•

This hilltdbl)llk is lI1<1itllv illwnd

with <;OIllt~ tr<1llling 11\ 1
M meth(lQ., whu wish to develop the
1

h('r, thnu~h it will also hI:' cd intt>rl'~t to .illlhoi:;e invlJlwd in

nin~ dnd polk.. dt!v-elopmenL

UBAf.DA

-­ ~ __I
\.

',\,
Participatory
Rural
Appraisal
Practical Experiences

J. Nabasa, G. Rutwara, F. Walker and C. Were

r-

Mil! •

Natural Resources Institute

ACTIONAID Overseas Development Administration
 © Crown copyright 1995

The Natural Resources Institute (NRI) is an internationally recognized

centre of expertise on the natural resources sector in developing
countries. It forms an integral part of the British Government's overseas
aid programmes. Its principal aim is to alleviate poverty and hardship
in developing countries by increasing the productivity of their
renewable nataural resources. NRI's main fields of expertise are
resource assessment and farming systems, integrated pest
management, food science and crop utilization.

NRI carries out research and surveys; develops pilot-scale plant,

machinery and processes; identifies, prepares, manages and executes
projects; provides advice and training; and publishes scientific and
development material.

Short extracts of material from this publication may be reproduced in

any non-advertising, non-profit-making context provided that the
source is acknowledged as follows:

Nabasa, J., Rutwara, G., Walker, F. and Were, C. (1995) Participatory

Rural Appraisal: Principles and Practicalities. Chatham, U.K.: Natural
Resources Institute.

Permission for commercial reproduction should be sought from: The

Head, Publishing and Publicity Group, Natural Resources Institute,
Central Avenue, Chatham Maritime, Kent, ME4 4TB, United Kingdom.

Price £7.50

No charge is made for single copies of this publication sent to

governmental and educational establishments, research institutions
and non-profit-making organizations working in countries eligible for
British Government Aid. Free copies cannot normally be addressed to
individuals by name but only under their official titles. When ordering
(, please quote ECN3

Natural Resources Institute

ISBN 0 85954 392 7

ii
\

CONTENTS

Page

Preface iv

Introduction 1

When and by Whom? 4

Preparations before a PRA Exercise 5

Conducting a PRA Exercise 12

Processing Information 38

Developing PRA Techniques 39

Summary 50

Further Reading 52

('

iii
· .

PREFACE
In recent years, participatory rural appraisal has become the new
'buzz/ phrase in agricultural development. But what is it? How is it
applied? How can it assist you?

Several handbooks explain the different techniques which have been

developed, but less has been written on how they might be applied
and used in real situations. This handbook draws on the experiences of
an ACTIONAID and Natural Resources Institute (NRI) multi­
disciplinary team, in using participatory rural appraisal techniques to
establish and monitor a participatory research project with fanners in
Uganda. It is mainly intended for development workers who have had
some training in participatory rural appraisal (PRA) techniques, but
wish to develop their skills in using PRA in development work.

Some of the principles behind the practice of PRA and participatory

approaches are outlined. They are of practical use to all development
workers, but of particular relevance to agricultural extensionists and
research scientists. Through the use of PRA, development workers are
able to learn directly from the communities about the rural
environment and with them design more appropriate and focussed
intervention strategies. Throughout the handbook, examples from
actual case studies illustrate some of the range of participatory rural
appraisal techniques used, their application and interpretation. In
particular, the last, extended case study shows how different PRA
techniques were used/ in succession and in combination, to establish,
monitor and evaluate a set of on-fann trials.

As with other information-gathering or appraisal techniques,

participatory rural appraisal is not an end in itself, but rather a set of
tools which should be selected, modified and adapted to the situation.
Central to the successful application of these techniques is a shift
towards an attitude and approach which recognize that community
members and agricultural development workers must work together
as partners in the wider development process.

Based directly on the experiences gained by the ACTIONAID / NRI

team, this handbook gives practical guidelines on how to prepare for
and conduct participatory rural appraisal exercises.

iv

____I
\

INTRODUCTION
Most farmers in developing countries face a multitude of
problems which outsiders'" often fail to appreciate, or even
identify. The unpredictable natural environment most farmers
live in and the social, political, economic and personal
hardships they face daily are difficult for outsiders to grasp.
Outsiders often have severe limitations on the amount of time
they can spend in the field or at any particular site. As a result,
many studies are conducted and projects implemented without
proper consultation with the intended beneficiaries. This often
leads to an inappropriate use of resources and in many cases
the failure of projects to meet their objectives.

The 'conventional' approach to collecting information from

rural environments has been through the use of questionnaire
surveys, to generate quantitative data which can then be
analysed statistically. Data from different sites can then be
compared. Information can be collected by relatively
inexperienced people with the minimum amount of training or
knowledge of an area. Information which is outside the scope of
the questionnaire often cannot be accommodated and is usually
unprocessed and ignored. At best this can affect the perceived
relevance of any development project and at worst paint an
incorrect picture.
~

In order to address this concern there has in recent years been a

concerted move towards the development of a more
participatory style of development. This type of approach
requires the active involvement and participation of the rural
people being targeted. This approach has involved the further
development of different techniques of information gathering,
known collectively as methods of rapid rural appraisal (or
RRA), into techniques which involve the active participation of
the rural communities: PRA techniques.

The use of either conventional or PRA / RRA approaches, can

in many cases give quite different results and lead to different
conclusions, as is illustrated in Case Study 1.
• Outsiders = individuals from outside the locality, including urban educated scientists

1
 Case Study 1
Bac::kground
Two students were asked to evaluate problems of maize storage in a
village in Tabora region, Tanzania.

Method A
Student A designed a formal questionnaire to explore a number of
issues relating to maize production and storage. Twenty farmers
were selected at random from the village secretary's list and
interviewed over three days. Analysis of the data suggested that, on
average, farmers grew about 2.3 acres of maize, but this ranged from
less than 1 acre to more than 10 acres. On average, 5.2 bags per farm
were harvested, ranging from 2 bags to over 25 bags. Six out of the
twenty farmers sold some maize and 15 stored maize for up to six
months. Fourteen of these 15 reported that there was a problem
with insects in the stored maize after two months of storage.

It was concluded that the farmers have problems with low maize
yields and a serious problem of maize storage pests which affects
the food security situation of most farmers.

-, Suggested Action
The Ministry of Agriculture should conduct a major campaign to
teach farmers about increasing maize production and improved
maize storage practices.

PRA is not mainly a matter of techniques. The most important

quality an individual needs in order to conduct a successful
(,
PRA is an appropriate attitude towards participatory methods
and the members of rural communities. This is often quite
different to the attitude of many development professionals in
which the educated 'expert' knows what is best for the less

II1II:
 I

Investigating Maize Storage Problems

Method B
Student B used a more participatory, less structured approach. Two
hours were spent with the village chairman discussing the life and
some of the activities in the village. Student B asked the village
chairman to suggest five households to visit to include a range of
type of farmers; rich and poor, including some who are mainly
cattle keepers, maize growers and tobacco growers_ Student B also
visited the maize mill and discussed maize production with some of
the women waiting in the queue.

Student B discovered that there was a great variation in the amount

of maize people grew and their storage problems and requirements.
Some people in the village did not grow enough maize for family
consumption and bought maize when they needed it. A few grew
and sold a lot, and others grew only enough for their family. Storage
~
problems were mainly the concern of the few who grew a lot of
maize because they had most to lose.

Suggested Action
The Ministry of Agriculture should focus its activities on redUcing
storage losses with the farmers who are self-sufficient in maize.

educated farmer. Central to participatory approaches is the

need to learn directly from the rural community members.
Examples of different attitudes are summarized in Table 1.

3
Table 1 Differences in attitudes towards information gathering

Inappropriate Attitude Appropriate Attitude

Farmers are reluctant to adopt, Farmers have good reasons for
'lazy' and 'stupid' non-adoption
We know best Farmers know
their own working environment
Farmers should learn from us Leaming is a two-way process
with ourselves and the farmers
We must tell farmers We must listen to farmers
'Modem' methods must be 'Traditional' methods can be as
superior to 'traditional' good as 'modem' methods
Over-emphasis on quantitative Emphasizes use of qualitative data
data or indicators

WHEN AND BY WHOM?

Participatory rural appraisal techniques can be used by all
agricultural development workers. Different techniques can be
selected and adapted to suit different stages of either extension,
research or general development programmes; from the early
needs assessment stage, through monitoring to the evaluation
and final adaptation phases of any project.

Figure 1 outlines the variety of PRA techniques (described in

more detail on pp. 16-38), which can be used in the initial
'exploration' stages of an agricultural project when it is vital
that the outsider spends time learning as much as possible
about the farmers' environment.

There has been a tendency to use PRA techniques more often

during the earlier stages of the project cycle, but they can
equally well be used during the later monitoring and evaluation
stages of a project.

4
\

Exploration of the farming system

• Transect walks / direct observation. Seasonal calendars

• Social/resource mapping

Identification of target population

• Social mapping. Wealth ranking

""
Problem identification

• Pairwise / problem ranking

Investigate nature of problem

• Semi-structured interviews. Transect walks / direct observation

• Mapping

""

Identify possible solutions

• Semi-structured interviews. Direct observation

Figure 1 PRA techniques used by ACTIONAID farmer

participatory research unit to identify researchable issues

PREPARATIONS BEFORE A PRA EXERCISE

Participatory appraisal requires a different type of preparation
to formal ('questionnaire') information-gathering techniques.
Rather than spending time preparing and pre-testing
questionnaires, successful PRA requires more time spent on the
collection of background information.

5
\

.
Defining the study objectives

.
Defining the study area

Reviewing secondary data

.
(and making direct observations?)

.
Selection of a team to conduct the PRA exercise

Discussion & selection of information to be collected;

drafting of checklist (based on study objectives,

.
review of data & direct observations)

.
Discussion of timing and PRA technique(s) to use

.
Allocation of PRA team duties during exercise

Arrival at site

Figure 2 Preparing for a PRA exercise

6
\

Preparations will normally form a logical step-by-step

sequence, starting with the definition of the study and the study
area (Figure 2).

The objective of the study defines why the work is being

conducted (e.g. to design a development project for area X) and
the type of information to be collected (e.g. basic data on
agricultural needs and constraints).

The study area may be the entire area for which the objective is
relevant, e.g. the area covered by a development project. In
many other cases, it will be partially dependent on other
factors, such as the type of information required, or resources
available (time, funding), and will comprise a much smaller
area. If this is the case, care must be taken that the study area is
representative of the larger area concerned.

Having defined the study objectives and study area,

preparations will include a review of any existing general socio­
economic and production information or literature on the topics
to be covered during the exercise. PRA methods involve more
'discussion' with farmers than 'interviewing', so it is important
that the PRA team has a good background knowledge on likely
issues. Sources of such information include:

• libraries; especially universities, bachelor's and master's

theses, UN and World Bank reports;
• research stations;
• government offices;
• local co-operative or marketing organizations;
• NGOs;
• local research and development workers;
• local leaders: e.g. village chiefs, elders.

Such information is often in the form of reports, maps, papers

or even photographs. As with all situations, things are
continually changing and the information from such sources
should not be assumed to give the absolute or current picture.

7
\

Direct observation is another way the outsider can learn about an

area. Such observations can be made at any time prior to (and
during) the study. These observations cannot be assumed to be
anything other than a starting point but will give the outsider a
useful reference point for exploring during the appraisal
exercises.

Selection of the most appropriate team to conduct the PRA

exercise(s) is vital for the success of any PRA. Ideally, teams
should be small (no more than two or three individuals),
represent a good gender balance and include individuals from
different disciplines. Larger teams (of up to seven or eight
individuals) can also operate successfully and it may be
necessary to include individuals with a range of disciplines.
Sometimes, the exercise may require that the team be sub­
divided in order to explore certain issues (see Case Study 2).

Case Sludy2

Problem Identification with Butawata

Farmers' Group

Whilst conducting a group discussion on community problems

with It mixed group of fanners from Butawata village, it was
observed that some of the men were dominating the discussions
and many of the women were not participating at all. The PRA
team decided to change their approach. The fanners' group was
split according to gender and the PRA team divided into two sub­
teams.

Each group was asked to mention and then rank, the community's
priority problems. The women ranked labour shortage as their
most important problem, whereas the men failed to even mention
it!

In addition to gaining an insight into different gender issues,

many more topics were covered by sub-dividing the FRA team.
The issues discussed in the two separate sub-groups were then
summariud to enable feedback and discussion with the whole
group, as well as cross-ehecking on the gender issues.

8
 \

In many cases the size and make up of teams for PRA work will
be determined by factors concerning the individuals involved:

• level of experience in conducting PRAs;

• ability to communicate effectively with the local
community members;
• attitude to the community members;
• experience of and willingness to do field work;
• the presence of a visitor from outside the region in the
team.

Immediately prior to the PRA, members of the team should

decide on the type of information that is needed. A check list
should be compiled of the socio-economic, ecological,
environmental and technical areas to be covered.

. ,-~. f
For each area to be covered, the team should discuss the PRA
technique or techniques that will be most suitable for gathering
information. However, the choice of techniques may change in
the field, as information comes in and as the suitability of
different techniques in that situation becomes clearer (see Case
Study 7).

Ideally the number and 'types' of informants required should be

decided before hand. Initial contacts with any community are
often made through one or more key individuals, i.e. local
officials, community leaders, or other development workers.
These key individuals may be able to link the PRA team with
suitable informants or assist in identifying them through
wealth-ranking exercises.
. '
In the early stages of work with any particular community,
general 'background' information can be most successfully
" ' collected from groups and more specific, 'household' details
from individuals.
" is....'-"-,::··:
~<' •. ~ T"~"" ~

.~ ..
'.

9
 Information needed

~*
,
PRA team resources

Selection of PRA technique

Individual or group?

Identification of individual or group

Figure 3 Matching technique to type of informant

Prior to the actual PRA exercise, the PRA team should decide
amongst themselves the different responsibilities of the different
team members. Usually one member will lead the exercise,
another will record the information generated and if necessary
one may need to act as a translator.

Effective communication between farmers and the PRA team is

vital for the success of any PRA exercise. Often members of the
PRA team will need a translator to talk with farmers. A
translator can be part of the PRA team, or someone recruited
locally. A translator should take care not to 'interpret' what is
being said, but to record the direct translation or verbatim
responses, and other team members should be aware of this
potential problem. This is especially important when discussing
more specialist topics in other languages; some languages may
use words which cannot easily be translated into English, and
lack equivalents for some English words (notably many
scientific terms). The members of the PRA team should always
take care to converse with the farmer rather than the translator,
and to make sure that all the farmers present understand the
language used. Even if many of the farmers speak a national or
international language, such as Swahili or English, it does not

10
 \

mean that all of them do, and using such a language may
exclude many farmers, especially women and the less educated,
from the discussion.

The means of recording any information which is generated

should also be determined before hand. Ideally, this should be
written down in note form during the exercise and typed up as
soon as possible afterwards. At least one member of the PRA
team should be assigned this responsibility before the exercise.
In situations where notes cannot be taken during the exercise, i.
e. if the team is too small, notes should be made as soon as
possible after the exercise, or by using a tape recorder.

Time is a major factor which has to be considered. In different

cultures and countries farmers are busy at different times of the
day depending on the season and the time of year, e.g. during
religious festivals, market days, local holidays, funerals.
Generally speaking, mornings are times when farmers are most
busy working in their fields or around the home. Afternoons
tend to be the time when they can spare more time for visitors.
This can vary considerably between men and women farmers;
in many cultures there is a clear division of labour by gender
resulting in either the majority of men or the majority of women
".,':"
being absent at different times of the day.
.'<'.
,~

"
..

If the PRA is to involve more than one individual, prior

',"
1 ..
>, • warning should be given ahead of the meeting, to mobilize and
~.
encourage informants who may be reluctant or unwilling to
attend. The optimum length of time needed for a PRA session
varies with the exercise, the interest of the individuals involved
and the innovativeness of the PRA team; the more interesting
the exercise can be made, the longer farmers will be happy to
participate. During sessions of more than two hours duration
many PRA team members and farmers will lose concentration,
become hungry and tired; sometimes this can be averted by
providing snacks.

Other important factors which should be borne in mind is the

means of transport to the site and the dress of the team members.

11
.. .~~7"

Neither should be too 'flashy'! It may be difficult to avoid

having to travel to the site in a big four-wheel drive vehicle, but
it is possible to avoid wearing expensive-looking clothes and
jewellery and carrying sophisticated cameras and other
technological gadgets.

CONDUCTING A PRA EXERCISE

Successful PRA requires the relationship between the PRA team
and community members to be much closer than in other forms
of research. Ideally, this relationship should build up over a
period of time and similarly the PRA exercises should be
conducted and built on over a period of time.

In order for this relationship to develop it is important for all

involved to get away from the old 'us and them' attitudes
which are barriers to an effective dialogue. In many cultures,
visitors to rural communities are offered preferential treatment,
for example being offered seats or chairs which literally
'elevate' them above the community members (who are usually
seated on the ground). This immediately distances them,
physically and psychologically, from the community. In many
situations, it may be impossible to avoid such formal seating
arrangements without giving offence, especially when one first
begins work with a community (Figure 4a). During subsequent
PRA exercises and once people get to know the members of the
PRA team, it is important to develop a more informal
atmosphere in which the 'us and them' barriers become less
evident (Figure 4b).

There are no hard and fast rules regarding the need or use of
different PRA techniques to obtain certain types of information.
The selection of the technique to use will depend on each given
situation and the make-up of the PRA team members and the
community informants. No PRA technique is foolproof and
information obtained from one source or from one technique,
should where possible be cross checked with other sources
and / or by using other techniques.

12
 ~. \'1

'&,;','­
l,;.'.

Figure 4 Conducting meetings

Who are the development workers? Both pictures represent
a group of development workers meeting with farmers.
Which situation is the best for conducting a PRA exercise?

13
..
..
~~.:'

The key to success of any PRA lies with the participants. The
more relaxed both the PRA team and the informants are, the
better will be the flow of information between both parties. This
can be achieved by making the exercises as enjoyable as
possible.

The PRA exercise should be conducted at a location where the

informants feel comfortable, e.g. in their home compound or
field, and if possible at a site near the topic under discussion, or
in or near a field with the crop being discussed. If this is not
possible during the exercise, the PRA team should if possible
visit such a site immediately before or afterwards.

Table 2 Group versus individual approaches

Group Approach
Advantages
1. Good for general information
2. Less time required to solicit the views of given individuals
3. Provides an immediate cross-check
4. Generates debate and consensus on village-level issues

Disadvantages
1. A few individuals may dominate and exclude the views of
\ poorer, minority or marginalized members of the community
2. Group members need to be carefully selected to avoid bias
3. Information can be too general, how 'things should be' or even
misleading
4. Usually needs an advance appointment and start often very
delayed
5. Management of larger groups may be difficult

14

~I
 t"':'l\:~'"

Should PRAs be conducted with a group or

an individual?
Some PRA techniques are best conducted within a group and
others with individuals. Information of a more general nature is
best collected using a group, whereas information of a more
specific or sensitive nature is best collected using PRAs
conducted with individuals.

In many cases when time and resources are limiting, the group
approach is the most cost-effective method. Group approaches
. , • tend to be used more in the earlier stages of any rural appraisal,
with the exception of wealth ranking techniques which are best
done with individuals.

There are advantages and disadvantages to both approaches

(Table 2).

Individual/Key Informant Approach

. .. Advantages
1. Good for descriptive, specific household level information
2. Information more personal, less general
3. Encourages freer expression than if peers present. May reveal
conflicts hidden in group situation
4. Do not necessarily need an advance appointment
5. Easier for inexperienced PRA practitioner to manage

Disadvantages
1. More time (per individual) than group situation
2. Cross-checking of information important
3. Fewer individual's opinions can sought
4. Can be more intimidating for some individuals

15
· .

Management of larger groups (over 20 individuals) can

sometimes be difficult and often excludes the participation of
some individuals. Groups of between 10 and 15 are usually the
most productive, encouraging greater participation.
Participants are able to concentrate for longer and more subjects
or topics can be covered.

The selection of an individual or individuals for a PRA exercise

can be random, but more often information is required from a
known 'type' of person, for example those within a given socio­
economic group or range of groups. Identification of and
introduction to such 'target' individuals can be through
othercommunity members Oocalleaders, community leaders or
other development workers) if time is limited.

A more systematic but more time-consuming method is that of

selecting 'target' individuals through conducting a wealth- or
well-being ranking exercise (see p. 29 and Case Study 7). From
the results of such an exercise, individuals of different
backgrounds can be identified and selected as required, though
sufficient time should be allocated for the process. A minimum
of one morning should be made available to enable wealth­
ranking exercises to be conducted.

Participatory Rural Appraisal Techniques

There is potentially no limit to the number of exercises which
can be described as being participatory appraisal techniques.
Any interaction an outsider might have with the member of a
rural community could be described as 'participatory rural
appraisal', but it is the quality of the interaction which is
important for a successful PRA.

PRA techniques can greatly simplify the collection and analysis

of some types of data. But there is a risk that some techniques
might greatly oversimplify (see Case Study 5), or even paint a
false picture of the farming system. There is also a risk that
team members may get carried away by techniques and use
what they have learnt, regardless of its usefulness in the
particular situation. The skill and judgement of the PRA team in

16
collecting and interpreting information is vital.

No PRA technique is foolproof. Responses given by informants

during PRA exercises will vary depending on the social group
to which the individual belongs and the time of the
(agricultural) year, and will change with time. It is vital that
data is cross-checked or verified using a combination of
different informants (of known 'well-being' status within the
community), and different PRA techniques (or combinations of
techniques).

Where possible data from different times of the agricultural

year and different years should be used for cross-checking, but
this is not always practical.

Rather than being a weakness of the approach, discrepancy and

contradictions enable the PRA team to learn more about the
environment or community being studied. Such discrepancies
should be seen as opportunities for further learning rather than
errors to be glossed over or ignored. In such cases the PRA team
should return to the sources of any apparently contradictory
information and probe the issue further.

In broad terms there are two types of technique used in PRA,

based on discussion/ dialogue and diagramming.

Most PRA techniques involve elements of both interviewing

and diagramming and it is often possible (and usually
desirable) to use more than one technique at a time to get the
most out of the situation.

Discussion I dialogue
Interviewing, communication and interpersonal skills are all
important qualities needed by PRA practitioners. The PRA team
must be able to put the informant(s) at ease and give them the
confidence to share their knowledge and perceptions by
showing a sincere interest in learning from the community. The
PRA facilitator not only has to be an effective communicator but
also a good listener and quick thinker.

17
Unlike the more formal, structured interviews used in
conventional questionnaire surveys, during participatory rural
appraisal informal discussions or 'semi-structured interviews'
are conducted. At the start of the discussion it is it is not known
what new information will be brought to light and it is the role
of the interviewer to explore and probe to 'discover' and learn
new information. Leading or closed questions, (those which
require 'yes' or 'no' answers), should be kept to an absolute
minimum and informants must not be discouraged from saying
things.

Open-ended questions and probing questions are especially

useful. For example, rather than asking farmers for specific
information on banana acreage and yields (which in many cases
are not very meaningful given the range of management,
cropping and harvesting practices), explore the issue by asking
questions such as;

• why the farmer grows bananas

• what types are grown
• where and when are bananas are grown
• how they are grown
• who grows bananas.

A more participatory, semi-structured approach will often

produce different results from one that makes prior
assumptions and uses a questionnaire to gather information
(Case Study 1).

During the dialogue farmers will be encouraged to speak if the

interaction is more of a conversation than a 'question and
answer' session. More information will be obtained from
farmers when they are giving their observations on a topic in
front of them rather than by talking from memory. Where
possible analogy should be used to illustrate a point or concept.
Use simple terms and expressions with which they are familiar.
Abstract terms should be avoided.

18
 ·'f···-f;~·';T:''t.';:·: J.\,.,.'

Case Study 3
Guide to Interviewing Farmers
Involved in Soil Trials
Broad objective To interview farmers conducting
participatory on-farm trials to test the effect of different green
manure treatments of soil fertility

Specific objective To explore farmers' perceptions of:

.. soil fertility / infertility (1)
.~ .. cause of trial site infertility (2)
.. the current trial (3)
.% .. possible modifications to trial (4)

Sample checklist for soil trial interviews

General discussion on soils
• What is a poor / infertile soil
• How do soils become infertile / what leads to infertility?
• What local methods are used to improve fertility?
• Are there problems with these methods?

2 Discussion on soils using trial site as an example

• Previous use of site (cropping, inputs, any fallowing etc.)
• When was an effect noticed? What caused it?
• What improvement methods have been used in the past?
When?

3 Discussion on trial
• What were the objectives of the trial?
• What treatments were used?
• What observations were made during the trial?
How do the treatments compare?
What indicators are being used to assess effectiveness?

4 Way forward?
Are you planning to continue with the trial? (reasons for yes
or no)
How many seasons do you need to fully assess each
treatment?
What modifications are you planning to make? Why?
What are your neighbours' and other farmers' comments?

19
 Table 3 Common interviewing errors

Team Errors

1 Interrupting each other

2 'Helping out' another team member
3 Switching or dropping subtopics
4 Debating issues within PRA team and not with farmers
5 Too many individual interviews at expense of group meetings
6 Lack of team preparation
7 Lack of regular team reviews between interviews

Semi-structured interviews are based on a checklist of points or

topics to cover the broad and specific objectives of the
interview, and should be planned in advance (Case Study 3).

Check lists should be used during such exercises with care.

They are not a substitute for a questionnaire but rather an aide
memoire to remind the interviewer of all the topics to be
covered. If they are used in a 'step by step' manner the resulting
interviews can become too formal, too structured and rather
inflexible and less participatory, as well as too repetitive.

Semi-structured interviewing is the most basic of PRA

techniques and possibly the most difficult to master.
Interviewers must guard against a natural tendency to pursue
their own specialisms. Aspects of semi-structured interviewing
(, are involved in all PRA exercises and can be used as a substitute
for some techniques.

Errors can occur at any time during an interview either on the

part of individual members, or as a result of team dynamics.
Some common team and individual errors are summarized in
Table 3.

20
~' '~';~).,.,1~~"C
"

Individual Errors

1 Failing to listen closely

2 'Interviewing the translator'
3 Repeating question asked by someone else
4 Failing to probe
5 Failing to judge answers
6 Asking vague or insensitive questions
7 Asking leading questions
8 Failing to adequately record information
9 Failing to write-up from notes
10 Concentrating on one's own interests

Diagramming

" Diagramming exercises, such as social mapping and problem

ranking, and combinations of these with semi-structured
interviewing, can have an advantage over using interviews
alone. Diagrams can give clearer and more precise information,
especially with less articulate individuals. In situations where
consensus is possible, agreement can be quicker.
Communication is facilitated and a diagram is usually easier to
analyse and to explore. Another advantage is that it tends to
focus peoples' attention away from the interviewer and towards
the issue under discussion.

One of the most important features of diagramming exercises is

that they can be conducted with farmers who are illiterate or
semi-literate. Many illiterate farmers show a remarkable ability
to represent their situation through maps and diagrams. But it
is important to conduct the exercise in such a way that such
people do not feel intimidated. Some less experienced PRA
practitioners tend to conduct such exercises in a classroom-type
environment, using either a blackboard or pen and paper. The
use of such materials should be avoided as they can intimidate

21
 Case Study 4
Open and closed discussions

Two researchers are conducting an informal survey on farmers'

perceptions of their soils and problems of infertility. The two
extracts from the discussions with farmers demonstrate how the
outcomes can vary depending on the type of question asked.

In both the examples, the researchers are attempting to use a semi­

structured interview technique but they reach different
conclusions. The 'closed' questioner merely attempts to confirm
his or her own ideas, but the 'open' questioner learns more about
the farmer's ideas.

The researcher asking the closed questions, first suggests to the

farmer several possible descriptions of soil fertility. It is assumed
that the fa,rmer uses the same indicator for infertility (declining
,yield) and that it would only be corrected through the application
of inorganic or organic fertilizers. Questions are phrased in such a
way to eonfirm the researcher's own assumptions and to 'force' a
set 9f quantifiable data.
In the second case, the researcher goes in with a more open mind
and probes the farmer on a range of issues relating to soil fertility
in general. Points mentioned by the farmer are followed up using
open~ended questions, usually beginning with words like 'how,
~what', 'when', 'which' and 'why'. The fatIner is encouraged to
expand on previous conunents, rather than agree or disagree with
the researcher's suggestions. Information is generated on a range
of soil fertility-related issues which are of interest to the farmer
and in helps understanding the problem.

In this case, the farmer does most of the ta1kiIlg whereas in the
other case the farmer give very short answers.

22
 ,-='

t- Question: How would you characterize the fertility of your soil;

excellent, good, average or poor?

Answer: Fairly poor

Question: Have your yields reduced in recent years?

Answer: Well, yes a bit

Question: How many bags of maize did you harvest per acre

last year?

Answer: I harvested about 6 in total

Question: And that is from 2 acres, so you got 3 bags per acre,

yes?

Answer: Mmmmmm

Question: And how many did you get from the previous year?

Answer: Last year I was lucky with the weather so I harvested

8 bags

Question: Eight bags from the same 2 acres, so your yield has

gone down from 4 bags to 3 bags per acre in one

season

Answer: Well, I suppose so but .

Question: Do you use fertilizer?

Answer: No, it's too expensive

Question: Do you use compost or manure?

Answer: No, it's not available

Summary: So your maize yields are falling each year because

the soil is becoming more infertile. But you don't use

any fertilizer because it's too expensive and you can't

use compost and manure because it's not available.

Answer: Well, I suppose so.

",

Closed discussion
Short interview, asking specific questions with no deviation from
set questions and no follow-up from leads in answers

23
 Question, How do you classify different soils?

Answer: Well, usually by their colour

Question: What does the colour tell you about the soil?

Answer: Dark-coloured soils are usually better than lighter

soils?

Question: When are they not better than lighter soils?

Answer: Sometimes it depends on how long you have been

cultivating on the soil and the area that it is

Question: The area?

Answer: Yes, like in areas near swamps but you can usually

tell by looking at how good the crops are

Question: What do you mean by 'good'?

Answer: The crops on good soils are better than on poor

soils

Question: In what way are the crops 'better'?

Answer: They are healthier, look greener and produce more.

Question: Which types of crops do you look at?

Answer: You can look at all types of plants. On very bad

soils you get very few plants, on waterlogged soils

you get lots of ferns and similar plants: on poor

soils you get lots of poor grasses which the

animals don't like very much, but on good soils

you get many different types of plants and the

weeds grow very fast. I always look at how my

beans or maize do to assess the types of quality of

the soil. If they are both green then the soil is good,

but if they are yellow you will harvest almost

nothing.

Question: What about this soil?

\ I Answer: Well, it is still quite good but after this season I will

have to stop cultivating it.

Open-ended discussion
Questions asked around a given theme, content of answer dictates the
content of the next question, commonly beginning with why, what,
which, where, when or how

24
 Question: Why?

Answer: I think it will have to be left to fallow for several

years, so I'll move onto that patch over there.

Question: But that patch is much smaller than this.
Answer: I know, but it one of the big problems these days.
There's not as much land to go round as there used
to be so we cannot fallow land like we used to in
the past and everywhere the fertility is declining.
",
,1
"
Question: What else can you do to improve the soil?
Answer: In the past we used coffee husks and fertilizers on
some of the crops, but they are all too expensive
now or else not available. We still mulch the
bananas but there's not enough mulching material
around and it's a lot of work cutting and carrying it.

Summary: Let me see, if I follow what is going on. In the past

there was enough land to enable everyone to
cultivate some and leave some fallow. When a patch
began to lose its fertility, it was abandoned and a
new patch opened up and used until it too began to
lose its fertility and then you would return to the
original patch.
Answer: Well almost, except that you needed a lot more land
in fallow than under cultivation. It was best to
cultivate for only 3 or 4 years, but it took at least 9
or 10 years to get its fertility back. And if you were
rich you could afford to use fertilizers or hire a lorry
to bring some coffee husks.

1..

25
or even exclude the less literate participants. The simpler the
materials used <e.g. drawing with a stick on the ground), the
better. During diagramming, symbols are often used to
represent different features. It is important that when symbols
are needed they can be easily identified with the feature they
are representing. In this way, participants who are either

Table 4 Common objects used as symbols used during diagramming

exercises
Symbol To Represent
Stones Poor soils, rocks, 'counters'
Castor oil seeds Crop pests
Bank note Money
Pen / paper Training courses
Maize cob Food / taste
Green leaf Vigour
Dried leaf Disease
Leaves or fruits Different food crops
Banana fibre string / stick Paths or roads

illiterate or have difficulty in reading and writing are not

excluded from the exercise. Examples of such symbols are
summarized in Table 4. If the use of such materials is not
possible, other 'unrelated' (but locally collected) materials,
simple symbols e.g. +, * or ~, or small pictures can be
used.

Use of Diagramming Techniques

Commonly used techniques include:

• Local histories;
• Transect walks and direct observations;
• Wealth/well-being ranking;
• Pairwise and problem ranking;
• Seasonal calendars;
• Social and resource mapping.

All these techniques require some element of semi-structured

interviewing. Other techniques can be developed to fit each

26
situation by combining two or more techniques together. Some
combinations and possible combinations of PRA techniques are
summarized in Table 5.
Table 5 Possible combinations of PRA techniques

Mapping Calendar Ranking Wealth Walks Local

histories
Semi-
structured *** *** *** *** *** ***
interview
Local
histories ** ** ** * *
Transect
walk ** ** * ?

Wealth
ranking ** ? *
Matrix
ranking
?
*
Seasonal
calendars *
Key: *** Commonly used together, ** Can be used together,
*Might be used together, ? Unlikely to be used together
Local histories/Time lines This technique taps informants'
memories, to recall important historical events which have
taken place in the area. This can be extremely important in
highlighting some of the causes of certain problems or how
changes have occurred. Informants are often the older members
of the community.

The technique is an integral component of any rural appraisal

and can involve the use of semi-structured interviews, matrix
ranking, seasonal calendars and other diagramming techniques.

Transect walks / direct observations These are systematic

walks during which the researcher discovers local practices,
observes, asks, and listens. The farmers talk about and describe
what, how and why they do things, their perceptions and
beliefs. In the course of observation the researcher is not only

27
 I
I
I
I
I
I
I
I
I
I
I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I
Soils: Shallow, ". Red, thin soils
I
'.11
I
Reddish, good
stony I
I : soils
I I
I I
I I
I I
Crops: Cassava "~I'. Sweet potatoes, 1'11• Old rnatoke Beer bananas,
, Maize, Vegetables , bananas, Coffee Coffee
I I
I I
I I
I
Problems: Erosion I
Weeds, Fertility
I
I
I
I
I
I
Poor fertility, Very infertile
I
I
J
declining I
I Low production
I
I I
I I
I I
Opportunities: Bunds _iIIWeed control 'II"

.,::.-----Irnprove soil fertitity ~III.

Figure 5 Transect Walk: Summary diagram of information collected on a transect walk

'~!f'

!.

learning about farmers' standard practices, but also actively

looking for those practices which vary from the norm.

These techniques are important for learning about the farming

system, identifying constraints or problems, potential topics for
research, potential local solutions and other available resources.

Wealth/Well-being ranking Wealth ranking is a simple socio­

economic study technique by which researchers, planners and
rural development workers can learn the ways in which
people's wealth or well-being differs from one another. It gives
a quick understanding of the relative socio-economic status of
each household in the community, and the community's
definition of wealth. It can assist the PRA team in identifying
potential key informants for other PRA exercises.

The technique defines categories of the rich and poor, or the

better-off and worse-off, which are meaningful to members of
the community, and places community members into those
categories. If possible this should be done carefully by asking
three different members of the community to place all their
neighbours in different categories (see Case Study 7). The key
to the successful use of this technique is the selection of
informants. It is important that all informants live in the
community being ranked and are familiar with all the
households in the community. Selection can either be at random
or from key individuals within the community, I.e. leaders,
school teachers, birth attendants, or in both ways. At least three
informants should be selected to enable some form of validation
and cross-checking of the information to take place.

One possible limitation in the use of this technique is the

tendency for informants, depending on the culture, to
overestimate or underestimate their place on the scale of well­
being. This can be avoided by increasing the number of
informants who are consulted. Another problem is that in some
cultures it will not be acceptable for people to comment in this
way on their neighbours' well-being, and some individuals in
any culture are likely to object.

29
 Case Study 5
Problem ranking
When asked to mention their biggest agricultural problems the
members of a women's group at Nganjo village, near Mityana in
Uganda, mentioned six problems:

• lack of capital;
• changing climate;
• difficulty in planning farm activities;
• lack of a nearby market;
• crop pests and diseases;
• declining soil fertility.

A matrix diagram was drawn on the ground and each issue was
compared with all the others in turn. Farmers were asked to
mention 'the most serious problem' in each pairing, each problem
being scored for the number of times it occurred and the problems
ranked in order, with declining soil fertility being given the most
importance and lack of capital the least.

Table 6 Problem ranking matrix

Soil fertility Crop pests

and diseases

Lack of capital Soil Pests

Changing climate Soil Pests

Planning farm adivitie Soil Planning

Lack of market Soil Pests

~t~
Crops pests and diseases Soil

Soil fertility

30
If these difficulties are found, or if time is a significant
constraint on the information-gathering exercises, it may still be
possible to apply the same general principles less formally to
allow the researcher to differentiate between rich and poor
households. This is important as otherwise the poorer members
of the community may be excluded from the research process in
favour of the wealthier.

Pairwise and problem ranking These techniques can be used

either with items important to the community such as crops or
foodstuffs, or problems perceived by the community. In either
case each item is compared to various others and assigned a
ranking. Apparently unrelated issues or problems can be
compared using pairwise or problem matrices, and related
issues can be compared using a preference matrix. These
techniques are usually used in a group situation with the scores
being recorded in a matrix diagram, often on the ground using
different objects to symbolize different items.

Lack of Planning Changing Total

market farm activities climate

Market Planning Climate 0

Market Climate 2

Market 2

31
These techniques can be useful in discovering, weighting and
prioritizing problems as presented by groups of farmers during Case Study 6

the early stages of identifying and setting the research agenda. Interpreting preference·ranking results

The techniques can also be applied during discussions with

individuals, not necessarily using diagrams. The techniques In an exercise conducted with a group of Banyankole farmers

bring out the perceived seriousness of farmers' problems and during the millet-growing season, millet was ranked as the second

point towards their possible causes. They can be used during 'most important' food after beans. Cassava was ranked eighth.

the early stages of any research cycle in the prioritization of

research topics. After further analysis and discussion with other individuals the

importance of cassava became more apparent:

An individual's ranking of different problems and preferences

can be influenced by different socio-economic factors, such as: • as an important ingredient in the preparation of many
millet dishes;

• during dry spells when millet was often unavailable;

• their relative wealth status within a community, e.g. the as a drought-tolerant crop;
better off commonly place more importance on cash crops •
• as a very versatile food which can be eaten in a variety of
relative to annual crops; ways; cooked and uncooked.
• their ethnic group, e.g. by reflecting traditional food

preferences.
Had this exercise been conducted at a different time of year the
result would have been different.
Many other factors can influence how each item is ranked
relative to the others. These include:
to recognize others. Further discussion may show that there are
• lack of awareness about the (long term) implications of
linkages between items (Case Study 6).
certain issues;

• seasonal changes in perceptions, i.e. during the dry spells Seasonal calendars These are diagrammatic representations of
storage may have priority, during planting seed/hoes are different activities, problems and opportunities which occur
priorities etc.; during different times of the year and which have an effect on
• cultural beliefs which have not kept pace with recent
people's lives and farm activities, e.g. rainfall, labour demand.
changes;
They are a useful introduction to the researchers to learn about
• cultural beliefs or myths, which discourage certain
the farming environment and for the farmers to get to know the
activities or crops from being grown;
researchers and PRA techniques.
• expectation of 'handouts' or assistance from the PRA

team.
Seasonal calendars can used to indicate which times of the year
farmers are more occupied, need more labour on farm, receive
It is important that the PRA team are aware of such factors and more or less rainfall, when different farm activities are
structure their techniques accordingly. Failure to do so may conducted or when food or money is in most abundance (see
result in various biases. example in Case Study 7).

The ranking created by these exercises is one-dimensional and Social and resource mapping This is a simple mapping/
can oversimplify the relative importance of some items and fail diagramming technique in which farmers are asked to draw a

32 33

1 __ i£!!_
 .._----------------­
These techniques can be useful in discovering, weighting and
prioritizing problems as presented by groups of farmers during
the early stages of identifying and setting the research agenda. Case Study 6
The techniques can also be applied during discussions with Interpreting preference-ranking results
individuals, not necessarily using diagrams. The techniques
bring out the perceived seriousness of farmers' problems and In an exercise conducted with a group of Banyankole farmers
point towards their possible causes. They can be used during during the millet-growing season, millet was ranked as the second
the early stages of any research cycle in the prioritization of 'most important' food after beans. Cassava was ranked eighth.
research topics.
After further analysis and discussion with other individuals the
importance of cassava became more apparent:
An individual's ranking of different problems and preferences
can be influenced by different socio-economic factors, such as: • as an important ingredient in the preparation of many

millet dishes;

• their relative wealth status within a community, e.g. the • during dry spells when millet was often unavailable;
better off commonly place more importance on cash crops • as a drought-tolerant crop;
relative to annual crops; • as a very versatile food which can be eaten in a variety of
• their ethnic group, e.g. by reflecting traditional food
ways; cooked and uncooked.
preferences.

Had this exercise been conducted at a different time of year the

result would have been different.
Many other factors can influence how each item is ranked
relative to the others. These include:

• lack of awareness about the (long term) implications of

to recognize others. Further discussion may show that there are
certain issues;
linkages between items (Case Study 6).
. i
• seasonal changes in perceptions, i.e. during the dry spells
storage may have priority, during planting seed/hoes are Seasonal calendars These are diagrammatic representations of
priorities etc.; different activities, problems and opportunities which occur
• cultural beliefs which have not kept pace with recent
during different times of the year and which have an effect on
changes;
people's lives and farm activities, e.g. rainfall, labour demand.
• cultural beliefs or myths, which discourage certain
They are a useful introduction to the researchers to learn about
activities or crops from being grown;
the farming environment and for the farmers to get to know the
• expectation of 'handouts' or assistance from the PRA
researchers and PRA techniques.
team.

Seasonal calendars can used to indicate which times of the year

It is important that the PRA team are aware of such factors and
farmers are more occupied, need more labour on farm, receive
structure their techniques accordingly. Failure to do so may
more or less rainfall, when different farm activities are
result in various biases.
conducted or when food or money is in most abundance (see
example in Case Study 7).
The ranking created by these exercises is one-dimensional and
can oversimplify the relative importance of some items and fail Social and resource mapping This is a simple mapping/
diagramming technique in which farmers are asked to draw a
32
33

I •

Figure 6 Map of Butawata Village, Mubende District, Uganda. a, Drawn on ground by community members;
b, Formal copy identifying different households, water sources, places of worship and shops

RUTSIBA .
•
~

SH
... •
...
•
MUGOl\!JCO
... TBAi( ..
ETATUURWA

...
...
.. KIGANDO

-. . .-.
"'''.' .~

KAYUNCA

...
b

Key: .. FPR experiments • Dwellings ~ Water sources SHShop 'Ill" Fence

SC School TBA Traditional Birth Attendant
+
• Church ill Mosque \; Rocks
 a

Figure 6 Map of Butawata Village, Mubende District, Uganda. a, Drawn on ground by community members;
b, Formal copy identifying different households, water sources, places of worship and shops

RutSIBA

....
.. .•SH

•
MUQONGO
:ETATUURWA
.... TBAif
.:•

....
.... T8A
•
• IQGANDO

I
. .

•• 'iI-"".

1<AYUNGA

....
b

Key: .... FPR experiments • Dwellings ~ Water sources SHShop 111111 Fence

SC School TBA Traditional Birth Attendant

+
• Church .. Mosque " Rocks
 .lII!!";o • •• _

simple map or diagram of their area, locality or farm. This may

best be done in a group situation by drawing on the ground
with a stick (Figure 6a). One person at a time is given the stick
and asked to draw in certain features on the map, these are then
discussed or cross checked with the other group members for
verification. This type of exercise can generate a lot of local
knowledge and information regarding the: :a:::E
• physical features, e.g. hills, valleys; 00
~ ttl
• infrastructure, e.g. water points, paths, roads, places of 0 ~
ttl
worship, community meeting points, schools, markets, :::E / .c­
health units; ::§
...
• geographical distribution of households; ttl
OJ
~
• names of the household heads (and others);
• names of resource people in the community, e.g.
.c­
oS
;:l
carpenters, herbalists, builders, religious leaders.
... ~
§
f) \:1 V
)\ E
<J)
0
QJ u u
Once the map has been completed on the ground, a 'pen and "0 QJ
ooU
0 en ttl
paper copy' (Figure 6b) should be made for future reference ...
<0
f-<
<0
...., ?• ~

and if necessary future discussion with the group. :E

.~

'"
OJ
co
The technique explores people's perception of their community. ttl
The information obtained during such an exercise can be used ]
U \ I\.~=-)
oM ::5 \
lu~ ~CO
in a subsequent wealth ranking exercise. In addition, .~ ~

" information on other local features such as local shops, markets, .S

~
water sources, schools, places of worship, roads/paths and the 0
..c:
local divisions of the village can be gathered. '"
S
;:a
...CO
ttl

\
QJ

~ I
iI
This type of exercise can be successfully combined with most
other PRA exercises at all stages of a project cycle, including
u
:g
Il.. / I:::

ON
0,C
;:l
_
.~

"0;;. ;:l~
ttl
;.c;
§
needs identification, monitoring and evaluation. QJ 0 OJ
i
~U >
t-.
Social linkages can also be mapped out by means of Venn ...
QJ
::l
diagrams, which show the interactions and social organization ....CO
~
within a community, or a community's links to outside
institutions, as set of overlapping circles.

In social and resource mapping exercises, the results depend

heavily on informants' drawing accuracy and recall capacity.
Obviously many maps produced in this way are likely to be

36

1
__
 _ _ _ _ _ _ _ _ _ _IiiiIIIII..... J _
simple map or diagram of their area, locality or farm. This may
best be done in a group situation by drawing on the ground
with a stick (Figure 6a). One person at a time is given the stick
and asked to draw in certain features on the map, these are then
discussed or cross checked with the other group members for
verification. This type of exercise can generate a lot of local
knowledge and information regarding the:
:r:o
::E
• physical features, e.g. hills, valleys; bO
• infrastructure, e.g. water points, paths, roads, places of <t::
o ~
worship, community meeting points, schools, markets, ::E ~
health units; ::§
• geographical distribution of households; ...
1ll
• names of the household heads (and others); r::
• names of resource people in the community, e.g. C
carpenters, herbalists, builders, religious leaders.
Once the map has been completed on the ground, a 'pen and
paper copy' (Figure 6b) should be made for future reference
I

u
ttl
r::
and if necessary future discussion with the group. ~
.~
The technique explores people's perception of their community. iO
00
I The information obtained during such an exercise can be used 't
:E
12
I
in a subsequent wealth ranking exercise. In addition, u :E
information on other local features such as local shops, markets, gp
water sources, schools, places of worship, roads/paths and the
local divisions of the village can be gathered. 1s
ttl
This type of exercise can be successfully combined with most .§ to
ttl
"0
other PRA exercises at all stages of a project cycle, including p.,
~ :e
r::
needs identification, monitoring and evaluation. ~ ~
"E::E :>
Social linkages can also be mapped out by means of Venn B£

'tj ~
to­
e::
diagrams, which show the interactions and social organization ::l
within a community, or a community's links to outside
c: '"
o '"
u<t::
....00
~
institutions, as set of overlapping circles.
In social and resource mapping exercises, the results depend
heavily on informants' drawing accuracy and recall capacity.
Obviously many maps produced in this way are likely to be
36
I _

 _ _ _ _ _ _ _......:..._ _ ---.11
L

unreliable regarding scale, or even miss out important features It is useful to circulate such reports to the relevant offices or
of the community and its resources. There is also a danger of persons in the area, not only for cross checking but as a courtesy
the exercise being too dominated by ideas of the individual to those who have assisted in the fieldwork.
selected to draw the map.
Direct feedback to the communities is something which many
development workers neglect. This is not only another
PROCESSING INFORMATION opportunity to cross-check findings, but will be important for
fostering future collaboration.
PRA techniques have many advantages over conventional data
collection methods. Not only is there greater flexibility possible As the information contained in such reports will be the most
in data collection, but also it takes a shorter time to collect and up-to-date on any given community it should be used as a
analyse the data. It is relatively easy to use the different guide for future activities as well as being constantly upgraded.
techniques to generate a wealth of information about the local These reports should be readily available and not filed away as
situation, but without a systematic method for recording, a historical document!
processing and using the generated information most or all of
the advantages of PRA can be lost.
DEVELOPING PRA TECHNIQUES
Information collected during field work should be written up as
soon as possible after the event. Some suggest that this can Participatory rural appraisal techniques are constantly
easily be achieved using a portable computer in the evening developing and evolving. Lessons learned from one exercise
after fieldwork, but this is not always possible. Rather than should be adapted and incorporated in subsequent exercises.
delegating responsibility the PRA team leader should take on Error should not be seen as 'failure' but rather as a learning
the task him- or herself. A draft field report should then be opportunity from which to refine and modify ones technique.
circulated to other team members for comments, cross checking
and final amendments. Case Study 7 gives an account of how different PRA techniques
can be used, applied and modified during rural development
The style, spelling, grammar and polish of this type of report is work. The case in point describes the use of PRA by a team of
not as important as the information it contains. Avoiding any researchers conducting on-farm bean trials in a village in
delay in writing up and loss (or misinterpretation) of western Uganda.
information is more important. It is useful to document as much
as possible of what was said, as well as giving a brief
summarized interpretation.

Ideally a report should contain three sections:

• introduction, outlining when, why and how things were

done, any problems met and the reliability of the results;
• presentation and analysis of data;
• conclusions and recommendations.

38 39
 ,- _I J _

unreliable regarding scale, or even miss out important features It is useful to circulate such reports to the relevant offices or
of the community and its resources. There is also a danger of persons in the area, not only for cross checking but as a courtesy
the exercise being too dominated by ideas of the individual to those who have assisted in the fieldwork.
selected to draw the map.
Direct feedback to the communities is something which many
development workers neglect. This is not only another
PROCESSING INFORMATION opportunity to cross-check findings, but will be important for
fostering future collaboration.
PRA techniques have many advantages over conventional data
collection methods. Not only is there greater flexibility possible As the information contained in such reports will be the most
in data collection, but also it takes a shorter time to collect and up-to-date on any given community it should be used as a
analyse the data. It is relatively easy to use the different guide for future activities as well as being constantly upgraded.
techniques to generate a wealth of information about the local These reports should be readily available and not filed away as
situation, but without a systematic method for recording, a historical document!
processing and using the generated information most or all of
the advantages of PRA can be lost.
DEVELOPING PRA TECHNIQUES
Information collected during field work should be written up as
soon as possible after the event. Some suggest that this can Participatory rural appraisal techniques are constantly
easily be achieved using a portable computer in the evening developing and evolving. Lessons learned from one exercise
after fieldwork, but this is not always possible. Rather than should be adapted and incorporated in subsequent exercises.
delegating responsibility the PRA team leader should take on Error should not be seen as 'failure' but rather as a learning
the task him- or herself. A draft field report should then be opportunity from which to refine and modify ones technique.
circulated to other team members for comments, cross checking
and final amendments. Case Study 7 gives an account of how different PRA techniques
can be used, applied and modified during rural development
The style, spelling, grammar and polish of this type of report is work. The case in point describes the use of PRA by a team of
not as important as the information it contains. Avoiding any researchers conducting on-farm bean trials in a village in
delay in writing up and loss (or misinterpretation) of western Uganda.
information is more important. It is useful to document as much
as possible of what was said, as well as giving a brief
summarized interpretation.

Ideally a report should contain three sections:

• introduction, outlining when, why and how things were

done, any problems met and the reliability of the results;
• presentation and analysis of data;
• conclusions and recommendations.

~ . ~
 Case Study 7 Participatory
in On-Farm rn.: - I.
Objective
Two 'promising' new bean varieties selected by plant breeders
from trials conducted at the national research station, are
forwarded for on-farm trials. The objective of these trials is to
assess the acceptability of these varieties to the majority of small­
holder farmers in the country.

Site Selection
The research team is conducting a series of similar trials at five
different sites, in five different districts of the country. Before
embarking on any field work the research team needed to select a
representative village of each area and farmers from within that
village who were representative of the majority of small-holder
subsistence farmers.

The team approached the District agriculture office and asked

one of the local agricultural officers to advise them. After
discussing the agricultural activities of the district and consulting
some of the recent reports from the district with him, a nearby
village was selected. The team arranged to visit the village and
meet with the village extension officer.

The immediate priority of the team was to identify and select a

group of farmers to work with.

The team chose to achieve this by first conducting a resource

mapping and then a wealth-ranking exercise.

Farmer Selection
The research team, accompanied by the village extensionist,
began by paying several informal, courtesy visits to several key
individuals in the village. The team observed many aspects of the
village life and the banana-based agricultural system of the area.

At one farm a group of old men were relaxing under a tree. The
team were introduced to everyone and began discussing life in
the village. One old man said that he had lived in the village for
This
over 50 years and that he had seen many changes. He had seen
separatl
many people move to the area. In the past most people grew
Each pile
bananas, today many now grow quicker-growing crops like
wrong pile.
I

40 41 I

_ _I
 M'· '7 II_lIlIliiI• • • • • • • • • • • •

Case Study 7
Objective
Two 'promising' new bean varieties selected by plant breeders cassava and sweet potatoes. New
from trials conducted at the national research station, are and new schools built.
forwarded for on-farm trials. The objective of these trials is to
assess the acceptability of these varieties to the majority of small­
holder farmers in the country.

Site Selection
The research team is conducting a series of similar trials at five
different sites, in five different districts of the country. Before
embarking on any field work the research team needed to select a
representative village of each area and farmers from within that
village who were representative of the majority of small-holder
subsistence farmers.

The team approached the District agriculture office and asked

one of the local agricultural officers to advise them. After
discussing the agricultural activities of the district and consulting
some of the recent reports from the district with him, a nearby
village was selected. The tearn arranged to visit the village and
meet with the village extension officer.

The immediate priority of the team was to identify and select a

group of farmers to work with.

The team chose to achieve this by first conducting a resource

mapping and then a wealth-ranking exercise.

Farmer Selection
The research team, accompanied by the village extensionist,
began by paying several informal, courtesy visits to several key
individuals in the village. The team observed many aspects of the
village life and the banana-based agricultural system of the area.

At one farm a group of old men were relaxing under a tree. The
team were introduced to everyone and began discussing life in
the village. One old man said that he had lived in the village for
over 50 years and that he had seen many changes. He had seen
many people move to the area. In the past most people grew
bananas, today many now grow quicker-growing crops like

_J
40 41
 •

Table 7 Seasonal calendar showing workload and rainfall

Jan Feb Mar Apr MayJun Jul Aug Sep Oct Nov Dec
Rain .... ..... ... .. ..... .... ...
Labour ::. .. .... .... .... : .... ... ..... .... .. ..

42
43
 •

research '
extensio:
weekly
house. I'
agricultural
agriculture
themselves

Table 7 Seasonal calendar showing workload and rainfall

Jan Feb Mar Apr MayJun Jul Aug Sep Oct Nov Dec

Rain ...... ..... ... .. ..... .... ...

Labour ::. .. .... .... .... : .... ... ..... .... .. ..

42 43

..
 exercises were conduct< During these vi$its. man~ observations were made and .~
') foods and lim: discussion with the fanners new information was obtained
that the peOple' regarding~
eir most important
also noted. • the main crops being grown;
• where they were grown;
Table 8 Preference ranking of common foods • the different soils and management practh::esi
• the main production constraintS;
Rank By 'Preference' By 'Imporlance' • research opportunities.
Matoke bananas Cassava
Maize Beans Beans appeared to play an important 'food security' role With all
Sweet potatoes Sweet potatoes the farmers and some used them for income generation. Six
Beans Matoke bananas different bean varieties were observed. Some were more
Cassava Maize susceptible to damage from insect pests than others, some were
grown only for market and others only for home consumption. It
also became apparent there were a number of othe.!
characteristics whi.ch were of importance to different farmers,. bUt
which characteristics were the most important?

Semi~structuretl gT0up interview Prior to a meeting in the

afternoon with the women's group, a check list was drafted.

Check list for discussion on beans

• Give feedback on mornings observations.

Comments?
• Review of the previous PRA results.
Comments?
• Confirm the relative importance of beans
• Why are they so important?
,lore further the general issue of crJ sts and • What qualities does the 'ideal' bean have?
'arch team arranged another visit to group,
week. • What bean varieties does everyone grow?
• What qualities does each variety have?
arrived in the village in the • Discuss possibility of evaluating two
women in their fields. The· new varieties
t only to get to know the
but also to make their own
and specifically di.scuss
roblems with individual

44 45

I II

 During these visits many observations were made and through

discussion with the farmers new information was obtained
regarding:

• the main crops being grown;

• where they were grown;
Table 8 Preference ranking of common foods • the different soils and management practices;
• the main production constraints;
Rank By 'Preference' By 'Importance' • research opportunities.
1st Matoke bananas Cassava

2nd Maize Beans

Beans appeared to play an important 'food security' role with all
3rd Sweet potatoes Sweet potatoes
the farmers and some used them for income generation. Six
4th Beans Matoke bananas
different bean va.rietieswere observed. Some were more
5th Cassava Maize
susceptible to damage from insect pests than others, BOn\e were
grown only for market and others only for home consumption. If
also became apparent there were a number of other
characteristics which were of imporhntre to different farmers, but
which characteristics were the most important?

Semi-structured group interview Prior to a meeting in the

afternoon with the women's group, a check list was drafted.

to understand how
Check list for discussion on beans
fanning system, the
of their concerns and
ing to severity. The
• Give feedback on mornings observations.
Comments?
"omen was the lack of a
pests and diseases and
• Review of the previous PRA results.
Comments?
tural inputs and finally poor
• Confirm the relative importance of beans
• Why are they so important?
• What qualities does the 'ideal' bean have?
• What bean varieties does everyone grow?
• What qualities does each variety have?
in the village in the. • Discuss possibility of evaluating two
.en in their fields. The new varieties
t only to get to know the
, but also to make their own
'own and specifically discuss
se problems with indiVidual

44 45

III

 During the meeting one team member gave a brief feedback A total of nine characteristics were mentioned by the women as
report on all the places they had visited and the things they had being important qualities in a bean varieties. These were listed
learned. The team expressed their interest in learning more about and a pairwise-ranking exercise was conducted to discoVli!1' which
beans. A discussion based on the issues from the checklist characteristics were more important than others (Table 9). The
brought new information to light. results demonstrated the importan<:e of a drought-resistant, high­
yielding variety, which was tolerant to pests and diseases and
• At least 13 varieties of beans are grown in the village. tasted good.Following this exercise the research team introduced
• Only one or two varieties were suitable for selling. They the idea of conducting some on-farm tests of two 'promising'
should be high-yielding, big-seeded and white-coloured new bean varieties from the national research station. Both were
and store as long as possible. said to be high yielding and tolerant of dry spells.
• Beans grown for home conswnption should be tasty, easy
to cook, small and dark-coloured. The group members greeted this suggestion enthusiastically and
• A mixture of both short- and long-maturing varieties are expressed their willingness to participate in the exercise. The
grown to ensure a spread of any risk and as continuous a research team distributed 0.5 kg packets of seed of the two
1.larV'est period as possible. varieties and asked the farmers to plant them using their normal
husbandry practices.

Table 9 Pairwise ranking of desirable bean variety characteristics

Drought Rain Softness Pest and Colour of Cooking Yield Total

tolerance tolerance on cooking disease 'sauce' time
tolerance

Taste on cooking Drought Taste Taste Pests Taste Taste Taste 5

Yield Drought Rain Yield Yield Yield Yield 4

Cooking time Drought Rain Softness Pests Time - 1

Colour of 'sauce' Drought Rain Colour Pest - 1

Pest and disease

tolerance Pests Pests Pests 6

Softness on cooking Drought Softness - 2

Rain tolerance Drought 3

Drought tolerance - 6

46 47

I •
 During the meeting one team member gave a brief feedback A total of nine characteristics were mentioned by the women as
report on all the places they had visited and the things they had being important qualities in a bean varieties. 'These were listed
learned. The team expressed their interest in learning more about and a pairwise-ranking exercise was conducted to discover which
beans. A discussion based on the issues from the checklist characteristics were more important than othersCTable 9). The
brought new information to light. results demonstrated the importance of a drought-resistant, high­
yielding variety, which waS tolerant to pests and diseases and
• At least 13 varieties of beans are grown in the village. tasted good.Following this exercise the research team introduced
• Only one or two varieties were suitable for selling. They the idea of conducting some on-farm tests of two 'promising'
should be high-yielding, big-seeded and white-coloured new bean varieties from the national research station. Both were
and store as long as possible. said to be high yielding and tolerant of dry spells.
• Beans grown for home consumption should be tasty, easy
to cook, small and dark-coloured. The group members greeted this suggestion enthusiastically and
• A mixture of both short- and long-maturing varieties are expressed their willingness to participate in the exercise. The
grown to ensure a spread of any risk and as continuous a research team distributed 0.5 kg packets of seed of the. two
harvest period as possible. varieties and asked the farmers to plant them using their normal
husbandry practices.

Table 9 Pairwise ranking of desirable bean variety characteristics

Drought Rain Softness Pest and Colour of Cooking Yield Total

tolerance tolerance on cooking disease 'sauce' time
tolerance

Taste on cooking Drought Taste Taste Pests Taste Taste Taste 5

Yield Drought Rain Yield Yield Yield Yield - 4

Cooking time Drought Rain Softness Pests Time - 1

Colour of 'sauce' Drought Rain Colour Pest - 1

Pest and disease

tolerance Pests Pests Pests
• - 6

Softness on cooking Drought Softness -

• 2

Rain tolerance Drought - • 3

Drought tolerance -
• 6

46 47

I •
 d make a meaningful comparison the Conclusions
to grow them adjacent to a local bean From the results of this exercise, variety 1 appeared to perform
well agronomically, but was the least tasty so was given the
poorest overall value. Variety 2. on the other hand, iUthough
yielding less well, was the most tolerant to pests and. ~es
season the research team visited. the farmers and was the tastiest
observations :regarding the varieties'
the views and observations of the The group members commented that they would definitely
e 01\ each occasion. continue growing variety 2. and would consider growing variety
1 if a market could be found.

the research team and the women's

leeting to evaluate the performance of
was done by conducting a matrix
was $letched out on the ground and
at the top of each column. The
:ferent characteristics for
were ranked according to
for serond best and three
selected to represent the
matrix emerged.

Table 10 Performance ranking of bean varieties

Symbol Meaning Local Variety Variety

Variety 1 2

Green leaf Vigour in the field : ..

Castor seed Tolerance to pests
and diseases ..
Bean pod Yield ..
Maize cob Taste ..
-­
Overall value ..

48 49

I •

 order that they could maft=1+ a tneaning'ful cotnparison the Conclusions

were asked to I 4:k2:Dw them adjacent to a local bean From the results of this exercise, variety 1 appeared. mperform
J- r
well agronomically, but was the least tasty so was given the
j~ poorest overall value. Variety 2 on the other hand.· although
yielding less well, was the most tolerant to pests and ~
season theJ i '~arch teatn visited. the farmers and was the tastiest.
their own oI:p:::f ~ -~~ ervatioos regarding the varieties'

,
and to elidt th ~EZO , news and observations of the
es were tnade on e-=" ~ "! ell occasion.
-
The group members commented that they would definitely
continue growing variety 2 and would consider growing variety
1 if a market could be found.

Table 10 Perfonnance ranking ~fbean varieties

Symbol Meaning Local Variety Variety

Variety 1 2

Green leaf Vigour in the fieI-===:3 ..

Castor seed Tolerance to pest=-;=r
and diseases ..
Bean pod Yield ..
Maize cob Taste ..
Overall value : ..

48 49

I •
 SUMMARY
Don't
Participatory rural appraisal techniques are not an 'end' in
themselves, but rather a 'means' by which all involved in the • Violate taboos and nonns
• Demand appreciation
development process can quickly gain an insight into the • Use abstract language
sometimes complex and highly variable world in which many •
Interrupt, blame, suggest or promise
rural communities live. • Side with opinion leaders or agitate
• Manipulate or create needs
They can be adapted and applied by all types of rural
development worker to a variety of different situations. There Do
are distinct advantages in using PRAs alongside ot instead of
conventional survey methods. A lot of information can be • Find out about taboos and nonns
obtained from community members and analysed with them in • Stimulate farmers to talk
a short space of time. If conducted in a systematic way
• Provide facts and infonnation
• Be neutral and objective
information generated can be a lot more focussed and 'client­ • Build up a dialogue
orientated', leading to the planning, design and implementation • Assist fanners to evaluate
of more appropriate strategies for development. • Be patient
• Be creative, adaptable and innovative
Although there are no hard and fast rules governing the • Learn from 'errors'
• Use analogy
application of PRA techniques, there are some guidelines • Use a variety of PRA techniques
which, with an appropriate attitude on the part of the • Cross-check infonnation
researcher, will increase the likelihood of success.
Above all
• Listen and be interested
• Respect the farmers, their perceptions
andthcirknowledge

50 51

I •

SUMMARY
Don't
Participatory rural appraisal techniques are not an 'end' in
themselves, but rather a 'means' by which all involved in the • Violate taboos and norms
• Demand appreciation
development process can quickly gain an insight into the • Use abstract language
sometimes complex and highly variable world in which many • Interrupt, blame, suggest or promise
rural communities live. • Side with opinion leaders or agitate
• Manipulate or create needs
They can be adapted and applied by all types of rural
development worker to a variety of different situations. There Do
are distinct advantages in using PRAs alongside or instead of
conventional survey methods. A lot of information can be • Find out about taboos and norms
obtained from community members and analysed with them in • Stimulate farmers to talk
a short space of time. If conducted in a systematic way
• Provide facts and information
•
Be neutral and objective
information generated can be a lot more focussed and 'client­ • Build up a dialogue
orientated', leading to the planning, design and implementation • Assist farmers to evaluate
of more appropriate strategies for development. • Be patient
• Be creative, adaptable and innovative
Although there are no hard and fast rules governing the
• Learn from 'errors'
• Use analogy
application of PRA techniques, there are some guidelines • Use a variety of PRA techniques
which, with an appropriate attitude on the part of the • Cross-check information
researcher, will increase the likelihood of success.
Above all
• Listen and be interested
• Respect the farmers, their perceptions
and their knowledge

50 51

I •

 Selected Reading
Gilling, J. and Cropley, J.P. (1993) Needs assessment for
agricultural development: practical issues in informal data collection.
NRI Socio-economics Series 1. Chatham, UK: Natural resources
Institute

Grandin, B. (1988) Wealth Ranking. London: Intermediate

Technology Publications.

Heim, F.G. (1990) How to work with farmers: a manual for field
workers, based on the case of Thailand. Tropical Agriculture 4.
Weikersheim: Margraf.

International Institute for Environmental Development (1991)

Participatory Rural Appraisal. Proceedings of the February 1991
Bangalore PRA Trainers Workshop. London: lIED and
Bangalore: MYRADA.

Khon Kaen University (1987) Proceedings of the 1985 International

Conference on Rapid Rural Appraisal. Khon Kaen, Thailand: Rural
Systems Research and Farming Systems Research Projects.

McCracken, J.A., Pretty, J.N. and Conway, G.R. (1988)

Introduction to RRA for Agricultural Development. London:
International Institute for Environmental Development.

Nichols, P. (1991) Social survey methods: a fieldguide for

development workers. Development Guidelines, Vol. 6 Oxford:
Oxfam.

Roling, N .G. (1988) Extension Science: information systems in

agricultural development. Wye Studies in Agriculture and Rural
Development. Cambridge: Cambridge University Press.

Printed by Hobbs the Printers Ltd, Totton, Hampshire 5040 3Y5

52
 UGAND"

~
I