as cited in Adane, 2002 and Gujarati, 1995). But since there is no a serious multicollinearity or high degree of association problem among independent variables in this study all the screened variables were decided to be included in the models analyses.

After conducting and passing all these steps, all screened and verified independent variables were included in logit model analysis using SPSS computer software program. But a problem faced in tobit model analysis to include all these screened and verified independent variables in tobit model analysis using Limdep computer soft ware program due to the limitation of this soft ware program to accommodate all variables included in logit analysis. Therefore, there need to select and choose the variables that can be accommodated by the Limdep soft ware program and most important independent variables for the analysis than others. As a result, from those independent variables included in the logit analysis only leadership position and credit service were dropped from tobit model analysis based on practical and actual situations, researchers observation, relevance of the variables and by employing Limdep computer program to check the number of significant variables those can affect the dependent variable, intensity of adoption of improved bread wheat varieties.

In this process the number of significant independent variables were increased when the above two independent variables were dropped individually or together. At last, the remaining screened and verified hypothesized independent variables were included in tobit model analysis.

4.2.1. Analysis of determinants influencing probability of adoption of improved bread wheat varieties and their marginal effect

To identify factors among hypothesized independent variables that significantly influencing the probability of adoption of improved bread wheat varieties in the study area, Akaki, SPSS

85

computer soft ware program and the binomial econometric analytical model (the binary Logit model) was employed. In fitting the logistic estimation model, the higher significance of chisquare statistics (80.187) was taken as a measurement of goodness-of-fit. This indicates that the explanatory variables together influence the probability of adoption of improved bread wheat varieties in the study area. In addition, the model correctly classified the respondents into adopters and non-adopters at 81.33% of correct prediction percentage. The maximum likelihood estimate of the parameters and the direction of relationship and the effect of independent variables on probability of adoption were analyzed and presented in Table 38.

As indicated in the methodology and other previous sections, a number of independent explanatory factors were postulated to influence the probability of adoption of improved bread wheat varieties in the study area. Among the selected hypothesized explanatory variables and considered by the model, only four variables were found to have significantly affect farmers adoption decision of improved bread wheat varieties. The variables affecting probability of adoption were distance of DA-office from the farmers home (DISDAOF1), house hold social/leadership status (PRTILEDE), market accessibility (CRINMFF1), and house hold farmers experience in extension (YEXPEXTS) as indicated in the Table 38.

Among those significant variables, only one variable, which was market access, related with adoption of improved bread wheat varieties negatively and the sign was different from the expectation but statistically significant at 5 percent probability level. In this study, the negative relationship of market access and adoption of improved bread wheat varieties showed that those farmers in the study area who do not have access to market are more likely to adopt better the improved bread wheat varieties than those farmers who have a better access to market.

The possible reason for this situation might be, those farmers who have better and closer access to market area might create other income opportunity from their farm and they may give more attention and priorities to these other alternatives, production activities and other

86

substitutions, which may bring better income to them than using their whole wheat farm to produce improved bread wheat. But those farmers far away from market since they may not have any other alternatives they give more attention for improved bread wheat production and to their farming occupation. The other possible reason to these farmers who far away from market make them to adopt the improved bread wheat varieties than those who are closer to market might be the better production performance of the varieties to provide food for their family through out the year since they might not have other food sources of alternatives and means .On the other way, environmental situations, the soil fertility frost problem variations might be the possible reasons.

The remaining three significant explanatory variables namely (leadership position/status, experience in extension and distance of DA-office from the farmers home) related with adoption of improved bread wheat varieties significantly and positively, as of the expectations, at 1, 1 and 10 percent probability levels respectively as indicated in the Table 38.

87

Table 38.Factors affecting Probability of adoption of improved bread wheat varieties and the marginal effect of the significant explanatory variables Variables HHHSEX-sex HHHAGE-age EDUHHH-education HEALSTAT-health PRTILEDE leadership position HHOFFING-off-farm income DISDAOF1-DA-office TOTLIVUM -livestock SUMOWRE- farm land CRINMFF1-market OTSOLA1-labor OXTLU-oxen GEXSERVE -extension YEXPEXTS-extension experience GECRSERV -credit FAMILYSI family size Constant B 0.657 -0.023 -0.156 0.942 2.217 -0.688 1.490 0.076 0.361 S.E 0.777 0.021 0.536 0.901 0.756 0.945 0.910 0.146 0.301 Wald df Sign. 0.714 1 0.398 1.194 1 0.274 0.094 1 0.772 1.092 1 0.296 8.610 1 0.003*** 0.531 2.683 0.272 1.435 1 1 1 1 1 1 1 1 1 1 1 0.466 0.101* 0.602 0.231 Exp (B) 1.928 0.977 0.856 2.565 9.181 0.502 4.436 1.079 1.435

-1.636 0.819 0.3.993 0.110 1.228 0.008 0.103 0.278 0.137 7.811 23.647 0.109 0.475 0.200 0.024 -4.053 0.113 0.788 0.135 1.744 17.690 0.064 0.030 5.399

-0.046** 0.195 0.928 1.117 0.712 1.108 0.741 2467.236 1.608 1.221 1.024 0.017 0.800 0.862 0.020

1 0.000***

Notes: Exp (B) shows the predicted changes in odds for a unit increase in the predictor *Omnibus Tests of model coefficients: Chi-square=74.97, Sign.0.000; * Percentage of correct prediction=81.30; and *, **and ***Significant at 10%, 5%, and 1% Significant level.

The variable leadership position affects adoption significantly in the study area as indicated in Table38. Farmers who have a leadership position in the society might give a better opportunity to access resources and inputs such as labor, fertilizer, seed, to contact with DA for better information, better access to credit providers, as a result of their leadership position and, hence, are likely to adopt improved bread wheat varieties better than those who did not have leadership position in the society this result is in line with the result of Rauniyer and Goode (1996) as cited in Legesse (1998). This implies that there need to give attention and identify what those farmers who do not have leadership position lack due to their lower leadership

88

position to design a strategy to provide access, support, encourage them to achieve better adoption of improved bread wheat varieties by them.

The variable, experience in extension influences adoption of improved bread wheat varieties. Farmers who have longer years of experience in extension have adopted better-improved bread wheat varieties than those who have the lower years of experience in extension participation. This showed that the farmers with longer years of experience in extension may use their experience to using and taking the advantages obtained from new agricultural innovations or technologies and also they may develop, the confidence in handling the risk, skills in technology application, and may developed better economical status and better income from out put of using of these improved agricultural technologies.

Regarding, the distance of DAs office, from farmers home showed influential effect in adoption of improved bread wheat varieties in the study area as revealed in this study. The farmers who are nearby the DAs office are likely to adopt the improved bread wheat varieties than those who are far. This implies that the near by farmers to the DAs office would have an opportunity to get better and up dated information on the availability and benefit of improved varieties easily and better than those far farmers. As a result, they can use these opportunities to adopt the improved bread wheat varieties than those farmers far away from DA office.

The remaining hypothesized independent variables were not statistically significant to influence the probability of adoption of improved bread wheat varieties at less than 10% significant level as indicated in Table 38. Even though they were not significant below 10% significant probability level in logit model analysis practical and experience situations, literatures and many research works as well as the test statistics of this study showed that they have influential impact on adoption of improved technologies and innovations. The result of the logit analysis and their change or marginal effect of explanatory variables on dependent variable, probability of adoption of improved bread wheat varieties showed and presented in Table 38.

89

The marginal effect of the variable of the distance of DA-office closer to farmers home by one unit might improve the probability of adoption of improved bread wheat varieties by a factor of more than four times fold. In the social system the farmers leadership position can improve farmers agricultural technologies as observed in this study. In this study farmers leadership position improves the probability of adoption .As there is a change of farmers leadership position from non-leadership to leadership, there is an improvement of adoption of improved bread wheat varieties by the factors of nine times fold. The farmers experience in any of extension activities and use of improved technologies and innovations, can also improve the probability of adoption of improved bread wheat varieties. The changes and improvement of farmers experience in extension participation by one year or by one unit can increase adoption of improved bread wheat varieties by the factor of 61%.

In this study, it was revealed that the market access has a negative relationship with adoption of improved bread wheat varieties. When farmers are closer to market access by one unit, there is a decrease of probability of adoption of improved bread wheat varieties by a factor of 0.2 or by a factor of 20 percent. This implies that as mentioned in the above of this section, farmers who are closer to the market centers and facilities might be influenced and attracted by other substitution factors created by the market center facilities and might inclined to involve in these activities and business tasks with out totally leaving the farming occupation. As a result, they become reluctant to adopt improved bread wheat since improved wheat demand intensive management and labor work.

4.2.2. Analysis of determinants influencing intensity of adoption of improved bread wheat varieties and their marginal effects

Parameter estimates of the Tobit model for the intensity of adoption of improved bread wheat varieties (measured in terms of size of land in hectare used for growing of improved bread wheat varieties over the total wheat land in hectare). The Tobit model was used or applied to analyze the factors that determine the intensity of adoption of improved bread wheat varieties

90

because the mean proportion of land allocated to improved bread wheat varieties is a continuous variable but truncated between zero and one.

The main purpose of this section is to identify the hypothesized independent variables among the selected and proposed to include in the tobit model analysis that significantly influence the dependent variable, intensity of adoption. The result of this study indicated and presented in Table 39. From the total hypothesized independent variables, only eight explanatory variables were significantly influencing and affecting the intensity of adoption of improved bread wheat varieties as presented and indicated in Table 39. These significant variables were, households sex (HHHSEX), age (HHHAGE), education (EDUHHH), health status (HEALSTAT), offfarm income (HHOFFINC), home distance from DA office (DISDAOF1), farmland holding (SUMOWRE) and extension service (GEXSERVE) were statistically the most important explanatory variables affecting intensity of adoption of improved bread wheat varieties in the study area.

The variable household sex was related with the intensity of adoption of improved bread wheat varieties positively and significantly at 5 percent probability level. The sign was in line with that of the expectation. AS it was indicated in the identification of hypotheses, probability and intensity of adoption was expected to relate positively with male sample and negatively related with female sample. Hence, the positive sign indicates that the male-headed households were better in intensity of adoption of improved bread wheat varieties than female farmers. This result showed that male farmers are more likely to allocate larger farmland to improved wheat than female farmers in the study area. This result is in conformity with the finding of (Thechane, 2002).

91

Table 39.The effects of changes (marginal effect) in the significant explanatory variables on the intensity of adoption of improved bread wheat varieties

Variables HHHSEX HHHAGE EDUHHH HEALSTAT FAMILYSI HHOFFINC DISDAOF1 TOTLIVUN SUMOWRE CRINMFF1 OTSOLA1 OXTLU YEXPEXTS GEXSERVE

Coefficient 0.1854436505 0.3862778245 0.1696807312 0.3813275935 0.1544088566 0.1409564409 0.1684878586 -0.8907270032 -0.3801474933 0.1719597193 -0.2466170095 -0.3023853019 0.4834808429 0.2589730719

Standard Error

b/St.Er

P (/Z/>z) 0.0237 0.0132 0.0026 0.0000 0.1845 0.0395 0.0580 0.4131 0.0983 0.9858 0.7590 0.8911 0.3779 0.0196

0.81995580 2.262** 0.15594257 2.477** 0.56390468 3.009*** 0.72099682 5.289*** 0.11636011 1.327 0.68463253 2.059** 0.88875721 1.896* 0.10883167 0.22993869 0.96915193 0.80390554 0.22090082 0.54835151 0.11095968 -0.818 -1.653* 0.018 -0.307 -0.137 0.882 2.334**

(Source Computed from own survey data, 2005) *, ** And***Significant at 10,5 and 1 percent probability level

From these significant explanatory variables only one variable namely size of farmland holding related with the intensity of adoption of improved bread wheat varieties negatively and significantly at 10 percent probability level. This variable has the different sign from that was hypothesized. The remaining of the seven significant explanatory variables namely household sex, age, education, health status, off-farm income, extension service, distance of DA office from farmers home showed statistically significant and positively related with intensity of adoption of improved bread wheat varieties in the study area, at 10 percent probability level.

92

The variable household age was also related with the intensity of adoption of improved bread wheat varieties positively and significantly at 10 percent probability level. As the result of this study showed that older farmers may have already developed better experience, face exposure opportunities with using of large size of improved agricultural technologies through their life experience and might develop experiences how to manage risks and taking of the first benefits from newly released varieties. This might help them to develop confidences to allocate larger farmland to improved bread wheat varieties production more than those lesser and younker age group farmers in the study area. The sign was different from that of hypothesized. The hypothesis formulation and establishment was conducted based on literatures, experiences and observation of actual, practical and existing situations.

Literatures showed that farmers expected to be reluctant to new innovations as their age increased. But in the context of this study area, it is different from that of literatures. In history of Ethiopian extension farmers in the study area have better exposure of opportunities to new agricultural innovations than other areas of Ethiopia. As a result they developed better experience through their life experience better than other areas of farmers who do not get the opportunities like farmers in the study area. This finding agreed with the finding of (Chilot, 1994).

Education was also has a positive and significant relationship with the intensity of adoption of improved bread wheat varieties at 1 percent probability level. In this regard, the proportion of farmland used for growing of improved bread wheat varieties by farmers who are literate is likely to be greater than farmers who were illiterate. This suggests that being literate would improve access to information, capable to interpret the information, easily understand and analyze the situation better than illiterate farmers. So, farmer who are literate were likely to allocate larger size of farmland proportion than those illiterate farmers. The sign was as expected. This result has supported by other previous studies such as the findings of Lelissa (1998), Techane (2002), Lelissa and Mulate (2002), Yitayal (2004).

93

The variable health status had a positive and significant influence at 1 percent probability level relationship with the intensity of adoption of improved bread wheat varieties in the study area.. The result of this study showed that farmers who have better health status are likely to allocate larger farmland size to improved wheat varieties production. It is obvious from practical and actual situation of the ground that managing and operating the improved agricultural innovations demanded intensive labor and management practices. Then, health farmers can do these practices than unhealthy farmers. There fore healthy farmers are likely to allocate larger farmland size than unhealthy farmers. The sign was as expected. Low intensity of adoption by un-healthy farmers may be due to the shortage of labor and the problem to conduct intensive management that the improved bread wheat demanded.

The explanatory variable, off-farm income influenced the dependent variable, intensity of adoption of improved bread wheat varieties positively and significantly at 5 percent probability level as hypothesized under the section of hypotheses description. As it is true that farmers who have better income can adopt new agricultural innovations because their income allowed them to purchase the new technological inputs, can with stand risks if appear and can cover labor costs. Off-farm income is one of the alternatives to improve farmers income. From these grounds of realities farmers who have off-farm income can adopt new technologies in larger proportion than those who do not have off-farm income.

The size of farmland holding, affected the dependent variable, intensity of adoption of improved bread wheat varieties in the study area negatively and significantly at 10 percent probability level. The sign was different from that of postulated. This finding is in conformity with the finding of (Bekele et al, 2000; and Chilot, 1994).

As it is supported by many literatures, those farmers who have larger land size are expected to adopt improved and new agricultural technologies in larger proportions than those farmers who have lesser farmland. Since these farmers have larger farm land they do not have fear of risks, can get credit because they are believed that they can pay their credit, or some part of their land may serve as mortgage to take credit, seed loan from other farmers and can adopt

94

new agricultural technologies than those who have lesser land size. But in this study the result showed the different situations from many literatures even though it is in line with some few literatures as mentioned in the above. The possible reason for this result can be that the actual situations in the study area different from other areas in that the situations of the area where this study has conducted is closer to Addis Ababa, the capital of the country where in the mean time of this study highest construction investment on land were conducted.

Farmers who are closer to the town lease their part of farm land and might reluctant to increase their farm land allocation to improved bread wheat land since they got better income from land contract than they got from improved wheat land production. Another possible reason also for this situation might be that people at the edge boarder of Addis Ababa and the rural part of the study area have the experience of producing crop for their consumption and profit purpose by contracting land from the surrounding farming community. As result of the existence of this situations in the study area farmers might contract their land for these types of par time farmers due to many situations like for better income than they used for improved wheat production or for the reason they may face different problems and cash constraint that could not give some time in the future. Then, those par time farmers might have their own interest of crop type production and objectives. As a result those first owners and adopters of improved bread wheat varieties might unable to increase their farmland allocation for improved bread wheat varieties.

And also in the study area there is an introduction and promotion of white check pea, which has high price in market. This also shift the wheat adopters to allocate their wheat farm land to this new crop variety rather than increasing of their land allocation to the improved bread wheat varieties since wheat land can use for check pea crop interchangeably. It was also observed that credit and input providers greatly reduced their service provisions due to the reluctant effect of farmers to return their previous credit loan as a basic reason of the highest interest rate of the loan. These are some of the possible reasons for the inverse relationship of the independent and dependent variables in this case.

95

Extension service influenced dependent variable, intensity of adoption of improved bread wheat varieties positively and significantly at 5 percent probability level as hypothesized. The finding of this study is in agreement with the findings of Adesina and Zinnah (1992), Chilot (1994), Techane (2002), Lelissa and Mulate (2002) and Yitayal (2004). Theoretical and practical realities showed that extension services provided to the farmers in different forms like training demonstration field day DA visit on the field and on spot field support can motivate, empowers kill and knowledge, increase information access and create interest to improve farmers use of improved agricultural technologies.

The independent variable, distance of DA office from the farmers home, influenced the independent variable, intensity of adoption of improved bread wheat varieties in the study area positively and significantly at 10 percent probability level. The sign was as postulated. The intensity of adoption of improved bread wheat varieties is higher to the farmers who are closer to the DA office than those farmers who found far. This result is in line with the result of (Chilot, 1994). This is also true from theoretical, practical and experience realities when the Das assigned closer to the farmers village farmers can easily and from near by distance can get the required information such as availability of inputs, credit services, market situation government and other development organization supports on time and sufficiently.

The results of the Tobit model analysis also showed the effects of changes or marginal effects in the explanatory variables on the dependent variable, intensity of adoption of improved bread wheat varieties, in the study area as indicated and presented in Table 39.

Literatures showed that adoption and intensity of utilization of improved agricultural innovations has relations with gender. As it was hypothesized male farmers were likely expected to show better intensity of farmland allocation for improved bread wheat production than female farmers. The marginal effect of Tobit model analysis showed that male farmers were better in allocation of farm land as compared to female farmers. The intensity of farmland size allocation for improved bread wheat varieties production by male farmers was larger by a factor of 19 % than female farmers.

96

Age is one factor to influence intensity of adoption. The marginal effect of tobit analysis showed that as age of adopters of improved bread wheat increase by one unit, intensity of farmland size allocation for improved bread wheat varieties production can improve by the factor of 38.63%. As mentioned in the above, since farmers in this area have a long years exposure to extension services than other areas of farmers, they developed better extension experience in the process of their lifetime experience that plays a great role in intensity of adoption of improved bread wheat varieties in the study area.

Education plays a positive and significant role in the intensity of adoption of improved bread wheat varieties in this study. An improvement in education or a change by a unit i.e. from illiterate to literate can improve farmers allocation of farm land for improved bread wheat production from their total wheat land can improved by a factor of 17%. In the other way, as there is an improvement in educational level of adopters of improved bread wheat varieties by one unit, there will be an increased allocation of farm land for production of improved bread wheat varieties by 17%.

In this study it was identified that as farmers health situations improved from unhealthy to healthy situation, the intensity of adoption of improved bread wheat varieties can increases by a factor of 38%, because the health farmers can directly involve in every activities of improved bread wheat production and can them selves manage their farm. As a result the allocation of intensity of farmland for improved bread wheat production can increase by a factor of 38%.

The variable household farmers off-farm income contributes its own part in the intensity of adoption of improved bread wheat varieties in the study area positively and significantly. As the involvement of farmers in off-farm income and consequently their income improved by one unit, their allocation of farmland for improved bread wheat production can increase by the factor of 14%.

97

When the farmers income improves by one unit from off-farm income source, the allocation of farmland by them for improved bread wheat production can improve by a factor of 14%. This is due to the fact that the off-farm income can solve farmers financial constraints and increase their purchasing power of improved bread wheat seed, other agricultural inputs such as fertilizer and other production means relevant to the production of improved bread wheat. Consequently, farmers might be encouraged in allocating larger area of their wheat farmland for the production of improved bread wheat varieties than those who do not have off-farm income.

The distance of DA office plays its role in intensity of adoption of improved bread wheat varieties in the study area positively and significantly. As the distance of DA office decreases or closer to the home of farmers by one unit, intensity of adoption could increase by a factor of 17%. This implies that farmers who are closer to the DAs office can get easy access to extension support and agricultural information that can give a chance to analyze situations and allocate their larger farmland for growing of improved bread wheat varieties than those who are far from DAs office.

As discussed in the above, the independent explanatory variable, farmland holding related with the intensity of adoption of improved bred wheat varieties negatively and significantly as indicated and presented in Table 39. As revealed in this study, when the size of the farm land holding of farmers increased by one unit intensity of adoption of improved bread wheat varieties decreased by a factor of 38%. This may be due to the fact substitution of some part of their wheat land to other highly market demanded crops like for example white chick pea production, land contracting by receiving larger amount of money, and the farmers themselves might involve in other activities and reluctant to allocate increased farm land for improved bread wheat production as a result the result of tobit model analysis showed the inverse relationship between farm land and intensity of adoption of improved bread wheat varieties in the study area.

98

As it is generally true, the better the extension service can improve the intensity of adoption and utilization of improved agricultural technologies. The tobit analysis in this study showed that the change in improving the extension service by one unit can improve the allocation of farm land for improved bread wheat production by adopter farmers can improve by the factor of 26% as indicated in Table 39. This implies that when farmers get support from extension agent, in various forms such as information provision, practical support on the spot of the field or in the form of demonstration, field day and skill development, can improve farmers knowledge, interest, motivation and confidence to allocate larger extent of farm land than those who do not get or who got less extension support.

To summarize the two analytical model results, that the purpose of data analyses using the econometrics models as discussed through out this section and in the previous sections is to know which independent variable most important and powerful to affect the intended dependent variable to which they hypothesized to influence. In this study two-econometrics models logit for identification of factors affecting probability of adoption of improved bread wheat varieties for those non-adopter farmers need to adopt in the future and tobit model for estimation of factors to influence adopters intensity of adoption or allocation of farm land size intensity for improved bread wheat production.

As mentioned at the beginning of this analytical section due to various reasons such as theoretical, actual, practical, technical reasons some hypothesized independent variables were dropped from further analyses like for example leadership position and credit service due to tobit model limitation to accommodate all independent variables included in Logit model analysis did not included in Tobit model analysis. The result of logit model analysis for probability of adoption and result of tobit model analysis for intensity of adoption of improved bread wheat varieties indicated and presented in Tables 38 and 39 respectively.

The most important and significant independent variables below 10% probability level to influence probability of adoption of farmers who did not adopt improved bread wheat varieties in the past but expected to adopt in the future, those identified by logit model analysis were

99

four namely distance of DA-office from the farmers home, house hold head social/leadership position, market accessibility, and house hold farmers experience in extension as indicated in Table 38. Regarding the result of tobit model analysis as indicated in Table 39, eight independent variables namely sex of household head, age, education, health status, off-farm income, distance of farmers home from DA office, farmland holding and extension service were statistically significant below 10% probability level and most important to influence adopters intensity of farm land allocation to improved bread wheat varieties production from their total wheat farm land as indicated in Table 39.

In this study, the two models used for two purposes as mentioned in the above. Though the purpose of this study is not to identify the significant common independent variables among those variables used in the two models analyses to identify influencing significant factors for the two dependent variables as mentioned in the above, it is very important to see whether there is a common influencing independent variables that affect significantly the two mentioned dependent variables. As a result, it was identified that there was only one independent variable, distance of DA-office from farmers home that commonly and significantly affected both probability of adoption and intensity of adoption below 10% probability of significant level. It doesnt mean that the remaining independent variables totally do not have any relationship with the dependent variables rather they are not statistically significant below 10% significant level.

100

5. SUMMARY AND CONCLUSION

5.1. Summary

In this study to identify factors influencing probability and intensity of adoption of improved bread wheat varieties among smallholder farmers the study area, Akaki was selected based on its wide practices of improved bread wheat production and its suitability for this research .In this area, agricultural extension and rural development activities like other rural parts of the country, conducted by Agricultural and Rural Development Unit which comprises several agricultural professionals in different disciplines at office level and Development Agents (DAs) at Center and Peasant Association (PA) level. According to the structural framework of the Addis Ababa Administration, the unit is accounted to and organized under Akaki-kality sub-city.

In this study, data were obtained from 150 randomly selected respondents through personal interview schedule conducted by employed and trained enumerators using pre-tested interview schedule and from group and individual discussions, as well as the researchers personal observations. The respondents, involved in the interview were selected randomly and proportionally from two sample Peasant Associations (PAs), constituted 99 (66%) adopters and 51 (34%) non- adopters.

Data were analyzed, and presented quantitatively using different statistical methods such as percentage, frequency, tabulation, Chi-squaretest (for dummy /discrete variables) and (t-test for continuous variables), Logit, Tobit models and qualitatively through interpretation, explaining, summarizing of ideas and concepts. T-test and Chi-square test were employed to test the variation of the sample group they have towards adoption and also used to describe the patterns of the sample data. Logit and Tobit econometrics models to estimate the effects of hypothesized independent variables they have on dependent variables, probability and

101

intensity of adoption. Computer soft ware package programs such as SPSS and Limdep were employed for statistical analyses.

Among the hypothesized independent variables, sex, health, education, extension service, DAoffice, market access, labor source, off-farm income, leadership, were treated as discrete variables and tested using chi-square-test. In this test the independent variable health, distance of DA-office, labor source, access to off-farm income were not significant below 10% significant level. And family size, years of extension experience, age, livestock ownership, oxen ownership, and farmland holding were considered as continuous variables and tested using t-test. The t-test result showed that except others only age was not significant below 10% significant level.

The t-test and chi-square test results showed that there were variations between adopters and non-adopters sample category in family size, extension experience, livestock ownership, oxen ownership, farm land holding, extension service, sex (gender) and leadership position in adoption of improved bread wheat varieties. According to the result of test statistics male are better in adoption of improved bread wheat varieties. On the other hand adopters have larger family size, livestock ownership, oxen ownership and farmland and they got better extension service than non-adopters. Due to their variation in these independent variables sample farmers vary in their adoption behavior in relation to dependent variables.

Except those hypothesized independent variables dropped due to various cases as mentioned in previous section all screened and verified independent variables were subjected to Logit model analysis. In the case of Tobit model analysis, all verified hypothesized independent variables included in Logit model analysis were not included due to the limitation of the model to accommodate all these independent variables. As a result, leadership position and credit services were dropped from further Tobit model analysis due to their less importance in the study as compared to other independent variables.

102

The Logit model result of this study showed that the significant independent variables affecting probability of adoption were distance of DA office, leadership position of household head, market access and years of house hold heads experience in extension and those independent variables significantly influencing intensity of adoption of improved bread wheat varieties were, household heads sex, age, education, health status, off-farm income, distance of DA office, size of farmland holding, and extension service resulted from tobit analysis. The distance of DAs office from farmers home was the only explanatory variable influencing both adoption and intensity of adoption of improved bread wheat varieties in this study.

The farmers selection and evaluation criteria of improved bread wheat varieties and ranking of the improved bread wheat varieties disseminated in the study area were also conducted through the summary of the survey data, group and individual discussions as well as researchers observation.

In this respect, white grain color, large grain size, straw quality, market demand were the first most important characteristics; germination capacity, cooking quality good yield performance were the second most important; water lodging resistance, tillering capacity, good food quality were the third most important; short maturity date fourth; disease and pest resistance fifth; frost resistance sixth; harvesting quality and storage quality ranks seventh most important characteristics as grouped and ranked based on the result of the survey data group and individual discussions and researchers observation.

Based on the selection and evaluation criteria, the result of the survey summary and group discussion the ranking result of improved bread wheat varieties disseminated in the study area has presented as HAR-1685 variety ranks firs, Paven 76 second and HAR-1709 ranks third.

Farmers in the study area got agricultural information from different sources. The most important information sources as summarized were, neighbors and colleague farmers got the 1st rank, DA and Community leaders the 2nd rank, farmers field day 3rd, PA leaders,

103

demonstration and radio 4th, News paper/News letter and other publications 5th, poster 6th, Training 7th, TV8th and religious leaders ranks 9th sources of information.

104

5.2. Conclusion and Recommendations

In this study several issues were observed and revealed in relation to adoption of improved bread wheat varieties disseminated in the study area, Akaki. The result, description and interpretation of the data were mainly depended on, the context of the research objectives and the situation of the study area.

The truthfulness of the information provided by the sample farmers for this study was also depended on the sample farmers voluntaries and credibility. Since the study area closer to Debrezeit research center, Addis Ababa (the capital of the nation) and subjected to long years of extension services in the past, the result of the study should be seen from this perspectives. This study may serve as an initial input for further study in this and other similar areas of the country.

Like other parts of the country, several agricultural innovations were disseminated in the previous years and extension services were offered to the farmers that have an influential impact on adoption and use of the disseminated agricultural innovations. From those disseminated technologies in this area, improved bread wheat varieties was the one on which this study was focused to identify factors affecting adoption of improved bread wheat varieties by non-adopter farmers, and to identifying other factors influencing the adopter farmers to increase the intensity of farm land size allocation to improved bread wheat production from their total wheat farm land.

Determinants that limit probability of adoption of improved bread wheat varieties were identified using descriptive statistics (t-test and Chi-square test) and logit model analysis. They were gender, extension service; leadership position, market access, farmers extension experience and distance of DA office from farmers home were the influencing factors affecting non-adopters to adopt improved bread wheat varieties in the study area.

105

According to the findings of this research, it is necessary to establish appropriate extension strategy to bring those non-adopters to adopt improved bread wheat varieties. In this regard, attention should be given; to encourage, support and motivate female and less extension experienced farmers to achieve their adoption decision behavior. Some farmers in this area

who have leadership position are better adopters since their position allowed for better access to information, resources and innovations. Therefore, there need to give attention to support those people who do not have resources access opportunities.

As it is confirmed in this study distance of DA office from the farmers home has an influential effect on adoption and intensity of adoption. Therefore, attention should be given to the close assignment and placement of DAs to the rural villages where the farmers can get them easily for extension advises and supports for better adoption.

In the study area, when farmers closer to the market showed reluctant behavior to adopt the improved bread wheat varieties. Some of the possible reasons may be due to weak extension service provided for them or due to fear of intensive management and labor requirement to operate practices, or may be due to substitution effect and their involvement in other par time works etc., created by the market facilities. The market in this area showed a negative effect rather than motivation farmers to adopt improved bread wheat varieties. Therefore, there must be efforts to formulate appropriate extension service for this area, improve market situation for bread wheat and improvements of the varieties qualities for better market demand.

The other aspect of this study was to identify factors influencing those adopters to increase and extend their improved wheat production by allocating larger area of farm land for improved bread wheat production from their total wheat farm land. According to the result of descriptive statistics and tobit model analysis gender, extension service, family size, experience in extension, livestock and oxen ownership, farm land holding, age, education, health, off-farm income, distance of DA-office were factors affecting intensity of adoption of improved bread wheat varieties. Attention should be given to improve farmers intensity of adoption by

106

designing of compatible extension strategy by considering the findings of this research as an input.

Regarding farmers variety selection and evaluation criteria, it is advisable to involve farmers through various techniques like for example using group discussion to evaluate and identify the best suitable varieties that can fit their interest, farming system and environmental situations. In-group discussion farmers from different angels such as gender, age, ecological area and educational levels should be involved to get various ideas and opinions. The idea reflected during group discussions should get attention and need to be incorporated and used in agricultural technologies development, extension programs formulation and policy preparations.

In the study area there is a shift of farmers to involve in improved chickpea production as a result of high price of improved chickpea. There fore, it is necessary to give attention to improve the quality of improved bread wheat varieties that can bring high market demand through breeding and genetics improvement programs. It is also necessary to improve the market facilities for improved bread wheat varieties.

Agricultural information and extension communication are powerful and crucial to achieve better adoption and intensity of adoption of improved agricultural innovations like improved bread wheat varieties in this case. Appropriate and timely information should reach to the intended farmers group to achieve better adoption and intensity of adoption of improved agricultural technologies. Appropriate information and communication strategy compatible with farmers and the study area should be designed and practiced.

Suitable strategies for better extension service are another important issue that should get proper attention. In the study area as observed and the survey data showed, the extension service is at lower and weak position due to various reasons such as transfer of DAs to other lateral offices, low motivation, poor credit service low educational background of extension

107

workers. In these respect it need attentions to solve these problems for better improvements of agricultural technologies adoption and production growths that can bring better living standard of the farmers in the rural areas. Attention also should be given to the research and extension linkages, to the empowerment and training of extension people and farmers, to achieve high level of improvement in adoption of improved agricultural technologies.

108

6. REFERENCES

Adam Bekele and Bedru Beshir, 2005.Adopting Improved Haricot Bean Varieties in the Central Rift Valley of Ethiopia. Ethiopian Agricultural Research Organization (EARO), Addis Ababa, Ethiopia Adams, M.E .1982.Agricultural Extension in Developing Countries, Longman Scientific and Technical Publications Xii, 108p. Adane Nabso Fufa.2002.Magnitude and determinants of rural household poverty in central Ethiopia.The case of Bereh-Aleltu district M.Sc Thesis in Agricultural Economics. Alemaya University, Ethiopia. Adane Nigus Mekonnen.2002.Economic analysis of adoption of soil conservation technologies:A case study of Kobo woreda in North Wollo, M.Sc. Thesis in Agricultural Economics. Alemaya University, Ethiopia. AdesinaA.AandZinnah, M.M.1993.Technology Characteristics, Farmers Perceptions and Adoption Decisions; A Tobit model application in Sierra Leone, Elssevier Science Publishers B.V., Amsterdam,U.S.A. Agricultural Economics.9(1993):297-311p. Adugna Haile, Workineh Negatu and Bisrat REta., 1991.Technology transfer for wheat production in Ethiopia, In Hilu Gebremariam, Tanner Doughlas G. and Mengistu Hulluka (eds). Wheat Research in Ethiopia: A Historical Perspective, Addis Ababa, and IAR/CIMMYT. Amemiya, T.1985.Advanced Econometrics, T.J.Press, Padstaw Ltd. Great Britain. Anderson, J.R. and Feeder, G.2002.The Rural Extension Services, World Bank, Agricultural Development Department, Washington D.C. Arnon, I., 1989. Agricultural Research and Technology Transfer, Elsevier Science Publishers ltd.,London and New York. Asfaw Negassa, Kisa Gunjal, Wilfred Mwangi and Beyene Seboka.1997.Factors Affecting the Adoption of Maize Production Technologies in Bako Area, Ethiopia, Ethiopian Journal of Agricultural of Economics 1(2) 52-72p. Bekele Hundie, Kotu hugo Verkuijl, Wilfred Mwangi and Douglas Tanner, 2000. Adoption of Improved Wheat Technologies in Adaba and Dodola Woredas of the Bale High Lands,

109

Ethiopia, Mexico D.F.:International Maize and Wheat Improvement Center (CIMMYT) and Ethiopian Agricultural Research Organization (EARO). Bisanda Shekania and Wilfred Mwangi.1996.Adoption of Recommended Maize Technologies, in Mbeya Region of the Southern Highlands of Tanzania, CIMMYT/ the United Republic of Tanzania, Ministry of Agriculture. Chandan J.S., 1998. Statistics for Business and Economics, Vikas Publishing house Pvtt Ltd., New Delhi. Chilot Yirga., 1994 .Factors Influencing Adoption of New Wheat Technologies in the Wolemera and Addis Alem Areas of Ethiopia, M. Sc. Thesis, Alemaya University. CIMMYT.1993.The Adoption of Agricultural Technologies: A Guide to Survey Design, Mexico, D.F.: CIMMYT. Curtis B.C. 2002.Wheat in the World. In Curtis Rajarm B.C.S. and Gomez H. McPherson, (Edits), Improvement and Production, FAO, Rome. Destaw Berhanu Nega.2003.Non-farm employment and farm production of smallholder farmers study, in Edja district of Ethiopia., M.Sc. Thesis, in Agricultural Economics, Alemaya University, Ethiopia. EARO.2004.Agricultural technology evaluation, adoption and marketing, Part 2.In: Tesfaye Zegeye,Legesse Dadi and Dawit Alemu .Proceeding of the Workshop held to discuss on The Socioeconomic Research Results of Farmers Participatory Research; Attempts and Achievement in the Central Highlands .1998-2002 August 6-8,2002,Addis Ababa, Ethiopia. Endrias Geta.2003.Adoption of improved sweet potato varieties in Boloso Sore woreda, southern Ethiopia, M.Sc. Thesis, Alemaya University, Ethiopia. ERSHA (Ethiopian Rural Self Help Association).2000.Evaluation of Bread Wheat Technologies on the Farmers Farm Condition Using Farmers Criteria in West Shoa Zone, Ambo Woreda Birbisa and Cherech Service- Cooperative, ERSHA Addis Ababa, Ethiopia. Feder, G. L. Just R.E. and Zilbernran D.1985.Adoption of Agricultural Innovation in Developing Countries; A Survey Economic Development and Cultural Change 32(2): 255 298p. FAO . 1993. Ethiopia. In: FAO:The State of Food and Agriculture .Rome. Franzel Steven. 1992. Features of smallholder farming systems. In Franzel and Helen van houten (edts). Research with Farmers: Lesson from Ethiopia. C.A.B International, UK.

110

Freund, John E.1967. Modern Elementary Statistics, third edition, prentice-hall, Inc.,Engle Wood Cliffs ,New Jersey. Getachew Agegnehu, Brhane Lakew and Kassa Getu.2002.On Farm evaluation of bread wheat varieties at Ginchi Watershed Site, Towards Farmers Participatory Research, at Central High lands of Ethiopia. Proceedings of Client Oriented Research Evaluation Workshop, 1618october2001, Holeta Ethiopia, EARO, Addis Ababa, Ethiopia. Gujarati D.N.1995. Basic Econometrics.3rd (Ed), Mc Graw-Hill.

Gujarati.D.N.1999.Essencial of Econometrics, Mc Graw Hill Company, Singapore, 220p Gujarati, D.N.2003.Basic Economics .4th(ed), McGraw Hill,New York. Hanson H., Noman E.B.and R.G. Anderson.1982.Wheat in the Third World. International West View Maize and Wheat Improvement Center, Press;Boulder, Colorado,USA. Harlan R.J. 1981.The early history of wheat earliest traces to the sack of Rome, In L.T. Evans and W.J. Feacu UK (edits), Wheat Science: Today and Tomorrow, Cambridge University press, Cambridge. Jha. Dayanatha, Behjat Hajjati, and Stephen Vosti.1991.The Use of improved agricultural technologies in eastern province of Zambia, In Celis Rafael, John T. Milimo and Sudlir wanmali, Adopting Improved Farm Technology, A Study of Small Holder Farmers in Eastern Province of Zambia, Rural Development Studies Bureau (University of Zambia), National Food and Nutrition Commission, Eastern Province Agricultural Development Project (Government of the Republic of Zambia), International Food Policy Research Institute (IFPRI), Washington, D.C. Kelejian, H..H..and W.Outes,1981.An introduction to economic analysis.2nd (ed) Horpeer and Row Publishers. Kiflu Bedane and Brhanu Kuma 2002.Farmer participatory research. An Overview .In Gemechu Keneni,Yohannes Gojjam,Kiflu Bedane,Chilot Yirga and Asgelil Dibabe (eds).Towards Farmers Participatory Research ,Central High lands of Ethiopia. Proceedings of Client Oriented Research Evaluation Workshop, 16-18october2001, Holeta Ethiopia, EARO, Addis Ababa, Ethiopia. Langer M.R.H. and Hill G.D.1982. Agricultural Plants, Cambridge University Press, Cambridge.

111

Legesse Dadi.1992.Analysis of Factors Influencing Adoption and the Impact of Wheat and Maize Technologies, In Arsi Negele, Ethiopia, M. Sc Thesis, Alemaya University, Ethiopia. Largesse Daddi .1998.Adoption and Diffusion of Agricultural Technologies: The Case of East and West Shoa Zones, Ethiopia, A Thesis Submitted to the University of Manchester for the degree of Doctor of Philosophy in the Faculty of Economics and Social Studies, School of Economic Studies. Largesse Daddi, Senait Regassa,Asnake Fikre and Demissie Mitiku .2005..Adoption of Chick Pea Varieties in the Central High Lands of Ethiopia, Ethiopian Agricultural Research Organization (EARO), Addis Ababa, Ethiopia. Lelissa Chalchissa and Mulat Demeke.2002.The Determinants of Adoption and Intensity of Fertilizer Use in Ejera District, West Shoa zone, Ethiopia , Institute of Development Research (IDR), Addis Ababa. Lucila,Ma.,A.Lapar and Sushil Pandey.1999.Adoption of soil Conservation :The case of the Philippine Uplands .The Journal of International Association of Agricultural Economics 21(3) 241-256. Maddala, G.S.1992.Introduction to Econometrics, second edition, Macmillan publishing company, New York. Maddala, G.S.1977. Econometrics, Singapore, Mcgrow-hill Book Company. Mergia Beyene .2002 Farmers participatory on farm research: An Alternative approach to agricultural technology promotion. In Gemechu Keneni,Yohannes Gojjam, Kiflu Bedane, Chilot Yirga and Asgelil Dibabe (eds.):Towards Farmers' Participatory Research: Attempts and Achievements in the Central Highlands of Ethiopia . Proceedings of Client-Oriented Research Evaluation Workshop, 16-18 Octobert2001, Holetta Agricultural Research Center, Holetta ,Ethiopia. MillionTadesse and Belay Kassa.2004.Determinants of Fertilizer use in Gununo area, Ethiopian Agricultural Research Organization (EARO), Addis Ababa, Ethiopia. Mullugetta Mekuria.1994.An Economic Analysis of The Smallholder Wheat Production and Technology Adoption in the Southeastern High lands of Ethiopia Ph.D Thesis, Department of Agricultural Economics Michigan State of University USA. Mussei,A., J. Mwanga, W.Mwangi, H. Verkuijl, R. Mongi, and A. Elanga. 2001.Adoption of Improved Wheat Technologies by Small Scale Farmers in Mbeya District, Southern Highlands of Tanzania, Mexico, D.F., International maize and wheat Improvement center (CIMMYTY) and the united Republic of Tanzania.

112

Nanyeenya,William Ntege,Mary Mugisa-Mutetikka, Wilfred Mwangi, and Hugo Verkuijl.1997.An Assessement of Factors Affecting Adoption of Maize Technologies in Iganga District ,Uganda.National Agricultural Reserch Organization/CIMMYT. Pearson C. Lorentz, .1967.Principles of Agronomy, Reinhold, New York. Rogers 1971.Rogers Everett M with F.Floyd Shoemaker.1971.Communication of Innovations; A Cross Cultural Approach, 2nd edition,The Free Press ,New York, USA. Rogers, Everett. M. 1983.Diffusion of Innovations: 3rd Edition .The Free Press, New York, USA. Sarantakos. 1998. Social Research, second edition. Macmillan Press Ltd. London Sherif Aliy Geda. 2001. The Social Nature of Agro-Technological Change: The Trajectory of the BBM Technology in the Joint Vertisol Project in Ethiopia, M. Sc Thesis, Wageningen University and Research Center. Solasya, B.D.S. Mwangi, H. Verkuijl, M.A. Odendo, and J.O. Odenya.1998.An Assessment of the Adoption of Seed and Fertilizer Packages and the Role of Credit in Small Holder Maize Production in Kakamaga and Viliga Districts, Kenya Tanner D. and R. Raemaekers.2001.Wheat: Triticum Spp., In Romain H.Raemalkers (edit.) Crop production in tropical Africa Directorate General for International Cooperation, Brussels, Belgium Techane Adugna Wakjira.2002.Determinants of Fertilizer Adoption in Ethiopia: The Case of major cereal production areas. M.Sc thesis, Alemaya University, Agricultural Economics Tesfaye Beshah .2003.Understanding Farmers: Explain soil and water conservation in Konso, Wollaita, and Wello, Ethiopia, Ph.D. Thesis, Wagneningen University and Research Center. Tesfaye Zegeye ,Bedassa Tadesse and Shiferaw Tesfaye.2001.Adoption of High Yielding Maize Technologies in Major Maize Growing Regions of Ethiopia ,research report no.41.,Ethiopian Agricultural Research Organization (EARO),Addis Ababa, Ethiopia. Tesfaye Zegeye and Alemu Haileye,. 2001. Adoption of Improved Maize Technologies and Inorganic Fertilizer in Northern Ethiopia: Research report no. 40. EARO, Addis Ababa, Ethiopia Tesfaye Zegeye, 2004.Adoption of Inorganic Fertilizer on Maize in Amhara, Oromiya and Southern Regions. In Abebe Kirub (edt.), Agricultural technology evaluation, adoption and marketing Part 2.EARO, Addis Ababa, Ethiopia

113

Vijayaragavan, K. and Singh, Y.P.1997. Managing Human Resources Within Extension .In Swanson, B.E.,Bentz,R.P. and Sofranko,A..J. (Eds.): Improving agricultural extension: A Reference manual, Rome: FAO, p.p.127-134. Yitayal Anley Mengistu.2004.Determinants of Use of Soil Conservation Measures by Small Holders Jimma zone: the case of Dedo District, M.Sc Thesis Alemaya University, Ethiopia.

7. APPENDICES

114

Appendix.1.Information on sample household demographic and socio-economic characteristics

Table 1.The distribution of sample respondents by age group Age (year) 19-30 31-45 46-64 Above or Equal to 65 Total N 11 41 34 13 99 Adopters % 11.11 41.45 34.34 13.13 100 N-adopters N 11 16 15 9 51 % 21.57 31.37 29.41 17.65 100 Total N 22 57 49 22 150 % 14.67 38 32.67 14.67 100

(Source: Computed from own survey data, 2005)

Table 2.Educational statuses of sample house hold head farmers Educational Status Illiterate Read & Write Grade1-6 Grade7-8 Grade9-12 Above grade 12 Total N 64 19 4 5 3 4 99 Adopters % 64.65 19.20 4.04 5.05 3.03 4.04 100 Non-adopters N % 33 64.71 14 27.45 4 7.84 51 100 Total N 97 33 8 5 3 4 150 % 64.67 22 5.33 3.33 2 2.67 100

(Source: computed from own survey data, 2005)

115

Table 3 .The sample household family size Family Size 1 2 3 4 5 6 7 8 9 10 11 Total Adopters N 2 2 8 17 17 18 13 11 5 2 4 99 % 2.02 2.02 8.08 17.17 17.17 18.18 13.13 11.11 5.05 2.02 4.04 100 N-Adopters N 4 11 10 5 6 5 4 5 1 51 % 7.84 21.60 19.61 9.80 11.76 9.80 7.84 9.80 1.95 100 N 6 2 19 27 22 24 18 15 10 3 4 150 Total % 4 1.33 12.67 18 14.66 16 12 10 6.67 2 2.67 100 Total Family Members N % 6 0.72 4 0.48 57 6.80 108 12.87 110 144 126 120 90 30 44 839 13.11 17.16 15.02 14.30 10.73 3.57 5.24 100

(Source: Computed from own survey data, 2005) Table 4.The sample household family size Family Size Maximum Minimum Range Average St.d. Adopters N 11 1 10 5.85 2.192 Non-Adopters N 10 1 9 5.10 2.385 Total N 11 1 10 5.6 2.27

(Source: Computed from own survey data, 2005)

116

Table 5.Total Family members of sample households in age group Age- group 0-14 age 15-64 age Above 64 N 361 452 26 % 43.03 53.87 3.10 100

Total 839 (Source: Computed from own survey data, 2005) Table 6.Respondents farming experience Age group (Year) 1-10 11-20 21-30 Above 30 Total Adopters N % 17 44 21 17 99 17.172 44.444 21.212 17.172 100 Non-adopters N % 9 25 12 5 51 17.65 49.02 23.53 9.80 100

Total N 26 69 33 22 150 % 17.33 46 22 14.67 100

(Source: Computed from own survey data, 2005) Table 7.Types of livestock and owners and the number of respondents Types of Livestock Oxen Cow Bull Heifer Calves Sheep Goat Horse Mule Donkey Poultry Bee-in-Hive Adopters N % 98 68 68.69 46 46.46 41 41.41 39 39.40 48 48.48 8 8.08 21 21.21 25 91 80 5 25.25 91.92 80.81 5.05 Non-adopters N % 42 33 64.71 19 37.25 14 27.45 20 39.22 29 56.86 4 7.84 6 38 29 1 11.76 74.51 56.86 1.96 Total N 137 101 65 55 59 77 8 20 31 129 109 6

% 91.33 67.30 43.30 36.70 37.70 51.33 5.33 16.67 20.67 86 72.67 4

(Source: Computed from Owen survey data, 2005)

117

Table 8.Sample house hold oxen ownership Number Of oxen No oxen One ox Two oxen Three oxen Four oxen Five& above Total Adopters N % 1 3 3.03 46 7 30 10 99 46.47 7.07 30.30 10.10 100 Non-adopters N % 9 3 5.86 25 3 6 4 51 49.02 5.88 11.77 7.84 100 Total N 10 6 71 10 36 14 150

% 4 47.33 6.67 24 9.33 100

(Source: computed from own survey data) Table 9.Sample house hold land ownership Types of Land ownership I. Own Land owners Cultivated Land Grazing land Home stead land Forest land Un-used land II .Shared/Rent land owners Cultivated land Grazing land III Growers of variety HAR-1685 variety HAR-1709 variety Paven-76 variety Adopters N 99 52 18 4 4 20 2 83 5 75 % 100 52.52 18.18 4.04 4.04 20.20 2.02 83.84 5.05 75.76 Non-adopters N 51 29 14 2 3 4 3 1 % 100 56.86 27.45 3.92 5.90 7.84 5.88 1.96 N 150 81 32 6 7 24 2 86 5 76 Total % 100 54 21.33 4 4.67 16 1.33 57.33 3.33 50.67

(Source: Computed from own survey data, 2005)

118

Appendix Table 10.Size of farmland holding of sample household

Farm Size In Ha. <1Ha 1-2Ha 2-3Ha 3-4Ha 4-5Ha 5-6Ha Total

N 12 32 30 18 5 2 99

Adopters % 12.12 32.32 30.30 18.20 5.05 2.01 100

Non-adopters N % 15 29.41 14 27.45 15 29.41 6 11.76 1 1.96 51 100

Total N 27 46 45 24 6 2 150 % 18 30.67 30 16 4 1.33 100

(Source: Computed from own survey data, 2005) Table 11.Respondents average land area and yield of wheat crops in 1996/97 E.C. cropping season. Item/List Area of land in Hectares Total farm Land Total wheat land Total improved B.W.L. Area of Paven -76 Area of HAR-1685 Area of HAR-1709 Area of Durum Wheat Area of Local Wheat Yield of wheat in Quintals Yield of Paven-76 -Yield of HAR-1685 Yield of HAR-1709 Yield of Durum wheat Yield of Local wheat Max 6.5 4.5 3.25 1.75 2.00 2.00 1.00 2.00 25 28 8 15 26 Min 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.50 3 .00 4.00 4.00 2.00 Range 6.25 4.25 3 .00 1.50 1.75 1.75 0.75 1.75 24.5 25 4 11 24. Average 2.44 1.125 1 .00 6.30 0.64 0.535 0.415 0.65 9.12 9.33 5.5 9.165 8.255 St.D 1.13 0.605 0.457 0.29 0.34 0.44 0.285 0.41 4.67 4.445 1.91 7.03 5.365

(Source: Computed from own survey data, 2005) N.B=Improved B.W.L. (Improved Bread Wheat Land)

119

Table 12.Respondents farm land ownership and crop type grown in 1996/97 E.C. cropping season Items/ list T.Farm Land Grazing Land Forest Land Improved Bread Wheat land (sum) - HAR-Paven - HAR-1685 -HAR-1709 -Durum -Local variety T.W.L. Other crop -Teff -Chick pea Lentils -Pea -Faba bean -Vegetables Adopters Ha. N % 309.26 99 100.00 20.45 54 54.55 101.00 101.75 44.75 53.50 3.5 5.5 53.8 161.05 80.25 42.01 2.63 20.38 2.25 0.69 4 99 75 83 5 4 86 99 92 98 9 59 6 6 4.04 100 .00 75.76 83.84 5.05 4.04 86.87 100 92.93 87.88 9.09 59.60 6.06 6.06 N-adopters Ha. N % 90.50 51 100.00 9.95 29 56.86 0.35 1.25 0.50 0.75 28.67 29.92 35.50 17.96 1.25 4.39 1.10 0.38 2 3 48 51 43 38 4 17 3 2 3.92 5.90 94.12 100 84.31 74.51 7.84 33.33 5.88 3.92 Total Ha. N % 399.76 150 100.00 30.40 83 55.33 1.36 100.00 45.25 54.25 3.50 5.50 64.55 190.97 115.75 59.97 3.88 24.77 3.35 1.07 6 99 4 104 150 147 136 76 9 8 5.33 4.00 100.00 2.67 69.33 100 98.00 90.67 13 8.67 50.67 6.00

(Source: Computed from own survey data, 2005)

120

Table 13.Respondents livestock ownership

Types of Livestock Oxen -Adop. - Nadop. Cow -Adop -N-Adop. Bull -Adop - N-Adop Heifer Adop - N-Adop Calf -Adop - N-Adop Sheep Adop - N-Adop Gaot Adop - N-Adop Horse Adop - N-Adop Mule Adop - N-Adop Donkey Adop - N-Adop Poultry Adop - N-Adop Bee-hive Adop - N-Adop

Livestock No. 301 110 85 42 64 22 52 18 42 29 223 106 25 21 4 26 7 178 61 510 131 18 1

Owners No owners 98 42 68 33 46 19 41 14 39 20 80 29 8 21 4 25 6 91 38 80 29 5 1

Owners (% ) 98.99 82.353 68.687 64.706 46.465 37.255 41.414 27.451 39.394 39.216 80.81 56.863 15.69 21.212 7.843 25.253 11.765 91.92 74.51 80.81 56.863 5.051 1.961

max

Min

Range

average per Owners 3.07 2.62 1.25 1.27 1.40 0.86 1.27 1.286 1.08 1.45 2.79 3.65 3.125 1.00 1.00 1.04 1.17 1 .96 1.60 6.37 4.572 3.60 1.00

10 6 3 3 3 2 3 2 3 3 18 13 5 1 1 2 2 5 4 45 12 10 1

1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1

9 5 2 2 2 1 2 1 2 2 17 12 3 1 1 4 3 44 11 9 -

(Source: Computed from own survey date, 2005)

121

Table 14.Respondents livestock ownership in Tropical Livestock Unit (TLU) Types of Livestock Oxen Cow Bull Heifer Calf Sheep Goat Horse Mule Donkey Poultry Total By- Adopters 301 85 48 39 10.50 29 3.25 23.10 28.60 124.60 6.63 698.68 Number of Livestock owned By- N- Adopters Total No. of livestock 110 411 42 16.50 13.50 7.25 13.78 4.40 7.70 42.70 1.703 259.533 127 64.50 52.50 17.75 42.78 3.25 27.50 36.30 167.30 8.333 958.213

(Source: Computed from own survey data, 2005) Table 15.Conversion factors used to estimate the households livestock ownership into tropical livestock units (TLU) Animals Calf Heifer & Bull Cows & Oxen Horse Donkey Ship & Goat Chicken/poultry Source: Strock et al., (1991) TLU-equivalent 0.25 0.75 1.00 1.10 0.70 0.13 0.013

122

Table 16.Discrete characteristics of respondents

Non Adopters (51) Characteristics Number Level of Education Illiterate Literate Read & Write Elementary School Junior Secondary High School House hold sex -Male -Female Health Status -Un-healthy -Healthy Access to credit -Yes -No Leadership /Social status -Yes -No Off-farm income -Yes -No Distance of credit institutions -Far -Close Distance of DA office -far -close Market Access -Far -Close Access to labor -No access -Employment and other Sources Access to Extension Service -Yes -No Education Level -Illiterate -Literate 33 4 14 11 40 7 44 43 8 48 3 42 9 43 8 48 3 43 8 48 3 51 18 33 Percent 64.71 7.84 27.45 21.57 78.43 13.73 86.27 84.31 15.69 94.12 5.88 82.35 17.65 84.31 15.69 94.12 5.88 84.31 15..69 94.12 5.88 100 35.29 64.71 Number 64 4 5 3 4 19 7 92 7 92 71 28 73 26 82 17 89 10 12 87 89 10 87 12 94 5 35 64 Percent 8.138 64.65 4..04 ..05 3..03 4..04 5..05 6.700*** 7.07 92..93 1.762 7.07 92.93 2.928* 71.72 28.28 8.965*** 73..74 26.26 0.005 82.83 17.17 0.994 89..90 10.10 1.456 12.12 87.89 0.994 89.90 10.10 1.456 87.88 12.12 2.665* 94.95 5.05 0.000 35.35 64..65 1 0.994 0.001 1 0.103 0.132 1 0.228 0.098 1 0.319 0.081 1 0.228 0.098 1 0.319 0.081 1 0.942 0.006 1 0.003 0.237 1 0.087 0.138 1 0.184 0.108 1 0.010 0.207 5 Adopters (99) Signifi cance (2sided) 0.149 Conting ency coeffici ent 0.227

X2

df

***, ** and* Significance at P<0.01, P<0.05 and p<0.10 respectively.

123

Table 17.Respondent farmers general information Adoption Category Adopters Summary of statistics Mean St.D Minimum Maximum Range Mean St.D Minimum Maximum Range Mean St.D Minimum Maximum Range House hold head Age 46.1010 13.2560 19.0000 80.0000 61.0000 46.4706 14.5305 20.0000 80.0000 60.0000 46.2300 13.6550 19.0000 80.0000 61.0000 House hold family size 5.85 2.19 1.00 11.00 10.00 5.10 2.39 1.00 10.00 9.00 5.59 2.28 1.00 11.00 10.00 Farmers extension experience (Ys) 7.8687 4.7866 2.0000 20.0000 18.0000 3.7647 1.7842 1.0000 9.0000 8.0000 6.4730 4.4850 1.0000 20.0000 19.0000 Farming experience 21.8990 11.0809 2.0000 55.0000 53.0000 20.7647 10.9829 4.0000 60.0000 56.0000 21.5130 11.0230 2.0000 60.0000 58.0000

Non-adopters

Total

(Source: Computed from own survey data)

124

Table 18.Factors affecting Intensity of adoption of improved bread wheat varieties (Maximum Likelihood Tobit Model Estimation) Variables HHHSEX HHHAGE EDUHHH HEALSTAT FAMILYSI HHOFFINC DISDAOF1 TOTLIVUN SUMOWRE CRINMFF1 OTSOLA1 OXTLU YEXPEXTS GEXSERVE
Sigma

Coefficient 0.1854701150 0.3863329500 0.1697049462 0.38138220125 0.1544308922 0.1409765567 0.1685119034 -0.8908541184 -0.3802017439 0.1719842596 -0.2466522041 -0.3024284552 0.4835498401 0.2590100298

Standard Error 0.82006662 0.15596286 0.56398485 0.72107639 0.11637647 0.68473026 0.88888172 0.10884712 0.22997223 0.96902492 0.80402014 0.22093233 0.54842929 0.11097534

b/St.Er. 2.262 2.477 3.009 5.289 1.327 2.059 1.896 -0.818 -1.653 0.018 -0.307 -0.137 0.882 2.334

P (/Z/>z) 0.0237** 0.0132** 0.0026*** 0.0000*** 0.1845 0.0395** 0.0580* 0.4131 0.0983* 0.9858 0.7590 0.8911 0.3779 0.0196**

0.2268484851

0.15882571

14.283

0.0000

Log likelihood function= 6.582514 *, **And*** indicate the level of significance at 10%, 5% and 1% respectively.

125

Appendix.2. Interview Schedule for data collection from. Farmers

The objective of this Interview Schedule is to collect information from farmer respondents on improved bread wheat production in Akaki area, rural part of Akaki Kaliti sub-city of Addis Abeba administration from December/ 2004 to March/ 2005. The study is conducted for academic purpose. Hence, we request your honest & fair responses to fill up this interview schedule. 1. General & personal information of the respondent 1. Respondents name.. 2. Sex; 0 = female 1 = Male

3. Age..years 4. Marital statuses; 1.Married, 2.Single or unmarried, 3.Divorced, 4.Widow/Widower. 5. Rural Kebelie Administration/ Peasant Association Village. 6. Previous or current leadership status; 0 = No, 1 = Yes 7. Educational Status: 0 = Illiterate, 1 = Literate 8. Educational level: 1. Read & Write, 2.Grade 1- 6, 3.Grade 7- 8, 4. Grade 9- 12, 5.above grade 12 9. Household Characteristics Information Table 19.Household characteristics No Name of house hold members Sex Age Educational status

126

10. Land holding and farm characteristics of the sample households Table 20.Land holding & Farm Characteristics of the sample households No 1 2 3 4 5 6 Types of land use Cultivated (farm) land Grazing land Homestead land Forest land Unused land Total land holding Own (ha) Rent (ha) Total (ha)

11. Livestock ownership Appendix Table 21.Livestock ownership

No 1 2 3 4 5 6 7 8 9 10 11 12

Types of Livestock Ox Cow Calf Bull Heifer Horse Mules Donkey Goats Sheep Chicken Bee in Hive

Number

12. Types of crop grown in the survey year

127

Table 22.Types of crop grown in the survey year No Types of crops Land in (ha)

13. Involvement in irrigation production; 0 = No, 1 = Yes 14. Land size for irrigation production---------- ha. 15. Involvement in improved bread wheat production: 1.only this year 2.This year & in the previous year 3.In the previous year but not this year 4. Never involve. 16. Reasons for involvement: 1.High yield 2.High market demand quality 3.Pest/disease resistance 4.Frost resistance 5.Short maturity date 6.High food quality 7. Good storage quality 8. Good quality of cook ability 9. Good straw quality 10.Seed availability 11.Seed availability 12. Good information service 13. Fertilizer availability 17: Reasons for un-involvement: 1. Low yield 2.Low market demand 3.Low pest/ disease resistance 4.Low frost resistance 5.Long maturity date 6.Poor food quality7. Poor storage quality 8.Poor cooks ability 9.Poor straw quality 10. High seed price 11.Shortage of seed 12.Shortage of fertilizer 13.Lack of information 14.Lack of money and credit 15.Late arrival of seed 16.Late arrival of fertilizer 17.High interest rate of credit 18. Reasons for discontinuity: 1.Poor yield performance 2.Poor pest/ disease resistance3. Poor market demand 4. Poor frost resistance 5.Poor storage quality 6.Long maturity date 7. Poor cook ability 8.Poor straw quality 9.Poor food quality 10. High seed price 11. Seed shortage, 12.Fertilizer shortage 13.Poor extension supports 14.Late arrival seed 15.Late arrival of fertilizer, 16.Lack of money & credit 17.High interest rate of credit. 19. Total farm land/ cultivated land ----- ha.

128

20. Total wheat land---- ha 21. Land for improved bread wheatha. 22. Do you know paven-76? 0=no; 1=yes 23. Do you know HAR- 1685 (Kubsa)? 0=no; 1=yes 24. Do you know HAR- 1709 (Mitike)? 0=no; 1=yes

25: Use of disease and pest control chemical: 0 = No, 1 = Yes

26. Your future plan of involvement in improved bread wheat production 0 = Discontinue, 1 = Continue 27. Presence of problems related to fertilizer: 0 = No, 1 = Yes 28. Problems related to fertilizer: 0 = No, 1 = Yes 29.If yes, 1.High fertilizer price 2.Lack of credit to purchase fertilizer 3.High interest rate of credit to use credit to purchase fertilizer 4.Far distance of distribution center 5.Poor quality (mixed with impurities and caked) 6.Shortage 7.Lately arrival 8.Lengthy process & complicated format 9. Poor distribution processes 30. Presence of problems related to improved bread wheat seed: 0 = No, 1 = Yes 31. Types of problems: 1.Shortage 2.Poor seed quality 3.Late availability 4.Far distance of distribution center 5 Impurity problems and 6.Poor germination problem.

32. Extension support: 0 = No, 1 = Yes 33. Extent of extension support: 1. Poor 2. Medium 3.Good

34. Improved wheat seed rate application: 1.The recommended rate 2. Below the recommended rate 3.Above the recommended rate

129

35. Fertilizer rate of application 1: Apply the recommended rate, 2: Below the recommended rate 3. Above recommended rate 36. Chemical application: 1. Apply the recommended rate 2. Below the recommended rate, 3.Above recommended rate 37. Reasons for Below & Above recommendation use of agricultural inputs 1.Low quantity of input availability 2.High price of inputs 3.High interest rate of credit 4.Lack of credit & money 38. Frequency of weeding: 1.One 2.Two 3.Thrice 4. Four & above 39. Frequency of plowing: 1. One 2.Two 3.Three 4. Four & above

40. Characteristics of improved bread wheat varieties: (1 = High, 2 = Medium, 3 = Low) Appendix Table 23.Improved bread wheat varieties characteristics No 1 2 3 4 5 6 7 8 9 Characteristics Frost resistance Pest/Disease resistance Seed size Cocking time Storage quality Yield performance Market demand Food quality Color quality Varieties HAR-1685 KAR-1709

Fovon-76

41. Seed selection criteria: 1.Pest/Disease resistance 2.Frost resistance 3.High yield performance 4.high market demand 5.Attractive color 6.Short maturity data 7. Good food quality 8.Low time taking 9. Good straw out put and good quality 10. Good storage quality 11. Good Germination and till ring capacity 42. Improved bread wheat seed source: 1.Purchase from market, 2.Exchange from other owners, 3.Own seed from previous product, and 4. Borrow from owner formers, 5.Cooperative, 6.MOA 7. Seed enterprise, 8.Research organization

130

43. Other Agricultural input sources: 1 Cooperatives, 2.MOA, 3.others 44. Access to credit service: 0= No 1: yes 45. Credit sources: 1.Cooperatives, 2.Ethiopian 3. MOA, 4.Other- Credit Institution, 5 Individual/ private lenders 46. Presence of credit problems: 0=no 1=Yes 47. Types of credit problems: 1.Shortage 2.Long and complex process, 3.high interest rate, and 4. Far distance 48. Support from relatives and other colleagues to solve financial constraints to purchase inputs: 0=no 1=yes 49. Distance of credit providers Institutions=far 50. Do you have Access to market? 0=no 51. Market distance=far 1=close 52. Do you have Access to extension Service? 0=no 1=yes 53. Distance of Development Agent Office: 0=far 54. Have you attended training? 0=no 1=yes 1=yes 1=close 1=yes 1= close

55. Have you attended demonstration and field day programs? 0=no

56. Can the DA call the farmers for extension meeting with out the permission of government authorities? 0=no 1=yes 57. What did you feel when called for extension meeting? 0=un-happy 1=happy 58. What are your Sources of Agricultural and input information sources? 1. DA 2. Radio, 3.Television, 4.Written materials, 5.Training, 6.Field, day and demonstration, 7.Posters, 8.PAleaders, 9.Community leaders, 10.neighbours and colleague farmers,

131

59. From the following, to which one assign your- self? 1. Mode farmer, 2.follower farmer, 3.Neither of them 60. How many years of Experience do you have in agricultural extension? Years 61. Total farming experience in years? ...................... Years 62. Have you got training and sufficient information on improved bread wheat? 0=no 1= yes 63. If you did not get training how did you perform production operations? 1. Using try and error methods, 2. By copying from other experienced farmers, 3. By asking support from DA 64. Purpose and use of off-farm income: 1.for house hold food consumption and other costs, 2.for input purchase, 3.for labor hiring, 4.for health cost covering, 5.for all 65. Do have access to labor outside the household labor? 0=no 1=yes

66. If yes, your sources of labor out side the household labors: 1. Hired labor 2. Cooperation labor from colleague and relative framers, 3. Exchange labor 67. Who make the decision on off-farm income? 1. Family head, 2. Husband 3. Wife, 4.Husband and Wife, 5.The household members 68. Which type of agricultural operation is critical to you and need higher labor? 1. Plowing, 2.Sowing, 3.Weeding, 4.Harvesting, 5. Threshing 69. Do you have plowing oxen? 0=no 1=yes

70. If no, how do you plow your farmland? 1. Using oxen plowing, 2.Through labor exchanges 3. By asking cooperation

132

Table 24.Cramers V and Pearsons R values for Discrete and Continuous variables

Hypothesized independent variables and their values Continuous variables Pearsons R value for Discrete variables Cramers V-value for Continuous variables discrete variables Age 0.257 Sex 0.211 Family size 0.167 Education 0.001 Farm land Livestock ownership in (TLU) Oxen ownership 0.292 0.257 0.247 Health status Leadership-position Off-farm income Distance of DA-office Extension service Other labor source Market access Credit service 0.108 0.244 0.006 0.099 0.133 0.099 0.081 0.140

(Source: Own computation) *Notice: 0 value=no association, 0-0.4 value= weak, 0.4-0.7= moderate and >0.7= strong association ((Sarantakos, 1998).

Table 25.Respondents leadership position Have leadership position Yes No Total (Source: own computation) ADs 48 (94.12%) 3 (5.88%) 51(100) NADs 73 (73.74%) 26 (26.26%) 99 (100) X2-test Total

121 29 8.965*** 150

133