LSM Segmentation

THE LIVING STANDARDS MEASURE AS A MARKET
SEGMENTATION TOOL FOR SELECTED RETAILERS
Nedia de Jager
(B.COM., HONS. B.COM)
Dissertation submitted in partial fulfillment of the requirements for the degree of
MASTER OF COMMERCE
in the school of
ENTREPRENEURSHIP, MARKETING AND TOURlSlul MANAGEMENT,

FACULTY OF ECONOMIC AND MANAGEMENT SCIENCES
at the
POTCHEFSTROOM CAMPUS. NORTH-WEST UNIVERSITY
SUPERVISOR: Prof. L.R. Jansen van Rensburg

ASSISTANT SUPERVISOR: Me. C.F. de Meyer November, 2004
I would like to express my sincere gratiude to the under mentioned who have
made the successful completion of this study possible.
Most important, God Almighty for giving me the strength, courage and
determination to initiate this study and seeing it through to the end.
My best friend, Heleen, for her faith in me and her continuous

understanding and support to help me succeed.
My parents, Andries and Lorraine, who have always supported and loved
me and provided me with the opportunity to study.
Professor van Rensburg, my supervisor for his expertise, guidance,

understanding and constructive criticism.
Me. de Klerk, for her support and faith in me to complete this study.
Christine, my assistant supervisor for her friendship, encouragement and

belief in my abilities.
Dr. S. Ellis for her expertise and guidance in the construction and
statistical analysis of the questionnaires.
The School of Entrepreneurship, Marketing and Tourism Management,

and the North-West University for the use of the study facilities.
ABSTRACT
LIVING STANDARDS MEASURE AS A MARKET SEGMENTATION TOOL

FOR SELECTED RETAILERS
Businesses have realised that they cannot appeal to all consumers in the
marketplace. Consumers are too numerous, geographically wide-spread, and
too varied in their needs and buying practices. Furthermore, the business&
themselves vary widely in their abilities to serve different segments of the
market. Businesses consequently conduct market segmentation to identify the
parts of the market that it can best serve and most proftably. Therefore, the
goal of this study was to determine the actual target markets of selected retailers
by means of the Living Standards Measure (LSM) tool, and to recommend
improved marketing strategies based on the Media Groups Measure (MGM) tool.
The objectives of the study were to describe market segmentation by

emphasising its importance and discussing the market segmentation process.
Further objectives include explaining the application of the LSM- and MGM-tools
in identrfying target markets and selecting media to reach these target markets,
and to conduct an empirical investigation on the application of the LSM tool for
the selected retailers.
A market segmentation strategy concentrates the business' marketing efforts on

consumers who have a similar interest in the business' values. Therefore,
market segmentation enables the business to identii consumer needs more
accurately. The aforementioned assists the business in creating a basis for
developing products to satisfy consumer needs. The basis is developed by the
identification of the attributes that are of value to consumers.
Market segmentation furthermore assists in the development of an effective

marketing mix, tailored to the business' strategii and objectives. A proper
marketing mix can determine the promotional appeal, media choices, and the
prices consumers are willing to pay.
The LSM and MGM are only two of the variety of segmentation tools that can be
used to segment a heterogeneous market. The LSM is a relatively new and
unknown instrument - in comparison with other segmentation tools - and was
specifically developed for the South African market. It is furthermore a
multivariate segmentation tool, consisting of 29 household variables, which
excludes discriminators such as race and income. The MGM-tool is used to
obtain media consumption information on market segments. Its primary function
is to assist businesses in selecting media that are most effective in reaching their
target markets.
The most important conclusions of this study are that the current target markets
of Shoprite Checkers and Pick 'n Pay in Potchefstroom differs from their actual
target markets. Gender and age do not differentiate shoppers significantly, and
although race and income are not included as differentiators in the LSM index,
these two variables correlate strongly with the various LSM-groups.
Some of the most important recommendations include:
- The headquarters of both retailers should consider developing

differentiated marketing strategies for stores in different locations.
Shoprite Checkers and Pick 'n Pay (in Potchefstroom) should both target
SU-LSM groups 6 to 10.
Both retailers should adapt their marketing efforts in order to provide for
the local market in Potchefstroom.
The recommendations will enable the retailers to enhance their business and
marketing efforts, and accordingly satisfy the needs of their customers.
OPSOMMING
"LIVING STANDARDS MEASURE" AS 'N

MARKSEGMENTERINGSINSTRUMENTVIR GESELEKTEERDE HANDELAARS
Ondernemings het besef dat alle verbruikers in die mark se behoeftes nie
bevredig kan word nie. Verbruikers is te veel, wyd verspreid, en te verskillend in
hul behoeftes en aankooppatrone. Verder verskil ondememings self te veel in
hul verrndns om onderskeie segmente te bedien. Ondememings voer dus
marksegmentering uit, om sodoende die onderskeie dele van die mark wat hul
die beste kan bedien en die mees winsgewende mark te identiiseer.
Die doe1 met die studie was dus om te bepaal wat die eintlike teikenmarkte van
geselekteerde handelaars is, deur gebruik te maak van die "Living Standards
Measure" (LSM) instrument, en om verbeterde bemarkingstrategi& voor te stel
wat gebaseer is op die "Media Groups Measure" (MGM) instrument.
Die doehviie van die studie was om marksegmentering te omskryf deur die
belangrikheid daarvan te beklemtoon en die marksegmenteringsproses te
verduidelik. Verdere doehviie slul in die verduideliking van die toepassing van
die LSM- en MGM-instrumente in die identiisering van teikenmarkte en
selektering van media om hierdie teikenmarkte te bereik, asook om 'n empiriese
ondersoek te loods op die aanwending van die LSM-instrument op die
geselekteerde handelaars.
'n Marksegmenteringstrategiefokus die ondememing se bemarkingspogings op

verbruikers wat soortgelyke belangstelling toon in die ondememing se waardes.
Gevolglik stel marksegmentering die ondememing in staat om sy kliente se
behoeftes meer akkuraat te identifiseer. Voorgenoemde ondersteun die
onderneming om 'n basis vir die ontwikkeling van produkte vir die bevrediging
van verbruikersbehoeftes te ontwikkel. Die basis is ontwikkel deur die
identisering van die eienskappe wat waarde toevoeg vir die verbruikers.
Marksegmentering ondersteun ook verder die ontwikkeling van 'n effektiewe
bemarkingsmengsel wat aangepas word by die ondernemingstrategie(! en
doelwitte. 'n Goeie bemarkingsmengsel kan die promosionele trefkrag, media
keuses en pryse wat verbruikers gewillig is om te betaal vasstel.
Die LSM en MGM is slegs twee van 'n verskeidenheid segmenteringsinstrumente

wat aangewend kan word in die segmentering van 'n heterogene mark. Die
LSM is 'n relatiewe nuwe en onbekende instrument - in vergelyking met ander
segmenteringsinstrumente - en is spesifiek vir die Suid-Afrikaanse mark
ontwikkel. Di is ook 'n multiveranderlike segmenteringsinstrument wat uit 29
huishoudelike veranderlikes bestaan, en diskriminerende veranderlikes soos ras
en inkomste weglaat. Die MGM-instrument word weer gebruik om media
verbruikersinligting oor die marksegmente te verkry. Die primere funksie is om
ondememings te ondersteun in die selektering van media wat die teikenmarkte
die beste sal bereik.
Die mees belangrike gevolgtrekkings wat gemaak kan word, sluit in dat die
huidige teikenmarkte van Shoprite Checkers Checkers en Pick 'n Pay in
Potchefstroom verskil van hulle eintlike teikenmarkte. Geslag en ouderdom
onderskei kopers nie betekenisvol nie, en alhoewel ras en inkomste nie in die
LSM-indeks ingesluit word nie, hou die twee veranderlikes we1 sterk verband
met die onderskeie LSM-groepe.
Sommige van die belangrikste aanbevelings sluit in:
Die hoofkantore van albei handelaars moet dit oorweeg om

gedifferensieerde bemarkingstrategiee vir die verskillende handelaars in
verskillende areas te ontwikkel.
Beide Shoprite Checkers en Pick 'n Pay (in Potchefstroom) moet SU-
LSM-groepe 6 tot 10 teiken.
Albei handelaars moet hulle bemarkingstrategie aanpas om sodoende
die plaaslike mark in Potchefstroomte bereik.
Die aanbevelings sal die handelaars in staat stel om hul besigheids- en
bemarkingspogings te bevorder, en sodoende hul kliente se behoeftes te
bevredig.
TABLE OF CONTENTS
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix
LIST OF KEY TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xi
CHAPTER 1: INTRODUCTION. MOTIVATION. GOAL AND METHOD ....1

1.1 INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 KEYTERMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2.1 Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.2 Target Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.3 Market Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.4 Living Standards Measure . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.5 Media Groups Measure. . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 FOR THE STUDY. . . . . . . . . . . . . . . . . . . . . . . . 3
MOTIVATION
1.3.1 Reasons for Selecting the Topic. . . . . . . . . . . . . . . . . . . . . . . 4
1.3.2 The Need for the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 AIM OF THE STUDY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4.1 Goal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 OF THE FIELDOF STUDY. . . . . . . . . . . . . . . . . . 6
DEMARCATION
1.5.1 Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.2 Sample Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.3 Sampling Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 RESEARCH ......................... 8
METHODOLOGY
1.6.1 Literature Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6.2 Empirical Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.6.3 Research Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
a) Questionnaire design . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table of Contents i
b) Data analysis procedure . . . . . . . . . . . . . . . . . . . . . . . . 10
1.6.4 Statistical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.6.5 Research Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.6.6 Limitations of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.7 OF CHAPTERS. . . . . . . . . . . . . . . . . . . . . . . . . . 13
DIVISION
CHAPTER 2: MARKET SEGMENTATION . . . . . . . . . . . . . . . . . . . . 16

2.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
INTRODUCTION
2.2 BACKGROUND SEGMENTATION. . . . . . . . . . . . . . 16
ON MARKET
2.2.1 Market Segmentation Defined . . . . . . . . . . . . . . . . . . . . . . . 17
2.3 IMPORTANCE
OF MARKET . . . . . . . . . . . . . . . 18
SEGMENTATION
2.4 MARKET PROCESS. . . . . . . . . . . . . . . . . . . 19
SEGMENTATION
2.5 MARKETS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.5.1 Market Defined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.2 Characteristics of Markets . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.3 Types of Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.6 TYPESOF SEGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.6.1 Mass Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.6.2 Market Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.6.3 Niche Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.6.4 Individual Market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.7 BASESFOR MARKET . . . . . . . . . . . . . . . . . 26
SEGMENTATION
2.7.1 Geographic Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.7.2 Demographic Segmentation . . . . . . . . . . . . . . . . . . . . . . . . 28
2.7.3 Psychographic Segmentation . . . . . . . . . . . . . . . . . . . . . . . 28
2.7.4 Behavioural Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.7.5 Other Segmentation Bases . . . . . . . . . . . . . . . . . . . . . . . . 30
a) Beneft segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . 30
b) Response elasticity segmentation. . . . . . . . . . . . . . . . . . . 30
c) Usage-rate segmentation. . . . . . . . . . . . . . . . . . . . . . . . 30
d) Living Standards Measure . . . . . . . . . . . . . . . . . . . . . . . 31
Table of Contents ii
e) Media Groups Measure. . . . . . . . . . . . . . . . . . . . . . . . . 31
SEGMENTATION . . . . . . . . . . . . . . . . . . . . . . 31
DESCRIPTORS
EVALUATING . . . . . . . . . . . . . . . . . . . . . . . . . . 32
SEGMENTS
SUCCESSFUL
MARKET . . . . . . . . . . . . . . . . . 33
SEGMENTATION
Basic Criteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Rules to Market Segmentation. . . . . . . . . . . . . . . . . . . . . . . 34
Characteristics of Market Segmentation . . . . . . . . . . . . . . . . . 34
Advantages of Market Segmentation . . . . . . . . . . . . . . . . . . . . 35
Disadvantages of Market Segmentation . . . . . . . . . . . . . . . . . 36
SELECTION. . . . . . . . . . . . . . . . . . . . . . . 37
TARGETMARKET
Evaluating Market Segments . . . . . . . . . . . . . . . . . . . . . . . 37
Target Market Strategies . . . . . . . . . . . . . . . . . . . . . . . . . 38
a) Undifferentiated targeting. . . . . . . . . . . . . . . . . . . . . . . . 38
b) Differentiated targeting . . . . . . . . . . . . . . . . . . . . . . . . . 39
c) Concentrated targeting . . . . . . . . . . . . . . . . . . . . . . . . . 40
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
POSITIONING
Positioning Strategies. . . . . . . . . . . . . . . . . . . . . . . . . 41
Characteristics of Positioning Dierences . . . . . . . . . . . . . . 43
MARKETMIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Marketing Mix Clarification . . . . . . . . . . . . . . . . . . . . . . 44
Market Mix Elements . . . . . . . . . . . . . . . . . . . . . . . . . 45
a)Product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
b) Price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
c) Distribution (place) . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
d) Promotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Marketing Mix Characteristics . . . . . . . . . . . . . . . . . . . . . . . 47
Marketing Mix Gaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
SUMMARY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Table of Contents iii

CHAPTER 3: LIVING STANDARDS MEASURE AND MEDIA GROUPS
MEASURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.1 ...............................
INTRODUCTION 50
3.2 SOUTH AFRICAN (SAARF) . 51
FOUNDATION
ADVERTISINGRESEARCH
3.3 THE NATUREOF THE LSM . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.3.1 LSM Detined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.3.2 History and Development of the LSM . . . . . . . . . . . . . . . . . . . 53
3.4 SU-LSM CLARIFIED . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.4.1 SU-LSM Compilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.4.2 Calculating the SU-LSM . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.4.3 SU-LSM Scale and Society . . . . . . . . . . . . . . . . . . . . . . . . 62
3.4.4 SU-LSM and Race . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.5 SAARF MEDIA GROUPSMEASURE(MGM) . . . . . . . . . . . . . . . 66
3.5.1 MGM Defined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.5.2 Analysis of MGM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.5.3 MGMs Scale and Soctety. . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.5.4 SU-LSM versus MGM. . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.6 BACKGROUND RETAILERS. . . . . . . . . . . . . . 76
ON SELECTED
3.6.1 Shoprite Checkers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.6.2 Pick 'n Pay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.6.3 SU-LSM-Groups Summarised . . . . . . . . . . . . . . . . . . . . . . . 77
a) SU-LSM-group 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
b) SU-LSM-gfoup 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
c) SU-LSM-group 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
d) SU-LSM-group 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
e) SU-LSM-group 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
f) SU-LSM-group 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
3.7 SUMMARY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Table of Contents
CHAPTER 4: EMPIRICAL METHODOLOGY AND EXPOSITION . . . . . . 85
4.1 ~NTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.2 EMPIRICAL . . . . . . . . . . . . . . . . . . . . . . 85
METHODOLOGY
4.2.1 Study Population and Sampling . . . . . . . . . . . . . . . . . . . . . 85
4.2.2 Questionnaire Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
4.2.3 Statistical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
a) Cronbach coefficient alpha analysis. . . . . . . . . . . . . . . . . . 88
b) Frequency analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
c) Multi-way frequency analysis . . . . . . . . . . . . . . . . . . . . . 89
d) Correspondence analysis . . . . . . . . . . . . . . . . . . . . . . . 89
4.2.4 Reliability of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.3 EMPIRICALEXPOSITION. . . . . . . . . . . . . . . . . . . . . . . . 90
4.3.1 Demographic Information. . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.3.2 Living Standards Information. . . . . . . . . . . . . . . . . . . . . . . . 97
4.4 SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
CHAPTER 5: SUMMARY. CONCLUSIONS AND RECOMMENDATIONS 106 ..

5.1 SUMMARY
AND OVERVIEW . . . . . . . . . . . . . . . .106
OF THE STUDY
5.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . .I 09
CONCLUSIONS
5.2.1 Overview of Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.2.2 SU-LSM-Groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.2.3 Pick 'n Pay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.2.4 Shoprite Checkers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5.3 . . . . . . . . . . . . . . . . . . . . . . . . . . 113
RECOMMENDATIONS
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119
APPENDIX A: QUESTIONNAIRE DISTRIBUTION SCHEDULE . . . . . . .127
APPENDIX 6: QUESTIONNAIRE . . . . . . . . . . . . . . . . . . . . . . . . .128
Table of Contents
APPENDIX C: SUMMARY OF THE SU-LSM CLASSIFICATION . . . . . . .I30
APPENDIX D: SAARF SUlSM CONCLUSIONS . . . . . . . . . . . . . . . .I35
Table of Contents vi
LIST OF FIGURES
CHAPTER 1: INTRODUCTION. MOTIVATION. GOAL AND METHOD ... I

Figure 1-1 Chapter division . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Figure 2-1 Market segmentation process . . . . . . . . . . . . . . . . . . . 20
Figure 2-2 Levels of market segments. . . . . . . . . . . . . . . . . . . . . 23
Figure 2-3 Undifferentiated targeting. . . . . . . . . . . . . . . . . . . . . . 39
Figure 2 4 Differentiated targeting . . . . . . . . . . . . . . . . . . . . . . . 39
Figure 2-5 Concentrated targeting . . . . . . . . . . . . . . . . . . . . . . . 40
Figure 2-6 The marketing mix. . . . . . . . . . . . . . . . . . . . . . . . . . 46

MEASURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Figure 3-1 The nature of SAARF . . . . . . . . . . . . . . . . . . . . . . . . 51
Figure 3-2 LSM timeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Figure 3-3 Education .10 year trends. . . . . . . . . . . . . . . . . . . . . 63
Figure 3 4 SU-LSM population distribution . . . . . . . . . . . . . . . . . . 64
Figure 3-5 North-West Province population percentage. . . . . . . . . . . 65
Figure 3-6 SU-LSM race profile for 2002.A . . . . . . . . . . . . . . . . . . 66
Figure 3-7 Media coverage of MGM-group 1 . . . . . . . . . . . . . . . . . 68
Figure 3-12 Media coverage of MGMgroup 6 . . . . . . . . . . . . . . . . . 71
Figure 3-15 Age profile of the MGM-scale . . . . . . . . . . . . . . . . . . . 73
Figure 3-16 Population group profile of the MGMscale. . . . . . . . . . . . 73
Figure 3-17 Monthly household income profile of the MGM-scale . . . . . . 74
Table of Contents vii

Figure 3-18 North-West Province media exposure . . . . . . . . . . . . . . 75

Figure 4-1 Correspondence Analysis: The similarity between the
respondents' population group and choice of store . . . . . . . 94
respondents' monthly income and choice of store . . . . . . . 96
respondents of Pick 'n Pay and Shoprite Checkers's SU-
LSM-groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Figure 4 4 Correspondence Analysis: The similarity between the
respondents' SU-LSM-groups and their population group . . .I01
Figure 4-5 Correspondence Analysis: The similarw between the
respondents' SU-LSM-groups and their education levels . . .I00
respondents' SU-LSM-groups and their monthly income. . . .I02
Table of Contents viii

LIST OF TABLES

Table 2-1 Bases for market segmentation . . . . . . . . . . . . . . . . . . 27

MEASURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Table 3-1 Original LSM variables . . . . . . . . . . . . . . . . . . . . . . . 55
Table 3-2 1993 SAARF LSMs . . . . . . . . . . . . . . . . . . . . . . . . . 55
Table 3-3 1995 SAARF LSMs . . . . . . . . . . . . . . . . . . . . . . . . . 56
Table 3 4 2000 SAARF LSMs . . . . . . . . . . . . . . . . . . . . . . . . . 57
Table 3-5 SAARF Universal LSM . . . . . . . . . . . . . . . . . . . . . . . 59
Table 3-6 SU-LSM variables and weights . . . . . . . . . . . . . . . . . . 61
Table 3-7 Range for each SU-LSM-group . . . . . . . . . . . . . . . . . 62
Table 3 8 Average household income by SU-LSM . . . . . . . . . . . . . 63
Table 3-9 Employment status (Januaty .December 2002) . . . 64
Table 3-10 SU-LSM profile of the MGM.scale . . . . . . . . . . . . . . . . . 75

Table 4-1 Amount of respondents . . . . . . . . . . . . . . . . . . . . . 87
Table 4-2 Cronbach's coefficient alpha of variables. . . . . . . . . . . . . 91
Table 4-3 Province of residence distribution for respondents . . . . . . . 92
Table 4 4 Gender distribution of respondents . . . . . . . . . . . . . . . . 93
Table 4-5 Population group distribution of respondents . . . . . . . . . . 93
Table 4-6 Education distribution of respondents. . . . . . . . . . . . . . . 95
Table 4-7 Age distribution of respondents . . . . . . . . . . . . . . . . . . 95
Table 4 8 Monthly income of respondents . . . . . . . . . . . . . . . 96
Table 4-9 Contingency: The similarity between the respondents of
Pick 'n Pay and Shoprite Checker's SU-LSM-groups . . 97
Table 4-10 Contingency: The similarity between the respondents'
SU-LSM-groups and their province of residence . 99
Table of Contents ix
SU-LSM-groups and their gender. . . . . . . . . . . . . . . . .I00
SU-LSM-groups and their population group . . . . . . . . . . ,100
SU-LSMgroups and their education levels . . . . . . . . . . ,102
SU-LSM-groups and their age. . . . . . . . . . . . . . . . . . .I
03
SU-LSM-groups and their monthly income. . . . . . . . . . . ,104
CHAPTER 5: SUMMARY, CONCLUSIONS AND RECOMMENDATIONS. .I06

Table 5-1 General household possessions for SU-LSM-groups 6 to 10. .I
15
Table 5-2 Media for SU-LSM-groups 6 to 10. . . . . . . . . . . . . . . . .I16
Table of Contents x
- - -
LIST OF KEY TERMS
Market
Target market
Market segmentation
Living Standards Measure
Media Groups Measure
Table of Contents xi
CHAPTER 1
INTRODUCTION, MOTIVATION, GOAL AND METHOD
The study of market segmentation is an intricate one, frequently raising many

questions not easily answered. Several studies attempted to determine the
relationship between particular consumer variables and consumers' purchasing
behaviour. Too often, however, some of these studies are disappointing, mainly
because they only reveal that these variables are redundant in explaining
consumers' behaviour (Levy, 1999:297).
Marketers soon realised that it would be practically impossible to satisfy every

customer or consumer, since each individual has different needs, preferences,
resources and behaviours (Recklies, 2001). Given that all people differ in their
requirements, each individual can be seen - taken to its extreme - as a unique
market segment (McDonald, 2000:81).
The change in demand patterns, encouraged by the aspiration for higher living
standards and individualism, originates from the fact that consumers are
becoming more demanding in exercising their choice (Lancaster & Reynolds,
1998:27). It is mainly these differences and changes that brought about the
need to segment large diverse markets (Levy, 1999:297).
Some terms that are frequently used throughout this study need clear
explanation. These key terms include the terms market, target market, market
segmentation, Living Standards Measure (LSM) and Media Groups Measure
(MGM).
Chapter 1: Introduction, Motivation, Goal and Method 1

1.2.1 Market
A market can be described as a group of customers with specific needs or

wants, which has the ability (income) and authority to purchase these needs or
wants (Lamb, Hair & McDaniel, 2000:212). A market can furthermore consist of
either consumers (individuals) or businesses (Nickels and Wood, 1997:192).
For the purpose of this study, the focus will be on the consumer market.
1.2.2 Target Market
A target market consists of a group of individuals (or businesses) with similar

characteristics, for which a marketing mix can be designed, implemented, and
maintained in order to meet the needs of that group (Lamb et a/., 2000:227).
1.2.3 Market Segmentation
Market segmentation entails the process of dividing a large, diverse market into
smaller groups (segments) that are similar in character (Gilbert, 1999:201).
1.2.4 Living Standards Measure
The Living Standards Measure (LSM) is a market segmentatin tool that

segments an individual or market, based on their socioeconomic status
(SAARF, 2004). Criteria such as degree of urbanisation, possession of major
assets and access to services are some of the variables used in the LSM
(Haupt, 2001a).
1.2.5 Media Groups Measure
The Media Groups Measure (MGM) is a tool designed to obtain media

consumption information on selected market segments. It furthermore assists
businesses in selecting media that are most effective in reaching their target
-

markets (SAARF, 2002b). Readers, listeners, viewers, cinemagoers, lntemet
users and those exposed to outdoor media, can be identified and profiled by
means of the MGM (SAARF, 2003).
1.3 MOTIVATION
FOR THE STUDY
The research topic, reasons for its selection and need for the study emphasise
the motivation for this study.
The focus of this study is to identify the target markets of selected retailers in
Potchefstroom, as well as to determine the media these stores can use to target
their related target markets. The target markets were identified by means of the
South African Advertising Research Foundation Universal Living Standards
Measure ( W R F Universal LSM), and the types of media, by means of the
Media Groups Measure (MGM).
The aforementioned formed a basis on which could be determined whether

these stores are targeting the correct market and whether they are making use
of the appropriate marketing activities for the related target markets. It was also
possible to compare the different stores' targeted LSM-groups.
Market segmentation, according to Rix (2004:93), is the process of dividing a

large heterogeneous market into smaller, relatively homogeneous segments.
Segmenting a market is of utmost importance for any type or kind of business. If
a business is able to divide a large market into smaller segments with different
preferences and adjust its product accordingly, the business has the opportunlty
to increase its competitive position (McDonald, 2000:81).
The SAARF Universal LSM (SU-LSM) is a market segmentation tool, assisting

businesses in directing their marketing activities. As with any other market
segmentation tool, the LSM was designed to divide a large market into smaller,
relatively homogeneous groups (Anon, 2002). The major difference, however, of
- - -
Chap6er 1: Introduction, Motivation, Goal and Method 3

this specific segmentation tool, is that it is based on a set of marketing
differentiators which groups people according to their living standards rather
than their income. The LSM is therefore a wealth measure using criteria such as
degree of urbanisation and ownership of major assets (SAARF, 2001b).
The population framework used for this study includes selected retailers in
Potchefstroom. Two major chain supermarkets, Pick 'n Pay and Shoprite
Checkers, form the basis of the research. Due to the large amount of
supermarkets, including local and chain supermarkets in Potchefstroom, only
thesetwo specific supermarkets were included in the framework.
Recommendations could, therefore, only be made to these participating stores
based on the research conducted.
1.3.1 Reasons for Selecting the Topic
The LSM as a market segmentation tool was selected for this study, mainly
because it is a relatively unknown instrument for Potchefstroom businesses.
However, according to Haupt (2001a), it has become one of the most widely
used market research tools in South Africa. Seeing that the LSM cuts across
obsolete methods of categorising people - such as race - the LSM became a
distinctive means of segmenting the South African market.
As mentioned, the LSM categorises people according to their living standards

rather than their income, and since this is a multivariate segmentation tool -
constructed from twenty-nine (29) individual variables - it is a much stronger
differentiator than any other single segmentation tool (SAARF, 2002b).
Market segmentation was selected as a study field, since it is evident that it is

becoming essential to divide such a large and diverse market as the South
African consumers into smaller and relatively similar groups. No single product
and associated marketing program can effectively serve an entire market (Kotler
Chapter 1: Introduction, Motivation, Goal and Method
-
& Armstrong, 2004:238), especially a heterogeneous market such as South
Africa.
Pick 'n Pay and Shoprite Checkers were chosen as research framework, since
these specific stores' formats are basically similar. Both stores are major South
African chain supermarkets, providing relatively similar types of food- and non-
food products (Terblanchb, 1998:121). They furthermore form the largest
supermarkets in Potchefstroom, and are both located in the CBD (central
business district) of this city. It was therefore important to identify whether these
two stores can be compared in terms of their customers, and whether their
marketing activitii are directed towards the appropriate segments.
1.3.2 The Need for the Study
The reason for undertaking the study is to determine whether Pick 'n Pay and
Shoprite Checkers are currently focussing on the appropriate target markets,
and whether the current media these stores use ffi the LSM-groups that actually
purchase from them. Finally, recommendations have been made to the
participating stores, in order to assist them in directing their marketing activities.
1.4 AIM OF THE STUDY
The following overall goal and objectives clanfy the aim of the study:
1.4.1 Goal
The overall goal of this study is to determine the actual target markets of
selected retailers in Potchefstroom by means of the LSM-tool, and to
recommend improved marketing strategies based on the MGM-tool.
Chapter I:Introduction, Motivation, Goal and Method

1.4.2 Objectives
To achieve the above aim, the objectives of the study are to:
- describe market segmentation by emphasising its importance and

discussing the market segmentation process
explain the application of the LSM- and MGM-tools in identifying target
markets and selecting media to reach these target markets
= conduct an empirical investigation on the application of the LSM-tool in
the selected retailers
1.5 DEMARCATION
OF THE FIELDOF STUDY
To achieve the goal and objectives of the study, the population, sample size and
sampling method needed to be determined.
1.5.1 Population
A population can be defined as the total group of elements (people or items) on

which research is directed (Blankenship, Breen & Dutka, 1998:152). The
population for this study is the customers of Pick 'n Pay and Shoprite Checkers
in Potchefstroom. However, due to Shoprite Checkers's agreement with its own
market researchers, the store did not want to disclose its retailing habits to this
study's researcher. Therefore, the precise number of this store's customers was
not available to the researcher, which consequently influenced the process of
determining the sample size for both stores.
1.5.2 Sample Size
A formula provided by the Statistical Consuttation Services (SCS) of the

Potchefstroom Campus of the North-West Universrty was used to obtain a valid
sample size. A sample, according to Berenson & Levine (1996:3), is the portion

of the population that is selected for analysis. The following formula can be used
to determine the size of a relationship sample, as in this specific instance.
Formula (Steyn, Sml, Du Toit & Strasheim, 1998:397):
However, as mentioned, the exact size of the population is not known, which
has the following impact on the formula. According to the SCS, the following
can be assumed:
'If the proportion (p) is unknown to the researcher, then p = %. '
The following formula was therefore used for this study:
In - sample size
II
1 e - deviation %
A 5% deviation was allowed, thus meaning that the information obtained from
the sample is 95% accurate of the entire population.
n - sample size I
5% - deviation % I
The sample size for this study should accordingly be a minimum of 400
customers for each store, thus 800 respondents in total.

1.5.3 Sampling Method
Due to Shoprite Checkers's inability to provide information on its retailing habits,

such as number of transactions per day or per month, no specific sampling
method could be used. Customers were randomly stopped at the respective
stores, and asked to complete the short questionnaire.
Questioning respondents on different days of the week, and at different times of

the day ensured the validity and randomness of this study. Therefore, to ensure
that all the customers had a chance to form part of the study, all the business
hours of the two stores have been covered.
The empirical study furthermore took place over a period of three weeks to
ensure the coverage of a large part of the diverse amount of customers, and to
make the study more viable. The days and times on which the customers were
questioned are included in Appendix A.
The Internet enables fast and economic access to a vast variety of data sources
and therefore played a major role in collecting secondary data (Bevan-Dye,
2002:5). Secondary sources were used to provide background information on
the specific topic of research.
I.6.1 Literature Study
According to Struwig and Stead (2001:38), a literature study involves the tracing,
identifying and analysing of documents that contain information related to the
specific research topic.
Literature on the background of market segmentation and the market

segmentation process was obtained from secondary sources such as books,

published articles, journals and the lntemet Non-published material relevant to
the study was gathered by means of questionnaires directed at the selected
sample.
A literature study on the nature of SAARF and the development of the LSM tool
was conducted. The functioning of the SU-LSM, as well as the role of the MGM
has been examined. In comparison with other segmentation methods, this
specific topic of research is relatively new and unknown, and secondary sources
on the LSM and MGM were therefore quite limited. Information on the
aforementioned was thus mainly obtained from the lntemet, the SAARF
handbook and -compact disk.
1.6.2 Empirical Research
lnformation was gathered by means of an empirical study, which according to

Struwig and Stead (2001:222) involves the obtaining of data by employing
questionnaires or conducting interviews.
A questionnaire was compiled for the customers of the selected stores to obtain
information regarding their living standards. Fieldworkers - which were placed
outside the entrances of the selected stores - questioned the respondents upon
exiting the store. The fieldworkers consisted of three Baccalaureus Commercii
students of the North-West Universtty (Potchefstroom Campus), which were pre-
informed on the study, its purposes and the workings of the questionnaire to be
completed. The researcher was present during each of the surveying sessions,
in order to ensure no misunderstandingsoccurred.
The questionnaire consisted of:
Multiple-choice questions, which provide a fixed number of answers, to

which respondents must select one or more options (Du Plooy,
2001:142).
Chapter I:Introduction, Motivation, Goal and Method 9

Dichotomous questions, which offer a choice between two options only
(Struwig & Stead, 2001:94).
lnformation on the selected stores' target markets and marketing activities were
obtained from their respective headquarters. lnformation that could not be
obtained from the headquarters was obtained by means of the Internet.
1.6.3 Research Design
Due to the exploratory nature of this study, a hypothesis was not formulated.
The questionnaires sewed as a basis for further research.
a) Questionnaire design
Since respondents were stopped for questioning during their shopping
process, the questionnaires needed be kept relatively short. The
compilation of the questionnaire was therefore verified by SCS to ensure
that all the required information and accurate, meaningful statistical results
could be obtained from the questionnaire. The questionnaire therefore only
consists of two sections - see Appendix B:
The first section gathered demographic information of respondents.

This included demographics such as area of residence, population
group, education, age, gender and monthly income. Multiple-choice
questions were used to gather this information.
The second part concentrates on the respondents' living standards.

The LSM-variables were applied as dichotomous questions in this
section.
b) Data analysis procedure

All the questionnaires were numbered for statistical purposes. After
completion of the questionnaires, the data was entered into the Epi Info 6

(Version 6.02 - October 1994) programme. The processed data was then
provided to the Statistical Consultation Sewices (SCS) of the North-West
University (Potchefstroom Campus) for further processing by means of the
SAS system (SAS Institute lnc., 1989). The SCS analysed the
questionnaires to ensure accuracy and correct interpretation of the results.
1.6.4 Statistical Analysis
The Cronbach's coefficient alpha analysis was conducted to determine the

reliability of the testing method used in this study. The average correlation of the
29 SU-LSM variables was determined to ensure that the average alpha
coefficient was large enough for the study to be reliable.
A frequency analysis was carried out to determine the strength of each of the
questions asked. This analysis consisted of the frequency, percentage,
cumulative frequency and cumulative percentage of the respondents' answers.
A relationship analysis was done between the variables of the two participating
stores (Shoprite Checkers and Pick 'n Pay), to determine the comparison of
these two stores' customers. In order to test the significance of these
relationships, an effect size needed be calculated by means of Chi-square tests.
Data with a large effect (w 2 0,5) is considered to be practically significant
(Cohen, 1988:109-111).
The multi-way frequency tables (relationship analysis) were further analysed by

means of a correspondence analysis. That is, plotting the row and column
categories of a multi-way frequency table on one diagram. Row and column
categories that have similar profiles will appear in close proximity on the plot. A
value close to zero on the plot indicates no interaction (association) between the
row and column categories. These interpretations can only be in relative terms,
and not in an absolute sense (Bartholomew, Steele, Moustaki & Galbraith,
2002:QO).

1.6.5 Research Procedure
After determining the research topic, the population and the sample size, the
next step was to gather as much information as possible on the subject, for a
complete literature study. After sufficient gathering of literature background, the
questionnaires were compiled in such a way that the study's objectives could be
achieved.
The questionnaire was evaluated and pretested until of high standard and
presentable. The questionnaires were then distributed in front of the main
entrance of the relevant stores to customers exiting the store. In total, 980
questionnaires (490 for each store) were completed. After completion of the
questionnaires, the data was entered into the Epi Info 6 programme and stored
on a diskette, which in turn was provided to SCS for statistical analysis.
The analysed information was then evaluated in order to make conclusions and
recommendations. The objectives of the study, literature information and the
empirical information obtained, determined the nature of the recommendations
made.
The information on the stores' current target markets and marketing activities
were obtained by means of the Internet (the retailers' homepage). Some
information was not available publicly and was acquired by contacting the
respective retailers' headquarters.
1.6.6 Limitations of the Study
A number of obstacles influenced the even course of events throughout the study:
1) Shoprite Checkers refused approval of questioning their customers in

their store.
- ---

2) The gathering of the empirical information was relatively time consuming,
due to the large sample that had to be obtained.
3) It was difficult to convince the respondents to be questioned, especially
during the busy times of the day.
4) It was difficult to find literature on the LSM- and MGM-tools, since these
topics are relatively new and sources are limited.
The abovementioned limitations were overcome by:
1) Surveying Shoprite Checkers's customers, on exiting the store, outside of

the store's entrance.
2) Making use of four fieldworkers to assist in the questioning of the
customers, and developing a carefully planned schedule to optimise the
available surveying time (see Appendix A).
3) Maintaining a friendly, persistent attiiude towards potential respondents.
4) Examining every possible source that contained information on the LSM-
and MGM-tools.
The study is divided into five chapters (see Figure 1-1). Chapters two and three
consist of a literature study on market segmentation and the LSM. The empirical
methodology and exposition are integrated into chapter four and ,concluded in
chapter five.
Chapter 1 provides a general scope of the entire study and explains the
procedures to be followed. It includes: an introduction, motivation
for the study, problem definition and objectives, more about the
questionnaire itself and method of sampling used.

Figure 1-1 Chapter division
Chapter 1
Introduction,Motivation,Goaland
Method
Chapter 2 Chapter 3 Chapter 4

Market Uving Standards Empirical
Measure and Media Methodology
Segmentation
GrouDs Measure and J;xposition
Chapter 5
Conclusions,Recommendations
and Summary
Chapter 2 defines market segmentation; outlines the importance of

segmenting and provides a detailed explanation of the different
steps in the market segmentation process.
Chapter 3 describes the nature of SAARF and why LSM was developed. The
development of LSM over the years is also discussed, with a more
in-depth look on the Universal LSM classification method. The
MGM-tool is explained to provide a background on the selection of
the most appropriate media for a specific target market.
Chapter 4 outlines the methodology of the empirical study and gives an

exposition of the empirical information. The questionnaire design is
explained in detail, containing information on the sample design,
sample size, and lastly the processing, analysing and evaluation of
the data. The empirical information on the LSM-categories of the
respondents is analysed.
Chapter 1: Introduction, Motivation,Goal and Method 14

Chapter 5 conclusions regarding the respondents and recommendations to
participating retailers in Potchefstroom are made in this chapter.
Finally, a summary of the entire study is provided.
Chapter I:Introduction, Motivation, Goal and Method 15

CHAPTER 2
MARKET SEGMENTATION
Customers form the most important aspect of any business. If a business is

unable to identii its customers, or what they prefer to buy, the business is
unlikely to survive (Newman & Cullen, 2002:121). Therefore, the basic
significance of the marketing concept is to place customers' needs at the centre
of the business' activities and decision-making (Lancaster & Reynolds, 1998:27).
In order for a business to plan its marketing strategy, it is important to identify its
customers and their purchasing behaviour (Newman & Cullen, 2002:121). The
heterogeneous nature of customers makes the aforementioned a challenging
task (Lancaster & Reynolds, 1998:27; Bagoui, Rosa, Celly & Coronel
1998:175). It is therefore necessaty to break this heterogeneous group of
customers down into meaningful groups that are large enough; and willing to
spend money on their specific needs or wants (Newrnan & Cullen, 2002:121).
To understand market segmentation, it is necessary to: define market

segmentation; distinguish between needs and wants; explain the market
segmentation process' sequential relationship with the rest of the chapter; define
markets and the different types of markets; indicate the various levels of market
segments; identii bases on which markets can be segmented; identify target
market strategies; clarify the meaning of positioning; and design an effective
marketing mix.
Market segmentation was first acknowledged in the 1950s, when product

differentiation was seen as the most popular marketing strategy. During the
Chapter 2: Market Segmentation
-
1970s and 1980s, market segmentation unintentionally commenced as a way of
increasing sales and obtaining a competitive advantage. In the 1990s, target
marketers discovered more sophisticated techniques to reach possible buyers
with the most customised offering possible, of which included market
segmentation (DSS Research, 2004).
Most businesses soon realised they cannot effectively serve all the segments in
a market, and must instead target their marketing efforts to a particular customer
group (Lancaster & Reynolds, 1998:28). Allen (1999) noted that the selection of
the most desirable target market presently forms an important part of planning a
marketing strategy. The marketing concept emphasises that marketing will be
more effective if it is customised according to the unique needs of each target
market.
2.2.1 Market Segmentation Defined
Since market segmentation plays an essential part in this study, it is imperative

to extensively define the term 'market segmentation' before comprehensively
examining its process and workings.
Market segmentation can be defined as the process of dividing a market into

meaningful, relatively similar. and identiable groups (Lamb et a/.,2000:212).
Market segmentation can also be seen as the process of breaking down the
total market, for a product or service, into different segments where each
segment may represent a separate target market to be reached with a unique
marketing mix (Lancaster & Reynolds, 1998:27).
According to Gilbert (1999:201), market segmentation entails the breaking

down of a large diverse market into smaller subsections in which there is
some similarrty of character.
- -
Chapter 2: Market Segmentation 17

In this study, market segmentation is defined as the process of dividing a large
heterogeneous market into smaller groups of customers that are internally
homogeneous, but distinct from each other (Bagoui et aL, 1998:157). These
groups can each be seen as a distinct target market that can be reached with its
own exclusive marketing mix.
Sometimes the differences in customers' needs, wants and behaviours are

rather small, and the same marketing mix can be used to satisfy these
customers. However, in most markets customers are reluctant to compromise
on the same marketing mix, and separate marketing mixes need to be designed
(Rix, 200493).
The terms 'needs' and 'wants' are generally used interchangeably, and are often
accepted to mean the same. Therefore, before attempting to proceed with the
clarification of markets and market segmentation, it is necessary to be able to
distinguish between customers' needs and wants.
Rix (20045) defines customer needs as the basic feeling of dispossession.

Thus, a need is a basic requirement that a customer desires to satisfy (Anon,
2004a).
Customer wants, on the other hand, can be seen as a desire for a specific
product or service that satisfies an underlying need (Anon, 2004a). 'Wants' is
thus a specific form that a need might adopt (Rix, 2004:5).
Market segmentation is a technique that is commonly accepted as one of the

prime pre-requisites for successful marketing (WestGroup Research, 2004).
Bagoui et al. (1998:175) notes that the underlying objective of market
segmentation is to improve marketing performance from what iwas in the past.
- - - -

Market segments vary in size and potential, marketers therefore make use of
segmentation to accurately define the business' marketing objectives and to
better allocate resources. With clear objectives, a business can more accurately
evaluate its performance, which in the end leads to increased efficiency (Lamb
et a/.,2000:213).
In addition, marketers can now more carefully analyse the wants, needs and
behaviours of potential customers, which provides the opportunity to develop
products and marketing programmes aimed at these specific needs and wants
(Nickels & Wood, 1997,221).
The market segmentation process consists of six distinct steps. However, steps
five and six follow market segmentation and are in fact marketing activities that
form the basis on which decisions about the product, its pricing, distribution and
promotion are made (Lamb eta/.,2000:226).
Although market segmentation is the focus of this study, it is necessary to look

into the steps preceding and following the actual segmentation procedure.
Figure 2-1 illustrates the sequential relationship between the various steps of the
market segmentation process and its ultimate contribution to the rest of the
chapter.
Marketers need to keep in mind that markets are dynamic in nature, meaning
that an entire market is subject to change. Since people's needs and wants can
change abruptly, segments within a market are also subject to changes in
growth (Allen, 1999). Thus, even though a business follows all the steps in the
market segmentation process, it should also be able to respond to environmental
changes that may have an influence (Nickels &Wood, 1997:225).

1
Figure 2-1 Market segmentation process
STEP 1 2.5 Markets

2.6 Levels of market
Select market to be studied
segments
STEP 2 2.7 Bases for market

Choose basis for segmentation segmentation
STEP 3 2.8 Segmentation

Select segmentation descriptors descriptors
2.9 Evaluate segments

STEP4 2.10 Successful market
Profile &evaluate segments segmentation
2.11 Target market

STEP 5 selection
Select target markets 2.12 Positioning
f
STEP6
2.13 Marketingmix
Design, implement and
maintain marketing mixes
Source: Adapted from Lamb et al. (2000:227).
2.5 MARKETS
One of the primary problems marketers are faced with when attempting to clarify
market segmentation is to determine what is meant with a 'market' and of whom
a market consists of (Monash University, 2004). Before a business can start
segmenting a market, it evidently has to be able to clearly define its market.

2.5.1 Market Defined
Marketers generally define a market as including all the actual and potential
buyers of a product or service. Many potential buyers, however, only purchase
the product or service several years after the product is launched (Monash
University, 2004).
A market can consequently be defined as a group of customers with specific

needs or wants, and have the ability, income and authority to buy the
specified needs or wants (Lamb et a/., 2000:212).
Defining a market is one of the first steps that need to be taken, since this is the
starting point for market segmentation, as well as the formulation of a marketing
strategy. An exact market definition can aid a business to narrow down the
market by eliminating segments that would not buy its product, regardless of the
benefits it offers (Bagoui et a/., 1998:177).
It is often assumed that countless individuals comprising the market are waiting
and willing to respond to a business' offerings. However, the 'market' is not a
single, consistent unit; it consists of contrasting, developing, and considerably
diverse human beings. Each one of these individuals lives in different
circumstances than another (Yankelovich, 1999).
2.5.2 Characteristics of Markets
Lamb et a/. (2000:212) identied the following general characteristics that

markets must have:
Markets must consist of people or businesses.

These people or businesses must have wants or needs that can be
satisfied by specific product or service offerings.

They must have the abiltty (usually money) to purchase the products or
services they desire.
They must be willing to trade their resources (usually money) for the
desired products or se~ices.
However, to identify a specific market can be quite challenging, since a market

might range from a single buyer to billions of buyers, concentrated or spread all
over the world. A business thus needs to take its resources, as well as
strategies, into account when making decisions on defining its market (Monash
University, 2004).
2.5.3 Types of Markets
Markets can consist of either consumers or businesses.
Consumer market
A consumer market consists of final consumers (individuals) and Ior
households that purchase products or services for personal use (Nickels
&Wood, 1997:192).
Business market
The business market consists of producers, resellers or governments that
buy products for the purpose of producing products, reselling to other
customers, or the carrying out of governmental functions (lownley, 2001).
Due to the nature of this study, research will focus only on consumer markets,
more specifically, consumer markets in the retailing industry.
To simplify the identication of a business' potential customers, the business

should decide whether it wants or needs to segment a market (Allen, 1999).
-
Various levels exist on which a business can segment a market. Figure 2-2
illustrates these levels, starting out with the largest segment (mass market), and
proceeding on to the smallest segment (individual market).
Figure 2-2 Levels of market segments
Mass Market
Market Segment
Niche Market
Individual
Market
Source: Thompson (2000).
2.6.1 Mass Markets
A business that prefers not to segment markets chooses a mass-market

approach, where the same product or service is offered to all customers. Thus,
marketers either assume that all customers desire the same product or service
or they deliberately ignore the customers' differences (Allen, 1999). With this
approach, only one marketing mix is used to aim at the entire consumer market,
and is often referred to as the 'one size fits all' marketing approach (Nickels &
Wood, 1997:220).
Production-, advertising- and marketing research costs are considerably lower

with a mass market approach, since it consists of only one product offering
(Brink, 1998:66). Although cost per unit may be lower, the overall costs are
higher with this approach, since the entire market needs to be targeted. Mass

marketing is also unable to meet the different consumer needs and wants
effectively (Deveau, 2004).
Mass marketing is thus generally more successful when competition is limited or

the product has mass appeal (Bull & Passewitz, 2003).
The market segmentation process commences by defining the overall mass

market (Belch & Belch, 2001:50). However, mass markets are usually too
heterogeneous to effectively target with a single marketing programme, which
means that it should be segmented into smaller sub-markets (Allen, 1999).
2.6.2 Market Segments
Since the goal is to identify groups of customers within a mass market that are
homogeneous in terms of their wants, needs and buying behaviour, the mass
market can be subdivided into a succession of segments. Each of these
segments is internally more similar in characteristics than the mass market
(Thompson, 2000).
To proceed on the definition of market segmentation - market segments consist

of individuals that have similar, but not necessarily identical, wants or needs
(Allen, 1999). Consumers within a specific market segment also hold similar
buying-related characteristics. However, even though consumers within a given
segment are similar to one another in certain areas, they will never be identical.
It might therefore be preferable to refer to customers within segments as having
'overlapping' needs, wants or purchasing behaviours (Thompson, 2000).
Market segmentation assists businesses to understand their customers' way of

thinking and behaving, thus enabling them to form a meaningful image of
consumers' needs, wants, perceptions and shopping behaviours (Belch & Belch,
2001:45). For a business to satisfy its customers' needs better than competitors,

the business needs to effectively implement market segmentation (Dunne, Lusch
& Griffth, 2002:72).
2.6.3 Niche Market
As mass markets are divided into smaller market segments, market segments in
turn can be further divided into even smaller and more defined segments. These
segments can be referred to as 'niches', since each niche includes consumers
with extensively more similar characteristics than market segments (Thompson,
2000).
According to Leduc (2000), a niche market is a group of consumers with similar

interests and needs, a strong desire for the product offering, and is easy to
reach.
Niches are extremely narrowly defined and are noticeably smaller than market
segments (Thompson, 2000). A niche should be large enough to generate the
required amount of business, but should also be small enough to minimise
competition (Leduc, 2000). By focusing on a niche, a differential advantage can
be obtained which competitors would not easily be able to match (Thompson,
2000).
2.6.4 Individual Market
Individuals form the most important part of a business and all the more in market
segmentation (Newman & Cullen, 2002:121). As mentioned, each individual in a
market can be seen as a unique market segment (McDonald, 2000:El).
Individual marketing is, however, a more general approach in the industrial

sector where individual businesses constitute market segments. Some types of
industrial products can be sold on a one-to-one basis to individual customers
that have unique production requirements that differ substantially from the

standard requirements. Consequently, a business might need to design an
entire marketing program for each industrial customer (Thompson, 2000).
2.7 BASES
FOR MARKET
SEGMENTATION
The consumer market is extremely diverse and complex, and even after the
unwanted segments are eliminated, the business will be left with a large market
to segment (Nickels & Wood, 1997:226). Table 2-1 lists the various
segmentation bases that can be used to segment the consumer market on.
In some situations, only one variable may be enough to segment a market on,
but marketers most frequently make use of a combination of approaches to
ensure accurate homogeneous segments (Belch & Belch, 2001:45). There is no
specific procedure to determine these bases, but there are certain criteria and
characteristics a business can keep in mind (Lamb eta/., 2000:226).
2.7.1 Geographic Segmentation
Geographic segmentation is one of the oldest forms of market segmentation.

During the 1970s, marketers realised that geographic location showed a
relationship with the numerous differences in consumer behaviour (Bagoui et
a/., 1998:181). People with similar social, economic and lifestyle characteristics
have a tendency to stay close to one another (Baker, 2000:272).
This form of segmentation can be seen as the division of markets into different
geographic sections. These sections may consist of nations, states or countries,
but mainly neighbourhoods (Belch & Belch, 2001:47). Other than the regions of
a world or country, markets can also be divided according to market size, market
density or climate (Lamb eta/., 2000:215).
Geographic segmentation is one of the most popular forms of segmentation, due

to its effortless nature to conduct. This form of segmentation is often also

combined with demographic segmentation to attain well-defined segments
(Bagozzi et al., 1998:181).
Table 2-1 B ----- --- ket

----- tati----
VARIABLE ExAMPLE
Geographic
Country South Africa; Zimbabwe; Kenya
Region Free State; North-West Province; Gauteng
City, metropolitanarea Potchefstroom; Cape Town; Johannesburg
Suburb Sandton;Centurion;Parow;Westville;Westering
Population size Lessthan20,000;20,000- 99,999; 100,000 - 249,999
Demographic
Age Under 6; 6-15; 16-20;21-29;30-39;40-50;50+
Gender Male; Female
Marital status Never married; Married; Separated; Divorced;
Widowed
Household size 1; 2; 3; 4; 5 or over
Less than R2,000; R2,000 - R4,999; R5,000-
Monthly income
R9,999; R10,000 - R15,999; R16,000 or more
Employment status Unemployed; Full-time job; Student; Retired
Education None;Primaryschool;Highschool;Diploma;Degree
Religion None; Catholic; Jewish; Moslem; Protestant
Race Asian; Black; White; Coloured
Nationality South African; American; Australian
Psychographic
Lifestyles Attitudes; Activities; Interests
Personality Introvert; Extrovert; Compulsive; Ambitious
Motivationor needlevel Orientation toward safety, family, self-esteem
Behavioural
User status Non-user; Ex-user; First-time user; Regular user
Buyer readinessstage Unaware; Aware; Informed; Interested; Intending
Usage rate Light; Moderate; Heavy
Loyalty status Brand loyal; Producer loyal; Store loyal
Benefit sought Economy; Performance; Service; Prestige
Source: Adapted from Belch & Belch (2001:46).

2.7.2 Demographic Segmentation
The segmentation of a market based on demographic variables such as age,

gender, income and education, is referred to as demographic segmentation
(Belch & Belch, 2001:47). Demographics are popular bases of segmentation,
since demographic information is standard, widely available and often related to
consumers' purchasing behaviour (Bagoui et a/. , 1998:180; Lamb et a/. ,
2000:215).
Some common bases of demographic segmentation include (Lamb et a/.,

2000:215; Lancaster & Reynolds, l998:3O):
age
gender (male or female)
income
ethnic background
social class
education (often related to social class, since better educated people tend
to get better jobs)
poliiical (a less obvious segmentation base)
family size (has an effect on the amount or size of purchases)
family life cycle
An intelligent marketer must be able to realise that demographics have certain

limitations (Belch & Belch, 2001:47).
2.7.3 Psychographic Segmentation
Demographic variables are useful in developing segmentation strategies, but

often they do not provide the complete picture (Lamb et a/., 2000:220).
According to Lamb et a/. (2000:220), demographics only form the framework,
while psychographics add more body.

Psychographic segmentation can be referred to as the division of a market,
based on factors such as personality, lifestyles, motives and geo-demographics
(Nickels & Wood, 1997:229; Lamb et a/., 2000:220; Belch 8 Belch, 2001:48).
According to Lancaster and Reynolds (1998:36), psychographic segmentation is
based on the assumption that consumers have specific ways of living, which are
reflected in the products or brands they buy.
Many marketers disagree with the usefulness of personality as a segmentation

basis. Lifestyle factors, on the other hand, seem to be the most popular and
effective criterion for psychographic segmentation (Belch & Belch, 2001:48;
Bagoui et a/., 1998:181).
The identification of consumers' lifestyles is generally based on an analysis of

their activities, attitudes, interests, and opinions. This lifestyle analysis approach
is generally used to compare consumers' lifestyles with their related product,
brand or media usage (Belch & Belch, 2001:48). The aforementioned procedure
assists marketers in understanding consumers' motivations or behaviours
(Bagoui et ab, 1998:181).
Psychographic segmentation provides, in general, more useful information on

consumers' thoughts and feelings, than demographic or geographic
segmentation (Bagoui et ab, 1998:181).
2.7.4 Behavioural Segmentation
According to Bagoui et a/. (1998:182), the most successful market

segmentation approach is one that is able to classify consumers according to
their product-class, brand-related preferences and behaviour patterns.
Behavioural segmentation relates to the aforementioned by segmenting a
market according to usage, loyalty, or buying responses towards a certain
product (Nickels &Wood, 1997:230; Belch & Belch, 2001:48).

Some general behavioural segmentation bases include product or brand usage,
brand loyalty, product category usage, buyer readiness stage, benefiis sought,
and level of product use (heavy versus light) (Lancaster & Reynolds, 1998:35;
Assael, 1992:561). Often one or more of the aforementioned bases are
combined with demographic or psychographic criteria in order to develop profiles
of market segments (Belch & Belch, 2001:48).
2.7.5 Other Segmentation Bases
Numerous different bases exist on which a market can be segmented. Some of

the less popular bases include:
a) Benefit segmentation
Benefk segmentation refers to the process of segmenting a market
according to the actual needs or wants of consumers, or the benefits
consumers seek from the product offerings (Lamb et a/.,2000:223).
b) Response elasticity segmentation

This method of segmentation is based on the segmenting of a market
according to consumers' se~sitivltyto the marketer's strategies. Some of
these 'marketer's strategies' might include price increases, changes in
advertising expenditures, or increased transaction effort (Assael,
1992:566).
c) Usage-rate segmentation
Usage-rate segmentation segments a market based on the amount of a
product bought or consumed. This segmentation method allows marketers
to focus marketing activities towards heavy users, or to develop several
marketing mixes aimed at different segments (Lamb et ab, 2000:224).

d) Living Standards Measure
The Living Standards Measure (LSM) is a ten-point scale (LSM-group 1 to
LSM-group 10) used to indicate the socioeconomic status of an individual
or group. LSM-group 1 refers to those of lowest status, whereas LSM-
group 10 refers to those of highest status (SAARF, 2002a:3). This
segmentation method is based on variables such as the possession of
certain household durables, water and electricity availability, the use of
various media types, and a range of demographics (SAARF, 2002a:56).
e) Media Groups Measure

The Media Groups Measure (MGM) is a market segmentation tool that
assists businesses in selecting media that are most useful in effectively
reaching their target markets. The MGM obtains customers' media
consumption information, by studying the amount and combinations of the
various media consumed (SAARF, 2002b).
As indicated in Figure 2-1, step 3 of the market segmentation process consists

of selecting segmentation descriptors. After the basis on which the market
should be segmented is selected, a business should find ways to further group
customers according to similar wants or needs (Belch & Belch, 2001:454).
A business' aim is to obtain a segment that consists of enough consumers that

are similar to each other, but that are also distinct from other segments.
Marketers therefore need to i d e n t i one or more variables on which to define
segments as accurately as possible (Nickels &Wood, 1997:225).
The business should therefore select segmentation descriptors on which to

segment the market more narrowly. Segmentation descriptors identii the
specitic variables to use, some of which might include city, age, lifestyle, or
usage rate (Lamb eta/., 2000:227).

After a business has defined its market and has selected segmentation
variables, it needs to evaluate the selected segments. Depending on the
business' market and product offering, one or more of the following criteria can
be used to evaluate the segments (Nickels &Wood, 1997:234):
Sales and protit

A segment that is more likely to purchase the business' product or sewice
offerings would be more valuable than a segment with less potential for
sales and profit.
Growth
A segment that has the potential to grow in sales and proffis would be
more valuable to a business than a segment that is shrinking.
Retention
Segments in which strong relationships can be built and customers can
be retained for a longer period are the most desirable.
Marketing costs
A business should estimate the amount it has available to spend for
marketing on each segment. Some segments may be too expensive or
even too difficult to reach effectively.
Risks
A business should consider all the external influences on a segment that
could form a risk.

Selecting the correct segments of a market is important to achieve adequate
sales volume and profitability. To be successful and useful, market
segmentation needs to comply with certain criteria, characteristics and rules
(Lamb etal., 2000:214).
2.10.1 Basic Criteria
Market segmentation cannot always be used in all situations. To be effective,

segmentation needs to meet the following basic criteria.
Substantial
The segmentation method selected, should lead to segments that are
sufficiently large enough to be economically valuable and to produce the
intended sales volume (Lamb et a/., 2000:214; Lancaster & Reynolds,
1998:29).
Measurable
The identified segment must be measurable in terms of both purchasing
power and size, such that the customers in that segment can be
communicated with by means of a unique marketing strategy (Lancaster
& Reynolds, 1998:29).
Accessible
The business should be able to reach the target market with customised
marketing mixes (Rix, 2004:96; Lamb etal., 2000:214).
Sensitive
The segmentation base should be sensitive to the planned marketing
spending (Anon, 2004b).

Some additional factors that could affect a business' success include (Anon,
2004b):
The strength of competitors.

Substitute products.
The rate of new product introductions by competitors.
Ease of entry into the market.
2.10.2 Rules to Market Segmentation
According to Business Bureau (2002), there are certain rules that a business
should follow in order to gain a viable market segment.
Be clear and specific in the identied segment.

Do not guess a segment. The selected segment must be a result of facts.
Segment only those customers relevant to the business' aspirations.
Do not 'over segment'. Aim for the largest and most relevant segment.
Be able to reach the selected segment realistically.
Ensure that a definite profit can be obtained from the selected segment.
Try and reach the largest segment first and place the others secondary.
2.10.3 Characteristics of Market Segmentation
Belch 8 Belch (2001:45) noted that using one or more of a multitude of

segmentation alternatives could segment markets. Suitable segmentation
bases, however, are those that differentiate customers according to dimensions
that are meaningful for further marketing programme development.
Segments derived from any segmentation process, will be more useful when the
following characteristics are evident (Bagoui et el., 1998:188):
Market segmentation is reliable, objective and logical.

Appropriate segmentation bases are selected carefully.
Segments are internally homogeneous and distinct from one another.
Segments are described effectively in relation to their characteristics,
sales and potential profits.
Segments are large enough to be sewed effectively.
Differences of segments are real and actionable in order to ensure that
customised marketing mixes are used.
Businesses are aware of the dynamic nature of markets.
2.10.4 Advantages of Market Segmentation
There are several reasons why a diverse market should be segmented into
smaller segments. Some of these reasons include:
Matching customer needs

Since customer needs and wants differ, it is rational to create separate
offerings for each different segment. This will therefore provide
customers with a better product offering (Brink, 1998:68).
Increased profits and growth

Customers' disposable income varies, which evidently influences their
sensitivity towards price. With market segmentation, businesses can
raise standard prices, which will consequently increase profits and lead to
increased growth (Anon, 2004a).
Retain more customers

Since customers' circumstances change over time, a business can retain
more customers by adapting marketing activities to their changing life
stages (Rix, 200494; Anon, 2004b).

Build long-term customer relationships
Enduring customer relationships can be developed if a business is willing
to devote its resources to do a better job than competitors in satisfying the
needs of customers (Nickels &Wood, 1997:222).
Efficient and easier marketing communications

A business can ensure that the unprofitable or inappropriate segments
are eliminated, to ensure that no time or money is wasted. With a market
segmentation strategy, a business can better allocate its resources to
satisfy its customers. This strategy also simplifies the marketing process,
since segments have many homogeneous characteristics (Brink, 1998:68;
Lancaster & Reynolds, 1998:39).
Increased response to changing consumer market

Consumers are dynamic in nature; their needs, wants and behaviours can
change from day to day. Market segmentation narrows down the mass
market, which makes it easier to attend to the changes that might occur in
consumers' needs, wants or behaviours (Nickels &Wood, 1997:222).
Designing marketing mixes

Segmentation provides marketers with information that can assist them in
designing specific marketing mixes that match the characteristics of one
or more of the selected segments (Lamb eta/., 2000:214).
2.10.5 Disadvantages of Market Segmentation
Market segmentation may have many benefts to a business, but this strategy
also has some drawbacks. Brink (1998:68) identified the following to be some of
the major disadvantages of a market segmentation approach:
Segmentation is more expensive, since the business needs to develop

different product offerings and marketing mixes for each segment.

Since only some of the segments are selected, market coverage might be
limited.
One business can take away the market share from another business of
the same enterprise if too high a degree of differentiation of the basic
product occurs. This is usually referred to as 'cannibalism'.
Lamb et a/. (2000:227) identifies a target market as a group of consumers or

businesses with similar characteristics, for which a marketing mix is designed,
implemented, and maintained, in order to meet the needs of that group.
The market segmentation process is complete once the business has identified
the segments it wants to address. The marketing process continues, since the
business now needs to make decisions regarding target marketing and
positioning (Nickels &Wood, 1997:236).
Thus, the business needs to evaluate the opportunities presented by the various
segments and then select one or more of these segments as a target market
(Belch & Belch, 2001:42).
2.11.1 Evaluating Market Segments
To effectively evaluate the various market segments, the business should

consider the following factors:
Segment size and growth

The business should collect and analyse data on the different segments'
sales, growth rates, and expected profitability. The segments will be
attractive to the business if they are potentially profitable, of considerable
size (depending on the size of the business) and have growth
characteristics (Kotler & Armstrong, 2004251).

Segment structural attractiveness
The business furthermore needs to examine major structural factors that
might affect long-run segment attractiveness. A segment that contains
strong and aggressive competitors (UTD, 2004), many substitute
products, or has strong bargaining power will be less attractive to the
business. Powerful suppliers, who can control prices or reduce the
quality or quantity of ordered products or services, will also reduce the
segment's attractiveness (Kotler & Armstrong, 2004:251).
Company objectives and resources

The selected segment will only be attractive if it improves the business'
long-run objectives, and complies with the available resources (Brink,
1998:77). The business should consider whether it possesses the skills
and resources to compete successfully in the selected segment, and
should only enter segments in which it can offer superior value and gain a
competitive advantage (Kotler & Armstrong, 2004:251).
2.11.2 Target Market Strategies
After evaluating the different segments, the business must decide which and
how many segments it will target. Decisions regarding target marketing will
determine the amount of segments to target, as well as the plans for the
marketing mix to effectively reach these segments (Nickels & Wood, 1997:236).
A business can target its selected market segments according to three main
market coverage strategies.
a) Undifferentiatedtargeting
With an undifferentiated targeting strategy, a business ignores the
differences between market segments, and rather approaches the market
as a whole with only one marketing mix (Belch & Belch, 2001:50). A
business therefore assumes that individual consumers have identical needs

and wants that can be met with one general marketing mix (Lamb et al.,
2000:227).
This strategy is designed to appeal to the largest amount of consumers

(mass market), since it focuses on similar consumer needs rather than on
those needs that are different (Monash University, 2004). Figure 2-3
illustrates an undifferentiated targeting approach.
Fiaure 2-3 Undifferentiated taraetina
Source: Gilligan & Wilson (2003:400).
b) Differentiated targeting
As Figure 2-4 indicates, a differentiated targeting strategy involves the
formulation of separate and distinct marketing mixes for each of the
selected segments (Belch & Belch, 2001:51). The product, price,
distribution and promotion are therefore altered as required for each distinct
segment (Nickels & Wood, 1997:237).
Figure 2-4 Differentiated targeting
[ Business

This targeting strategy has become increasingly popular, since it often
creates higher sales levels than undifferentiated targeting. However, as the
marketing mix needs to be customised along with each segment's needs
and wants, the overall costs will be higher (Monash University, 2004).
c) Concentrated targeting
A business using a concentrated targeting strategy, concentrates on a large
share of one or more segments (Belch & Belch, 2001:51). This strategy
can be useful to a business with limited resources (Nickels & Wood,
1997:237).
Because the business only concentrates on one segment, it can focus on

understanding the segment's needs, wants and behaviours to be able to
develop and maintain a specialised marketing mix (Lamb et al., 2000:229).
Figure 2-5 illustrates a concentrated targeting strategy approach.
Figure 2-5 Concentrated targeting
""'"
/'
Segment 1
r Business , mum M;k;.ji-Mm uuu u'

Segment 2
'"- Segment 3 J
2.12 POSITIONING
After the target market strategy is selected, a business needs to decide how to
position its product or service. This will be important, since the marketing
decisions made will affect the image of the business' product offering (Fill,
1999:212).

According to Brink (1998:78), positioning provides a focal point in the
development of a marketing strategy. However, positioning is not about the
product, but rather about what the consumer thinks about it. Furthermore it is
not the physical aspect of the product that is important for positioning, but the
way the product is perceived that is important (Fill, 1999:511).
Market positioning can be defined as the observed fitting of the product or

service offered to one or more of the segments of the broad market, in such a
way as to significantly distinguish it from competition (Belch & Belch,
2001:52).
Simply, positioning is the way the target market perceives the business
compared with relevant competitors. It is therefore essential for the business to
understand its product offering from the consumers' point of view (Determan,
1999).
Assael (1992:568) emphasises that market segmentation and product-

positioning strategies should be developed simultaneously. Market segments
should be branded by their related needs, and products should be positioned to
accordingly meet these needs. A product positioning strategy should also be
developed for each selected segment. This is required to ensure that a
business' products or services obtain a pertinent place in the chosen target
markets, so as to oppose competition in the marketplace (Lancaster & Reynolds,
1998:40).
2.12.1 Positioning Strategies
A business can distinguish its products or services from those of competitors

based on a set of meaningful differences. The aforementioned process is more
commonly known as differentiation (Cosenza, 2004).

Businesses can make use of one or more of the following types of differentiation
or positioning strategies.
Product differentiation
This strategy entails the positioning of the product according to its
characteristics.' The business can set its product apart from competitors
based on attributes, features or beneffis that the product has compared to
competition (Fill, 1999:515).
Pricequality differentiation
A business can obtain a measure of value-for-money by weighing quality
up against cost. Usually a high price indicates high quality, just as a low
price often mislead consumers into thinking a product to be of low quallty.
This strategy is therefore a popular product evaluation criterion, mainly
because price itself can be a powerful communicator of quality (Fill,
1999:515; Brink, 1998:79).
Use or application differentiation

Another way to position is by informing consumers when or how a product
can be used or applied. By means of this strategy, a position can be
created within the minds of the consumers (Fill, 1999:517; Determan,
1999).
Product-class differentiation
This strategy entails the positioning of products according to various
product classes. Productclass positioning is popular in situations where
the segments are essentially uninteresting, or when competitors occupy
other positions (Fill, 1999: 517; Brink, 1998:80).
Product-user differentiation
Another positioning approach is to associate the product with a certain
class or type of user. Thus, positioning the product in such a way, that

the targeted user can be clearly identified (Brink, 1998:80; Determan,
1999).
Cultural symbols differentiation

A business can make use of deeply entrenched cultural symbols to
differentiate its product from that of competitors. Symbols including
cultural heritage and tradition, heraldic devices or visual cues can be
implemented to communicate quality, experience and knowledge (Fill,
1999:518).
2.1 2.2 Characteristics of Positioning Differences
The differences on which a business positions its products or services must

have the following characteristics (Determan, 1999; Cosenza, 2004).
Important Seeing that the difference will deliver a great beneffi to

customers.
Distinctive. Competitors should not offer the difference, or the business

should be able to offer the difference in a more unique way.
Superior. The difference offered should be superior in its class. The

customer should not be able to obtain the same benefit elsewhere.
Communicable. The difference must be easy to communicate or be

explained to potential customers.
Preemptive. Competitors should not be able to simply copy the

business' difference.
Affordable. Customers should be able to afford to pay the difference

being offered.

Profitable. The business should be able to introduce the difference to be
offered proffiably.
After selecting the segment(s) to target and determining the positioning

strategies to be used, the business can proceed to the final step in the market
segmentation process (refer to Figure 2-1): the designing, implementing, and
maintaining of a marketing mix (Lamb eta/., 2000:227).
To be able to develop an effective marketing strategy, an effective marketing mix

should be narrowed down (Perreault & McCarthy, 1999596). The various
elements of the marketing mix are needed for a business to achieve its
marketing objectives and maintain a differential advantage in the market
(Webber, 2001).
2.13.1 Marketing Mix Clarification
The marketing mix includes the combination of the key demand-influencing

variables that are available to the business. The customer responds not only to
the product when making a purchase, but also to the series of variables that
composes the product offering (McCartan-Quinn, Durkin & Stewart, 2004).
The business needs to determine the needs and wants of consumers, develop a
product or service that satisfy these needs, offer the product at a certain price,
make the product available at a specific place, and develop an advertising or
promotion programme to create awareness and interest in the product offering.
The classic description of the aforementioned process is the 'four Ps of
marketing' - product, price, place and promotion -which comprise the marketing
mix (Belch & Belch, 2001:lo).

However, this simple formula of the marketing mix was designed when
businesses sold standardised products, and changes in the market were
relatively slow. These days, however, are characterised by change (Gilligan &
Wilson, 2003:7). The marketplace is changing continuously by means of
improved technology, changing consumer needs, new markets, new competitors
and new types of media (Nickels & Wood, 1997:20).
These changes entail that a more comprehensive method of marketing

management needs to be developed, that builds on the 'four PSI-model(Nickels
& Wood, 1997:20). Gilligan & Wilson (2003:7) therefore emphasis4 an
expanded marketing mix, with three additional elements, namely people,
physical evidence and process management.
2.13.2 Marketing Mix Elements
Although Gilligan & Wilson (2003:7) added three additional elements to the
traditional marketing mix; it is important to understand that there are many other
elements that can be added. The marketing mix can consist of numerous other
elements, due to continuous changes in the marketplace (Webber, 2001).
However, for the purpose of this study, only the four well-known elements - 4 Ps
- will be discussed. Figure 2-6 provides an illustration of the marketing mix
elements.
a) Product
The product forms the heart of the marketing mix. The business is unable
to design a distribution strategy, decide on a promotion campaign, or set a
price without a product. Thus, meaning that without a product, a marketing
mix will not exist (Lamb eta/.,2000:44).
The product offering not only includes the physical component, but also the
package, warranty, after-sale service, brand name and value of the product

(Rix, 2004:26). Products can furthermore be physical goods, services or
ideas, and should offer value to consumers (Lamb et al., 2000:44).
Figure 2-6 The marketing mix
MARKETING MIX I
[
I
I I I I
" , ""I
/' ...., ".
PRODUCT PLACE PROMOTION PRICE
Service Supplies Advertising Costs

Quality Inventory Direct mail Payment
Range Storage Sales catalog Structure
Brand Retailers PR
Warranties Export Publicity
Style -"
Import .J '- Research ,) -'
Source: Adapted from Perreault & McCarthy (1999:600).
b) Price
Price is what a customer must give up in exchange for the product or
service being purchased, which is usually determined according to
monetary value. A business should also determine the total cost of the
product from the customer's point of view (Webber, 2001).
The business can raise or lower prices freely, and therefore price is often
seen as the most flexible marketing-mix element (Lamb et al., 2000:44).
Price also includes the selection of the right base-price level, the setting of
special prices, and decisions regarding discounts, allowances, and whether
and how to charge delivery costs (Rix, 2004:26).
c) Distribution (Place)
Distributionis concernedwith providingproductsto consumerswhen and
where they want them. Distributionalso includes all business activities
Chapter 2: MarketSegmentation 46
concerned with the selection of intermediaries, physical handling, storing
and transporting of raw materials or finished products (Rix, 2004:26).
The primary goal of distribution is therefore to ensure that products arrive -

at the right time - at their destination in a usable condition (Lamb et a/.,
2000:44).
d) Promotion
Promotion includes personal selling, advertising, sales promotion, and
public relations (Webber, 2001). The role of promotion in the marketing mix
is to create relations with target markets by informing, educating,
persuading, and reminding them of the benefits of the business or a product
(Rix, 2004:26).
A good promotion strategy can increase sales considerably, but it does not
necessarily guarantee success (Lamb et ab, 2000:44). A business can
spend large amounts on promotion, this will, however, mean nothing if the
business' product is of poor quality, priced improperly, or lacks effective
distribution channels. It is thus clear that the marketing mix elements are
significantly interdependent and need to be combined to create a
successful marketingstrategy (Belch & Belch. 2001:lO).
2.13.3 Marketing Mix Characteristics
McCartan-Quinn et a/. (2004) identified a number of characteristics that an

effective marketing mix can contain.
A marketing mix increases the business' understanding of its customers,

which can improve marketing management in general.
= A competitive advantage can be obtained from decisions regarding the
four elements.

A successful combination of the four elements will create a reliable
subject.
The marketing mix simplifies the reality of marketing management.
It provides a realistic framework for making decisions regarding
marketing.
2.13.4 Marketing Mix Gaps
It would seem that if a business completely understands the needs, wants and
behaviours of its target market, the designing of a marketing mix should be easy.
However, there are three gaps in conflict with the aforementioned argument
(Perreault 8 McCarthy, 1999:597):
The business often has limited information on the needs, wants and
behaviours of its target market. It is sometimes also complicated to
obtain this information.
Competitors in the market are also trying to satisfy these or similar needs
of consumers, which may force a business to change its marketing mix.
Other dimensions in the marketing environment may change radically,
which might lead to even more changes in the marketing mix.
From this chapter, it is apparent that market segmentation is an important

requirement for successful marketing. It can be used to improve the business'
marketing performance by accurately defining the business' marketing objectives
and by identifying the wants, needs and behaviours of customers.
The entire chapter summarised the market segmentation process, which

consists of six distinctive steps that were discussed in detail. These steps
include the selection of the market to be segmented; selecting the basis on
which to segment the market; selecting the segmentation descriptors; profiling

and evaluating the segments; selecting the market segment(s) to be targeted
and finally the designing, implementing and maintaining of a marketing mix for
the selected target markets.
To be able to select a market, the types of markets, as well as the basic

characteristics that a market has to comply with must be taken into
consideration. The different market segmentation approaches that can be
followed were discussed, and some of the most renowned bases on which
consumer markets can be segmented, were examined.
To segment the market more narrowly, the various segmentation descriptors,

and some of the criteria that can be used to evaluate segments were discussed.
To entirely understand the use and successful application of market
segmentation, the overall criteria, rules, characteristics, advantages, and
disadvantages of segmentation were studied. The three strategies on which
selected segments can be targeted, and the various strategies a business can
use to position itself and their required characteristics were analysed. Finally,
the basic elements of the market mix - the 'four P's' - and some of the
characteristics of an effective marketing mix were identified. The gaps that
influence the effortless design of a marketing mix were recognised.
In this chapter the basis of market segmentation was discussed. In the next
chapter, two different types of market segmentation tools will be discussed,
namely the Living Standards Measure (LSM) tool and the Media Groups
Measure Tool (MGM). These two tools aid in the market segmentation process.

CHAPTER 3
LIVING STANDARDS MEASURE AND
MEDIA GROUPS MEASURE
When observing a market, it is important to understand that although people are

diverse in nature, they do have certain commonalities (Rix, 200493). A market
segmentation tool should be able to develop an index that will differentiate
between people with different behaviour patterns and group together those
people with similar behaviour (Haupt, Smit & Ncube, 2002).
South African businesses are increasingly making use of the LSM as a market
segmentation tool (SAARF, 2001a). Its popularity is increasing rapidly, since the
LSM is globally being recognised as one of the most sophisticated segmentation
models (Wortley & Maserame, 2002). Its main characteristic is that neither race
- which is a relatively sensitive point for the South African community - nor
income is used as segmentation variables (Haupt, 2001a).
Wth the LSM, the living standards of people are used as segmentation
descriptors. This seems to be understandable, since if a person is used to high
living standards, it is not likely that he Ishe will purchase cheap, lowquality
products. The same principle also applies to people with low living standards.
They will rather purchase inexpensive products, which are usually of lower
quality than expensive ones (SAARF, 2001b).
An explanation of the SAARF LSM as a market segmentation tool, how it was

developed and how it works, will be provided. The purpose of the Media Groups
Measure (MGM) tool in correlation with the LSM will also be analysed,
specifically from retailers' point of view.
Chapter 3: Living Standards Measure and Media Groups Measure 50

3.2 SOUTH AFRICAN ADVERTISINGRESEARCHFOUNDATION(SAARF)
The South African Advertising Research Foundation (SAARF) is an

independent, non-profit organisation, and the developer of the Living Standards
Measure (LSM) market-segmentation tool (SAARF, 2004).
This organisation was formed to provide detailed information about the

population's media and product consumption patterns. The primary objective of
this research was to effectively assist advertisers and advertising agencies in
making decisions regarding media selection for their advertising campaigns
(SAARF,2001b).
It was realised that, in order to provide unbiased research data that would be
accepted entirely, SAARF needs to be managed by a joint industry body. This
body would then be able to supervise fieldwork and report production to ensure
neutral research results (SAARF, 2001b). The aforementioned body is compiled
by the advertising, marketing and media industries in South Africa (SAARF,
2003). Figure 3-1 illustrates this tripartite nature of SAARF.
Figure 3-1 The nature of SAARF
MEDIAOWNERS
· Cinema
· Internet
· Outdoor
· Print
· Radio
· Television
ADVERTISING AGENCIES
Source: Haupt et al. (2002).
Chapter 3: LivingStandards Measure and MediaGroups Measure 51

The joint body - advertisers, advertising agencies and media owners - finance
and support SAARF in conducting research. SAARF, on the other hand, was
established to meet a single, clearly defined goal - to provide a common
research currency to its three stakeholder industries, which can aid them in
buying and selling advertising time and space (Haupt et a/.,2002).
SAARF furthermore developed a number of research products to aid in the

development of a 'common currency' for the joint body (Haupt, 2002).
1) All Media and Products Survey (SAARF AMPS)

2) Radio Audience Measurement Survey (SAARF RAMS)
3) Television Audience Measurement Survey (SAARF TAMS)
4) SAARF Universal Living Standards Measure (SU-LSM)
5) SAARF Branded Data
6) Media Groups Measure (SAARF MGM)
7) SAARF Development Index
These research products assist in the provision of a thorough analysis of the

demographic profiles of South African consumers (aged 16 years or older),
information about their media habits, and information on their consumption
patterns (Haupt, 2002).
3.3 THENATURE
OF THE LSM
Traditional studies conducted on South Africans, mainly used segmentation

descriptors such as geographic location, population group, education or income.
The problem, however, with single variable segmentation methods such as
these are that they generally do not distinguish the diverse market's living
standards sufficiently (SAARF, 2004). This lead to the emergence of South
Africa's first multivariate market segmentation index - the LSM (Haupt, 2001a).

3.3.1 LSM Defined
The LSM-index is based on a group of variables that segment people according

to their living standards, combined with demographic characteristics, into
relatively homogeneous groups. Criteria such as degree of urbanisation,
possession of major assets and access to services are some of the variables
used in the LSM (Haupt, 2001a).
Essentially, LSM can be defined as a scale that indicates the socio-economic

status o f an individualor a group (SAARF, 2004).
3.3.2 History and Development of the LSM
In designing the first LSM, SAARF aimed to combine a number of variables that
would be stronger discriminators than any single variable. SAARF aimed to
create an index that could be used to segment South Africa's diverse population
(SAARF, 2001a).
The process of designing the LSM index started out by assigning a score to
each respondent based on a wide range of variables. These variables, which
had been selected on judgement, included the ownership of a range of durable
household articles, access to water and electricity supplies, and a number of
demographics such as education and population group. Each respondent's
score was calculated according to an index, which in turn was divided into a
number of groups based on the distribution of the total population (SAARF,
2004).
As a result, the South African population could be divided into eight distinct
groups, from LSM-group one (lowest living standards) to LSM-group eight
(highest living standards) (HSRC, 2002:165). Figure 3-2 provides a concise
illustration of how the LSM-index developed over the years.

Figure 3-2 LSMtimeline
1989/1990
71 variables analysed & 13 variables construct LSM.
1991/1992
Variablesandweightsupdatedannually.
1993/1994
Variables and weights kept constant.
1995
39 variablesadded. 20 variablesconstruct1995LSM.
Variables& weightswere keptconstantuntil2000.
1997
LSM segments 7 and 8 divided.
2000
Variables and weights changed.
2001
SU-LSM- 29 variables. Scaleextendedto 10groups.
Source: Adapted from SMRF (2004).
· 1989/1990
From a list of 71 variables, thirteen were selected for their ability to
discriminateconsistently between respondents - see Table 3-1. In an
attempt to improve this measure, income and education were forced into
the index. It was, however, found that neither of these two variables
contributed sufficiently to be permanently included in the index (HSRC,
2002:165).

Table 3-1 Original LSM variables
1. Polisher IVacuum cleaner 8. Sewing machine

2. Fridge IFreezer 9. Non-supermarket shopper
3. Television set 10. No tumble dryer
4. Water IElectricrty 11. No domestic worker
5. Washing machine b2. No video cassette recorder
6. No car in household 13. Rural dweller (outside PWV 8
7. Hi-fi IMusic centre Western Cape)
7
Source: SAARF (2001a).
Since each one of these variables has a different degree of impact on the
classification of the respondent, certain 'weights' were allocated to the
variables. To determine the respondent's LSMclassification,the weights of
the durables that he Ishe owns are added together (SAARF, 2004).
During 1991 and 1992, SAARF increased the number and range of
variables to improve the LSM-index (SAARF, 2004). The weights were also
adjusted accordingly, since it seemed that some variables - which are
significant reflectors of a person's living standards - received too little
attention in relation to other less significant variables (SAARF, 2001a).
Table 3-2 indicates the variables used in the 1993 LSM.
Table 3-2 1993 SAARF LSMs
1. Fridge IFreezer 8. Hi-fi IMusic centre

2. Water IElectricrty 9. No domestic worker
3. Polisher I Vacuum cleaner 10. Washing machine
4. Non-supermarket shopper 11. Sewing machine
5. No car in household 12. Metropolitan dweller*
6. Television set 13. Rural dweller (outside PWV &
. 7. Microwave oven* Western Cape)
New. Video cassette recorder and tumble dver no longer used
Source: SAARF (2001a).

1995
In the mid-nineties, another 39 variables were added to the original list of
71 characteristics. After thorough analysis, this list of variables contributed
to the invention of a new list consisting of 20 indicator variables (SAARF,
2004). The list of 20 variables was used for the first time in the 1995
survey. From 1995 to 1999, the list was verified annually to ensure its
continuing validity. Table 3-3 shows the twenty variables of the 1995
survey, which were used until 1999 (HSRC, 2002:167).
Table 3 3 1995 SAARF LSMs
1. Flush toilet* 11. Dishwashing liquid*

2. Polisher IVacuum cleaner 12. Hwsehdd supermarket shopper
3. Neither water nor electricity 13. Hot running water
4. Fridge IFreezer 14. No credit facility*
5. No car in household 15. Television set
6. No financial services used* 16. Microwave oven
7. Non-supermarket shopper 17. Rural dweller (outside Gauteng
(personal) &Westem Cape)
8. No insurance policy* 18. Washing machine
9. Hi-fi Imusic centre 19. Hut dweller
10. Teleohone in home* 20. No domestic worker
* New. Sewing machine and metropolitandweller no longer used.
Source: HSRC (2002:167).
= 1997
In 1997, SAARF refined the LSMclassification even more, by splitting up
LSM-segments 7 and 8. These two segments were each divided into an
upper and lower stratum - 7 low, 7 high. 8 low and 8 high. This division
was required to extract more detail at the 'top end' of the market (SAARF,
2004).

The variables used for previous LSM-surveys were analysed once again to
ensure validity for the 2000 LSM-survey. A number of new variables were
included to aid with topend differentiation, which resulted in the
replacement of only four variables, with 16 remaining the same (HSRC,
2002:168). Table 3 4 lists the 2000 SAARF LSMs.
Table 3 4 2000 SAARF LSMs
1. Built-in kitchen sinkR 11. Stove IHotplate (electric)*

2. No car in household 12. Polisher IVacuum cleaner
3. Flush toilet 13. No insurance policy
4. Supermarket shopper (personal) 14. Hi-fi IMusic centre
5. Microwave oven 15. Video cassette recordeP
6. Credit facility including retail card 16. No domestic worker
7. Fridge IFreezer 17. Television set
18. Car ISedan IBeach buggy I
8. Washing machine
Hatchback I2-Seater coupe*
9. No financial services used -
1 19. Hot runnina water
1110. Hut 120. Home tele~hone 1
* New. Dishwashing liquid, water or electricity, rural dweller and
household supermarket shopper no longer used.
Source: HSRC (2002:168).
During 2001 SAARF developed an improved LSM, known as the SAARF

Universal LSM (SU-LSM). The SU-LSM was introduced to the SAARF
panel at the end of August 2001, and its development is based on the
previous LSMs' biased nature (HSRC, 2002:169). This new classification
instrument consists of ten distinct segments instead of only eight, and is
based exclusively on household variables (Haupt, 2001a).
Since the SU-LSM market-segmentation tool forms the focus of this study, it is
necessary to clearly understand what it is and how it works.

3.4 SU-LSM CLARIFIED
According to Haupt et a/. (2002), the SU-LSM is a target-marketing tool, which

can also be used in the selection of proper media. Since the SU-LSM does not
consist of any personal attributes (Haupt et a/., 2002), it can be used constantly
on various types of markets, without having to go through the annual
modifications (Wortley & Maserame, 2002).
The SU-LSM obtained its name from the fact that it is based on universally
applicable variables only. This means that all respondents (16 years and older)
can answer all the LSMquestions, and be indexed without biases. The SU-LSM
is furthermore designed to be universally applicable to all of SAARF's products
and to all future surveys (Haupt eta/., 2002).
3.4.1 SUlSM Compilation
The new SU-LSM is similar to the older versions, but starts out with ten
segments - from 1 (one) with the lowest living standards, to 10 (ten) with the
highest living standards - instead of only eight. What were previously referred to
as LSM 7 low, 7 high, 8 low and 8 high are now referred to as LSMgroups 7, 8,
9 and 10 respectively (Haupt, 2001a).
The SU-LSM can furthermore be adapted to remain up to date with societal

changes. As the South African society develops, the SU-LSM has the ability to
expand beyond group 10, into 11, 12 or more groups (SAARF, 2004). It is
sensitive to changes like these, and will therefore reflect the South African
market relatively accurate (SAARF, 2001a).
To ensure that all the members of a household belong to the same SU-LSM-
group, the SU-LSM is based entirely on household variables instead of personal
variables (HSRC, 2002: 169).

In order to refine the scale even more, the LSM-list of variables was expanded
from 20 to 29 household variables, of which 15 have been carried through from
the previous (2000) LSM-index (Wortley & Maserame, 2002).
Table 3-5 relates the latest SU-LSM-index that has been developed and that will
be used for the purpose of this study.
Table 3-5 SAARF Universal LSM
Source: SAARF (2004).
The abovementioned variables form the basis of the questions asked in the
SAARFquestionnaire. According to these variables, people can be segmented
into relatively homogeneous groups, which makes targeting and marketing to a
selected segment less complicated (Haupt, 2001b). These questions were also
asked in the questionnaire that has been designed for the purpose of this study
(see Appendix B).

3.4.2 Calculating the SU-LSM
As mentioned, each one of the SU-LSM-variables is connected to a certain

weight as determined by SAARF. The weight of a variable depends on the
importance of the variable in describing the respondent's living standards
(SAARF, 2001a). Table 3-6 lists the 29 SU-LSM-variables and their related
weights.
Some of the variables in Table 3-6 have negative weights. The following need to
be kept in mind with these 'negative' variables (SAARF, 2004):
1) 'No domestic in household: A score is only assigned if the respondent

does not have a domestic worker.
2) '0 or 1radio set in household: A score is only added if there is no or only
one radio set in the respondent's household.
3) 'No cellphone in household: A score is only given if no one in the
respondent's household has a cellphone.
4) 'Traditional hut'. The respondent only gets a score if he Ishe lives in a
traditional hut.
5) 'Rural outside Gauteng and the Western Cape: A score is only allocated
if the respondent does not live in Gauteng or the Western Cape, or in a
metropolitan area, city or town in any of the other provinces.
To allocate a respondent to one of the SU-LSM-groups, the following procedure

needs to be followed (SAARF, 2004):
Step 1: ldentii the variables in Table 3-6that apply to the respondent.

Step 2: Apply the weights of each variable that comply with the respondent.
Step 3: Add up the weights of the variables the respondent identified.
Step 4: Add the constant (1,340410) to the total score to obtain the final SU-
LSM-score.

Step 5: Apply the information in Table 3-7 to allocate the respondent to one
of the ten SU-LSM-groups.
Table 3-6 SUISM-variables and weights
VARIABLE WEIGHT
1. Hot runnina water 0.158200
2. Washing machine 0,138930
3. Fridge IFreezer (combination) 0,152515
4. No domestic in household -0,222364
5. 0 or 1 radio set in household -0,158252
6. Flush toilet in- Ioutside house 0,142228
7. Video cassette recorder 0,134488
8. Motor vehicle(s) in household 0,155217
9. Personal computer 0,132148
10. No cellphone in household -0,175184
11. Electric stove 0,163219
12. Tumble dryer 0,117338
13. Home telephone 0,097140
14. Traditional hut -0,201085
15. Vacuum cleaner IFloor polisher 0,135318
16. Television set(s) 0,133830
17. Microwave oven 0,126829
18. M-Net IDStv subscription 0,126068
19. Rural rest (excludingWestem Cape 8 Gauteng rural) -0,093228
20. Hi-fi IMusic centre 0,105378
21. Built-in kitchen sink 0,165505
22. Home secunty service 0,097632
23. Deep freezer 0.093849
24. Water in home or on stand 0,127671
25. Sewing machine 0,090320
26. Dishwasher 0,119925
27. Electricity 0, 128613
28. Live in Gauteng 0,056788
29. Live in Western Cape 0,079999

Table 3-7 Range for each SU-LSMgroup
SU-LSM SU-LSM
TOTALWEIGHT TOTALWEIGHT
GROUP GROUP
1 0,00000 to 0,72100 6 2,12701 to 2,68500
2 0,72101 to 1,05300 7 2,68501 to 3,01000
3 1,05301 to 1,35600 8 3,01001 to 3,32400
4 1,35601 to 1,72600 9 3,32401 to 3,65000
5 1,72601 to 2,12700 10 More than 3,65000
Some general-, demographic- and media usage information on the various SU-
LSM-groups are provided in Appendix C.
3.4.3 SU-LSMScale and Society
The SU-LSM is a socioeconomic descriptor, and since not all of a person's

characteristics are driven by socioeconomic factors, the SU-LSM differentiates
better between some aspects than others (SAARF, 2004).
Gender is relatively independent of the SU-LSM-scale. Each SU-LSM--

group consists of approximately the same amount of male and female
respondents (SAARF, 2004).
Household income, on the other hand, is highly related to socio-economic

standing, with SU-LSM-group 1 having the lowest household income, and
SU-LSM-group 10 having the highest (SAARF, 2004). Table 3-8
demonstrates the average household income of each of the SU-LSM-
groups, for 2001B (January - December 2001), 2002A (July 2001 - June
2002) and 20028 (January - December 2002). There is an increase in
average household income for every SU-LSMgroup (SAARF, 2003).

Table 3-8 Average household income by SU-LSM
20018 2002A 20028
SU-LSM Group 1 R 777 R 804 R 832
SU-LSM Group 2 R 885 R 962 R 1 075
SU-LSM Group 3 R 1 107 R 1 188 R 1 318
SU-LSMGroup 4 R 1 523 R 1 570 R 1 724
SU-LSMGroup 5 R 2 205 R 2 230 R 2 421
SU-LSMGroup 6 R 3 557 R 3619 R 3 897
SU-LSMGroup 7 R 5 509 R 5 675 R 5 859
SU-LSMGroup 8 R 7 428 R 7 587 R 8 053
SU-LSM Group 9 R 9 861 R10 245 R10732
SU-LSM Group 10 R13 788 R15 076 R15 931
· Education is also related to socio-economic standing, with the lower SU-

LSM-groups representing the less educated. Education has improved
significantly over the last ten years - as seen in Figure 3-3 - but the job
market is not growing fast enough to accommodate these highly skilled
people. Table 3-9 indicates the employment status as for 2002 for the
SouthAfricanpopulation(SAARFI 2003).
Figure 3-3 Education - 10 year trends

100
fs!. 80
g 60
i
'S 40
oa.
... 20
o
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Years
.No school a Primary completed .Matric +

Table 3-9 Employment status (January -- December 2002)
% '000
Work full-time 24,1 7124
Work part-time 10,1 2982
Not working 65,8 19 477
Classify self as "unemployed" 33,8 9991
Self-employed 8,6 2553
Source: Adapted from SAARF (2003).
· Population distribution is also relatively related to socio-economic

standing. SU-LSM-groups 2 to 6 account for 67% of the adult population
(16 years and older). Figure 3-4 compares the percentage of the South
African population for each SU-LSM-group over three years - 2001A, 20018
and 2002A (Wortley & Maserame, 2002).
Figure 3-4 SU-LSM population distribution

16
14
~ 12
= 10
i 8
1o
6
... 4
2
o
1 2 3 4 5 6 7 8 9 10
SU-LSMGroups
I .2001A C]20018 .2002A I
Source: Wortley & Maserame (2002).
The population percentage of the North-West Province for 2002A (July

2001 - June 2002) and 20028 (January - December 2002) is analysed in
Figure 3-5.

Figure 3-5 North-West Province population percentage
20
--
'i!
0
I: 15
-I
Do
10
0
IL
5
0
1 2 3 4 5 6 7 8 9 10
SU-LSMGroups
. 2002A 1:1 20028
3.4.4 SU-LSM and Race
According to Haupt (2001a), many people believe that LSMs are only a
courteous way of referring to race. However, race has never been used as a
determinant or as a variable in an LSM-index (Anon, 2002).
The LSMs do correlate closely with race, because it only reflects reality - for
which it was created. Although race is not an LSM-variable, LSM indicates the
reality of the South African society (Anon, 2002).
As can be seen in Figure 3-6, black adults dominate SU-LSM-groups 1 to 6, with

82% falling into SU-LSM-groups 1 to 5. In contrast, white adults dominate SU-
LSM-groups 8 to 10, with 98% falling into SU-LSM-groups 6 to 10 (Wortley &
Maserame, 2002). The higher LSM-groups are generally more multiracial than
the lower ones (Anon, 2002).
The SU-LSM can therefore indicate a society's race profile, but race is not a
living standard descriptor, only a demographic descriptor. The impact of race as
a differentiating variable will decrease when the balance in the South African

society has been restored. The relationship between race and the SU-LSM will
also decrease as societal standards increase (Haupt, 2001b).
Figure 3-6 SU-LSM race profile for 2002A

100
"- 80
I:
0 60
11
OS 40
a.
0
.. 20
0
1 2 3 4 5 6 7 8 9 10
SU-LSM Groups
I_slack \:IWhite.Coloured olndian I
Source: Adapted from Wortley & Maserame (2002).
3.5 SAARF MEDIAGROUPSMEASURE(SAARF MGM)
The SAARF Media Groups Measure (MGM), which was launched towards the
end of 2002, is another type of segmentation tool. The MGM segments in a
different way than the SU-LSM. The main purpose of this tool is to optimise the
selection of media in reaching specific target markets (Haupt, 2002).
In this study, the MGM-tool will be used to recommend the most appropriate
types of media the participating supermarkets should use to reach their
customers.
3.5.1 MGMDefined
The MGM assists businesses in selecting media that are most effective in
reaching their target markets (SAARF, 2002b). Aspects such as geographic
location (that restricts availability), poor education standards (that impact on

literacy) and the financial ability to afford certain media, are all limiting the reach
of media (Corder, 2001). It is mostly less well-off people that are affected by
these limitations, and therefore only reached by a few media (SAARF, 2004).
Wealthier, urban dwellers on the contrary, tend to have more disposable income
available, higher education levels and enhanced access to media (Corder,
2001).
The MGM is designed to obtain media consumption information, by examining

the amount and combinations of the different media consumed (SAARF, 2002b).
Readers, listeners, viewers, cinemagoers, Internet users and those exposed to
outdoor media, can be identified and profiled by means of the MGM. It therefore
makes it possible for businesses and their advertising agencies to select media
that would be best to reach the business' target market (SAARF, 2003).
3.5.2 Analysis of MGM
The MGM-scale consists of eight groups, which are related to the level of media
exposure that consumers get (SAARF, 2002b). Lower MGM-groups have less
duplicated media exposure than the higher MGM-groups. The duplication of
media types increase as the MGM-scale progresses to higher levels (SAARF,
2004). A summary of the characteristics of each of the MGM-groups follows.
· SAARF MGM-group 1
Except for extensive radio exposure, this group has relatively limited
exposure to other media types (SAARF, 2002b). Figure 3-7 illustrates the
coverage of the various media for MGM-group 1.
Radio exposure to this group is still high, but to a lesser extent than in
MGM-group 1, due to the presence of other media types (SAARF, 2002b).
Other media, such as television, outdoor media and monthly magazines
have some exposure - see Figure 3-8 (SAARF, 2004).

Figure 3-7 Media coverage of MGM-group 1
100
80
f1.
.
~
.. 60
o
S' 40
20
Source: Adapted from SAARF (2002b).
-
Figure 3-8
100
Media coverage of MGM-group 2
As Figure 3-9 indicates, there is an increase in readership for newspapers
and monthly magazines. Exposure to radio remains high, and television
reach increases persistently. Exposure to train advertising (as part of
outdoor media) is becoming relatively limited (SAARF, 2004).

100
f1. 80
!
:a. 60
:I
.a 40
20
..,....
o
(,\'(\8'«\3 <ft.3~(S _.,).'(\e'O f'C\e\
"e"''01'' ~3C)- \'(\\8
An increase in exposure to weekly newspapers and monthly magazines is
visible. Radio and television continue to have high exposure. Except for
exposure to trains, all forms of outdoor media have a high reach (SAARF,
2002b). Figure 3-10 illustrates the media coverage of MGM-group4.

100
f1.
80
!:I
I)
60
0
i 40

As shown in Figure 3-11, readership of newspapers and outdoor exposure
are higher. Exposure to daily newspapers is becoming apparent, with radio
and television exposure maintaining relatively the same (SAARF, 2004).
100
o
C\(\e«\9> ~9>~e'(S _.,\(\es {(\e\
"e s\" \tt9>q- \(\\e
Readership levels of newspapers and magazines continue to rise. Radio
and television exposure remain at high levels, with outdoor media rising
slightly (SAARF, 2004). See Figure 3-12 for the media coverage of MGM-
group 6.
As shown in Figure 3-13, radio and television remains at a high level, with
print and outdoor media increasing steadily. Cinema going and Internet
usage continue to expand its coverage (SAARF, 2002b).

100
80
'if!.
!:I 60
I:
.9' 40
20
o
C\(\e'«\9. ~9.getS af.\(\es
~e"'s ~9.C)

100
80
fs!.
!:I 60
II
& 40
.a
20
Source: Adapted from SAARF(2002b).
As seen in Figure 3-14, exposure to print and outdoor media is constantly
increasing. Reach of cinemagoers and Internet users are building up, with
exposure to radio and television remaining high (SAARF,2004).

100
oe. 80
!
::::I
.. 60
0
a.
.a 40
20
o
~a oftB(5 .~8S «\8\
C\(\8 "8 59a"'- ",aq,fP.\ \(\\8
3.5.3 MGMs Scale and Society
The MGM is, just as the SU-LSM, a stronger differentiator in some areas than in
others. The MGM has a definite relationship with demographics such as age,
population group, education and income (SAARF, 2004).
· Age is related to the MGM, with the youngest age group (16 to 24 years)
falling into MGM-groups 3 to 5. As people get older, however, they tend to
move into higher MGMs (SAARF, 2004). Figure 3-15 illustrates this
relationship.
· Population group: As shown in Figure 3-16, Blacks dominate MGM-

groups 1 to 6. The amount of Coloureds, Indians and Whites increases as
the MGM-scale increases. The correlation of population group with the
MGM is therefore much the same as with the SU-LSM. Most black people
fall into the lower MGM-groups, with the higher MGM-groups being more
multi-racial (SAARF, 2004).

Figure 3-15 Age profile of the MGM-scale
100
80
.!!oo
q::
e
...40
20
0
1 2 3 4 5 6 7 8
MGMGroup
I .16-24 D25-34 .35-49 050+ I
Source: SAARF (2002b).
Figure 3-16 Population group profile of the MGM-scale

100
80
~
,!tOO
q::
e
...40
20
o
1 2 3 456 7 8
MGM Group
. I_Blacks []Whites iiiColoureds Dlndiansl

· Education levels are directly related to the MGM-scale. As the MGM-scale

increases, the education level of the population also steadily increases
(SAARF,2002b).
· Monthly household income is also strongly related to the MGM-scale

(see Figure 3-17). People in higher income categories have more money

available to purchase media, and are therefore more exposed to the
various media types (SAARF, 2004).
Figure 3-17 Monthly household income profile of the MGM-scale

100
80
i/!.
.!! 60
r;:
e
a. 40
20
o
1 2 3 4 5 6 7 8
MGM Group
.. Less than R899

R2500-R6999
c R900-R2499
0 R7000+
. Province: The Western Cape and Gauteng provinces have an overall high
media exposure level, with the majority of their population falling into MGM-
groups 6 to 8. Kwazulu Natal and the Eastern Cape, on the contrary, have
low media exposure levels, falling into MGM-groups 1 to 4 (SAARF,
2002b). The North-West Province's exposure to media is relatively average
(see Figure 3-18), with the majority of its population falling into MGM-
groups 4 and 5.
3.5.4 SU-LSM versus MGM
The MGM and the SU-LSM are strongly related to one another. The SU-LSMs
increase the higher the MGM-scale goes. Table 3-10 indicates the positive
relationship between these two measures (SAARF, 2002b).

-
Figure 3-18 North-West Province media exposure
12
10
.
~8
i6
....
4
2
o
1 2 3 4 5 6 7 8
MGMGroup
Table 3-10 SU-LSM-profile of the MGM-scale

MGM-GROUPS
1 2 3 4 5 6 7 8
1 35 25 14 6 2 1 0 0
2 31 30 23 16 8 3 1 0
UJ 3 19 20 24 21 15 8 4 1
Q.
:J 20 15 8 2
0 4 10 14 20 22
0::
C) 5 4 7 11 18 21 17 13 5
.
:E 6 1 3 6 11 18 24 21 15
CI)
...I
. 7 0 1 2 3 6 11 15 14
:J
CI) 8 0 0 0 2 5 9 15 17
9 0 0 0 1 4 8 14 22
10 0 0 0 0 1 4 9 24
All values are expressed as percentages
Both the SU-LSM- and MGM-tools will be applied in this study. The SU-LSM will
be used to determine the LSM-groups into which the selected stores' (Pick 'n
Pay and Shoprite Checkers) customers can be classified. Accordingly, the MGM

will be used to determine the most appropriate types of media to be used to
effectively reach these LSM-groups.
To understand the functioning of the SU-LSM and MGM in practise, some

background information need to be provided on the selected stores. As
mentioned, the two stores selected for researching purposes are Pick 'n Pay and
Shoprite Checkers in Potchefstroom. Both stores are located in the central
business district of Potchefstroom, and are only three blocks apart. These are,
furthermore, the only two major supermarkets in Potchefstroom.
Both stores will be analysed in terms of their target markets and marketing
strategies to determine whether they are targeting the actual markets that
purchase from them, and whether they are targeting these markets in the most
appropriate manner.
3.6.1 Shoprite Checkers
The target market of Shoprite Checkers includes a relatively wide range on the
SU-LSM-index, from LSM-group 3 to 8 (Shoprite Holdings Limited, 2003). By
targeting such a large customer base (broad middle to lower market), Shoprite
Checkers is therefore making use of a mass-market approach (Shoprite. 2004).
3.6.2 Pick 'n Pay
Pick 'n Pay, on the other hand, has a more concentrated marketing approach, by
aiming only at LSM-groups 6 to 8. These LSM-groups include customers with
medium-high living standards (Competiiion Tribunal, 2002).

Certain conclusions can be made regarding the LSM-groups of each of these
stores. Only those LSM-groups relevant to the participating stores - SU-LSM-
groups 3 to 8 -will be discussed.
According to SAARF (2004), these LSM-groups' demographics and media

consumption patterns can be summarised as follows:
a) SU-LSWgroup 3 (14,1% of the adult population)
Demographics
More than 82% of this group is between the ages of 16 and 49 years old,
with 62% living in rural areas (SAARF, 2004). Education levels are low,
with the majority having only some high school (Anon, 2002). Average
household income for this group is approximately R l 318 per month
(SAARF, 2003).
Media consumption
Radio is the dominant medium for this group (92%), with the majority
listening to African Language stations such as Ukhozi FM (25%). Umhlobo
Wenene FM (19%), Lesedi FM (18%), Motsweding FM (16%) and METRO
FM (16%) (SAARF, 2004).
Television is also a popular medium, with the majority of the viewers (93%)
watching SABC 1 (SAARF, 2004).
The most popular newspapers and magazines being read by this SU-LSM
group, includes the Bona (13%), Soccer Laduma (8%), Kickoff (7%),
Sowetan (6%), City Press (6%), Drum (6%) and Sunday Times (4%)
(SAARF, 2004).

The reach of cinema advertising is barely visible, while outdoor
advertising's importance increases. Some outdoor media include notice
signs on stores (79%), billboards (69%), taxis (58%), trucks (35%) and bus-
borne signs with a 25% reach (SAARF, 2004).
General
Approximately 85% of this group has electricity in their homes, with 66%
having water on-site (SAARF, 2004). Furthermore, this group has only
minimal ownership of durables, with the exception of radio sets (82%) and
stoves (47%) (Anon. 2002).
b) SUlSMgroup 4 (14,0% of the adult population)
Demographics
The majority of this group (30%) is aged 25 to 34 years, with 37% being
rural dwellers and 30% living in metropolitan areas (SAARF, 2004). This
group also has only some high school education, with their monthly
household income averaging at R1 724 (SAARF, 2003).
Media consumption
Radio is still the most popular medium (93%), with Ukhozi FM (25%),
METRO FM (22%) and Lesedi FM (22%) having the highest listenership
numbers. Most of the television audience watches SABC 1 (96%) and
SABC 2 (69%) (SAARF, 2004).
The highest readership numbers include newspapers and magazines such

as Bona (15%), City Press (1I%),
Soccer Laduma (9%), Kickoff (9%), True
Love (9%), Sowetan (8%) and Sunday Times (8%) (SAARF, 2004).
Cinema advertising still has r i l e reach (2%). while outdoor advertising's

reach increases in all sectors - signs on stores (&I%),
billboards (77%),

taxis (67%) and trucks (43%). Train advertisements (11%) have the least
visible reach (SAARF, 2004).
General
Most of this group has electricity (97%), water (85%) and a flush toilet
(59%) on-site (Anon, 2002). Approximately 85% of this group owns a radio,
52% a hi-fi and 78% a television. The majority of this group owns an
electric hotplate or stove (79%) and a refrigerator (69%) (SAARF, 2004).
c) SU-LSMgroup 5 (12,4% of the adult population)
Demographics
Most of this group (55%) is aged 16 to 34 years and livss in urban areas.
Approximately 3% of this group is illiterate, with the majority (47%) having
reached matric or beyond (SAARF, 2004). The average household income
is at R2 421 per month (SAARF, 2003).
Media consumption
Radio exposure remains high (94%), with the most popular radio stations
being METRO FM (29%), Ukhozi FM (23%) and Lesedi FM (22%) (SAARF,
2004).
All SABC stations' popularity is increasing (SABC 1, 96%; SABC 2, 72%;

SABC 3,48%), with e.tv having a 69% viewership (SAARF, 2004).
The most popular print media includes newspapers such as Crty Press
(13%), Sunday Times (11%), Sowetan (10%) and Soccer Laduma (9%).
Popular magazines include Bona (14%), True Love (11%) and Drum (10%)
(SAARF, 2004).

Compared to SU-LSM 4, the reach of outdoor media continues to increase,
specifically bus shelter advertising (33%). The reach of cinema advertising
(3%) still has a relatively low reach (SAARF. 2004).
General
Almost everyone in this group has water (97%), electrictty (99%) and a
flush toilet (85%) available. The majority has a television (91%), radio
(88%) or hi-fi (72%), stove (72%) and fridge (89%) in their possession
(Anon, 2002).
d) SUSLMgroup 6 (12.5% of the adult population)
Demographics
Most people in this group are between the ages of 25 to 34 years (28%)
and reside in urban areas. The majority (37%) has passed matric, with only
2% having completed universtty (SAARF, 2004). This group's household
income averages R3 897 per month (SAARF, 2003).
Media consumption
METRO FM and Ukhozi FM are still the most popular radio stations,
respectively having a 35% and 17% listenership. E.tv is the most popular
television station with an 81% reach. M-Net is becoming more popular with
a 15% reach (SAARF, 2004).
Newspapers and magazines with the highest readership numbers include

Sunday Times (18%), City Press (17%), Edgars Club Magazine (14%),
True Love (14%). Sowetan (13%), Bona (13%), Soccer Laduma (lo%), The
Citizen (3%) and The Star (3%) (SAARF, 2004).
The reach of outdoor advertising remains relatively the same, with an

increase in some of the less widely seen sites - bus shelters (44%) and

trailers (39%). Cinema viewing is becoming more popular, with cinema
advertising having a 7% reach of this group (SAARF, 2002b).
General
Water, electricrty and flush toilets are generally available to this group
(Anon, 2002). This group has the average number of durables in their
possession, some of which include a television (96%), radio (90%) or hi-fi
(75%), video cassette recorder (48%), electric stove (91%), microwave
(45%). refrigerator (96%), washing machine (38%), cell phone (33%) and
motor vehicle (13%) (SAARF, 2004).
e) SU-LSMgroup 7 (6,3% of the adult population)
Demographics
This group's age ranges from 25 to 49 years (53%). The majority of this
group (99%) has been to school, with 41% having completed matric
(SAARF, 2004). The average monthly household income of LSM-group 7
is estimated at R5 859, and 60% lives in urbanised areas (SAARF, 2003).
Media consumption
Popular radio stations include METRO FM (28%), Radiosondergrense
(RSG) (10%) and East Coast Radio (9%). SABC Iis still the most popular
television station (87% reach), while e.tv, SABC 2, SABC 3 and M-Net
respectively have 85%, 81%, 73%. and 25% penetration of the total
viewership (SAARF, 2004).
Readership of newspapers is at 77%, with the most popular newspapers

being the Sunday Times (26%), C i Press (16%), Sowetan (10%) and
Rapport (10%). Magazines' readership level is at 63%. with Edgars Club
Magazine (18%), Sunday Times Magazines (17%) and You (17%) being
the most popular magazines (SAARF, 2004).
-
Cinema attendance - cinema reach - of this group is at 12%. Outdoor
advertising such as billboards, signs on shops and on taxis is of less
importance to this group. However, outdoor advertising such as bus
shelters, trailers, trucks, buses and trains have relatively more importance
(SAARF, 2004).
General
This group has general possession of the majority of familiar durables, and
has full access to services (Anon, 2002).
f) SU-LSMgroup 8 (5,8% of the adult population)
Demographics
The major part of this group (55%) is aged 35 years and older, with 95%
being urban dwellers. Education levels are slightly higher than SU-LSM 7,
with the majority having matric or higher (SAARF, 2004). The monthly
household income of this SU-LSM averages R8 053 (SAARF, 2003).
Media consumption
The reach of television (98%) is slightly higher than that of radio (91%), but
both mediums have exceptionally high reach. The highest listener
penetration is for METRO FM (17%), Radiosondergrense (RSG) (13%),
94.7 Highveld Stereo (13%), 5fm (12%) and Jacaranda 94.2fm (10%). E.tv
(83%) has the highest television viewership, with the SABCchannels
having relatively similar reach of 76% to 79%. M-Net has a 36% reach in
this SU-LSM-group (SAARF, 2004).
Newspapers' (81%) readership remains higher than that of magazines

(70%). The most preferred newspapers and magazines to be read include
Sunday Times (28%), Huisgenoot (21%). You (20%), Edgars Club
Magazine (19%), Rapport (15%), City Press (7%J Sowetan (5%), Cape
Argus (5%). The Star (5%) and Die Burger with a 5% reach (SAARF, 2004).

Cinema attendance increases to 17%. however, outdoor reach decreases
slightly. Advertising on bus shelters, trailers and trucks are the only outdoor
reach that increases (SAARF, 2004).
General
This SU-LSM generally has full access to services, as well as full ownership
of durables (Anon, 2002). The majority of this group owns durables such
as radio or hi-fi, television, M-Net (14%) or DStv-subscription (lo%), video
cassette recorder (78%), personal computer (22%). microwave (go%), fully
automatic washing machine (58%), freezer (50%) tumble dryer (21%),
dishwasher (3%), vacuum cleaner or floor polisher (73%), motor vehicle
(50%) and 52% have a cell phone (SAARF, 2004).
Shoprite Checkers and Pick 'n Pay have been analysed in further detail by
means of the questionnaires completed by their customers. According to these
questionnaires, conclusions were made regarding the actual LSMgroups in
which the customers of the stores can be classified.
The LSM is a market segmentation tool, specifically designed for the South
African market. However, neither race nor income is considered as variables in
this segmentation tool.
The LSM moved from its initial limited thirteen variables and eight groups in the
1989 1 1990 survey, to its current 29 variables and ten groups in 2001. Each
stage of the LSMdevelopments over time -from 1989 to 2001 -was discussed,
with reference to the various variables and number of groups.
The SU-LSM-variables were identified, and the method to calculate the SU-LSM-
score was explained. The connection between the SU-LSM and society was
discussed by analysing SU-LSM's relationship with socioeconomic factors.

Furthermore, the SU-LSM can be used in the selection of appropriate media for
a selected target market. The eight MGM-groups -that are related to the level
of media exposure of consumers - were analysed in terms of exposure to
different media types as well as in terms of its relationship with various socio-
economic factors. The connection between the SU-LSM and MGM were
compared, to ensure that both these measurement tools could be used in this
study.
Finally, the stores selected to participate in this study were analysed in terms of
their respective target markets, and marketing strategies. Consequently, these
stores' current target SU-LSMgroups were summarised, as to provide
background information for the empirical conclusions and recommendations to
be made.

CHAPTER 4
EMPIRICAL METHODOLOGY AND EXPOSITION
The theoretical aspects of market segmentation and the Living Standards

Measure (LSM) market-segmentationtool were studied in the previous chapters.
In this chapter, the theory will be implemented and tested in practice by
discussing the empirical methodology and the interpretation of the results of the
questionnaires.
4.2 EMPIRICALMETHODOLOGY
Selecting an appropriate sample for the study included identifying the most
suitable population, determining the sample to be included, and the size the
sample should be in order to be representative.
4.2.1 Study Population and Sampling
The population of this study consists of the customers of Pick 'n Pay and
Shoprite Checkers in Potchefstroom. As mentioned, however, the exact number
of customers for each of these stores is unknown, and it was therefore
impossible to accurately determine the population size.
In order to determine a valid and representative sample size, a formula provided

by the Statistical Consultation Services (SCS) of the North-West University
(Potchefstroom Campus) was used. This formula is commonly used to
determine the size of a relationship sample, such as this one.
Chapter 4: Empirical Methodology and Exposition 85

Formula (Steyn et a/. , i998:397):
According to SCS, if the population size is unknown, the p-value (proportion) can
be accepted to be %. Thus, if p = %, the final formula used to ensure the
accuracy of the study, was:
n - sample size
e - deviation %
1
The maximum deviation allowed for the calculation of the sample size was 5%.
Thus, with a 5% deviation (e = 0.05), the sample size should be a minimum of
400 customers for each store - 800 respondents in total.
Due to the nature of this study, no specific sampling method could be used. The
only means, by which a viable and representative sample could be ensured,
were to survey respondents at different days of the week, and at different times
of the day. By conducting the empirical research over a period of three weeks
furthermore ensured unbiased research. Appendix A provides the schedule
according to which the empirical research was conducted.
4.2.2 Questionnaire Design
The design of the questionnaire used in this study was extremely important.
Since respondents were surveyed while shopping, the questionnaire needed to
be kept relatively short. This was necessary to prevent that the respondents
became irritated and refused cooperation.

Due to the diverse nature of the population -different genders, ages, population
groups and income groups - the questionnaire was compiled in such a way as to
be presentable to the entire population of this study.
The questionnaire itself consists of two sections:
The first section gathered information on the demographic aspects of

respondents. This included demographics such as area of residence,
population group, education, age, gender and monthly income. These
questions are based on the demographic information the SU-LSM has on
its various LSM-groups (SAARF, 2004 - Table 38). Multiplechoice
questions were used to gather this information.
The second part was used to classify respondents according to their living
standards, by making use of the SU-LSMclassification scorecard. The
LSM-variableswere applied as dichotomous questions in this section.
The statistical consultation sewices of the North-West University (Potchefstroom

Campus) performed the technical preparation of the questionnaires to ensure
that the questionnaires were statistically correct and that they would provide
accurate and meaningful statistical results.
As mentioned, the minimum sample size for this study to be viable is 800
respondents - 400 for each store. As Table 4-1 indicates, a total of 980
respondents were surveyed, 490 at each store. The study is therefore even
more viable, since the deviation is less than the initial 5%.
Table 4-1 Amount of respondents
STORE AMOUNT PERCENTAGE

Shoprite Checkers 490 50
Pick 'n Pay 490 50
TOTAL 980 100
Chapter 4: Empirical Methodology and Exposition

The customers of the selected stores were surveyed at the entrance of each of
these stores. Surveying was conducted over a three-week period (from Monday,
26 July 2004 to Sunday, 15 August 2004) in order to obtain a high response rate
and to ensure that the diversity of the population is reflected.
Four fieldworkers (two at each store) performed the surveying to ensure that
each respondent understood the questions. Furthermore, the researcher was
continuously present to clarify any misunderstandings.
The data, gathered from the completed questionnaires, were entered into the Epi
Info 6 version 6.02 - October 1994) programme during the surveying period.
The aforementioned data was provided to the Statistical Consultation Services,
where this data was processed into meaningful and useful information, by
means of the SAS and Statistics systems.
The information obtained from the participating respondents was then analysed
and evaluated in order to make relevant assumptions that could contribute to
reaching the study's main objective.
4.2.3 Statistical Analysis
Various statistical tests were conducted to determine the relationships between

the identified factors, and to test the reliability and validity of the questions asked.
a) Cronbach coefficient alpha analysis

This was the first type of statistical analysis conducted in the study. This
analysis was used to estimate the reliability of the SU-LSM-variables used
in the questionnaire (Appendix 8, question 8), by determining the internal
consistency of the average correlation of these variables. If the coefficient
between each pair of variables has a value of 1, the coefficient alpha will
have a maximum value of 1. The larger the overall alpha coefficient, the
more likely those items contribute to a reliable scale (Anon, 2004a).

b) Frequency analysis
This procedure includes the general frequency tables on the various
questions asked in the questionnaire. These tables consist of frequency,
percentage, cumulative frequency and cumulative percentage of the
respondents' answers.
c) Multi-way frequency analysis

In this study, the customers of Shoprite Checkers and Pick 'n Pay were
compared in terms of their SU-LSMclassifications. Therefore, a
relationship analysis was done between the variables of these two stores.
To determine whether the relationships are practically significant, Chi-
square tests were done.
2
The effect size is given by w = g , where X is the usual Chi-square
statistic for the contingency table and n is the sample size (Steyn, 1999;
Steyn, 2002). In the case of a 2 x 2 table, as is the case in this study, the
effect size (w) - which is independent of sample size - is given by the
phi (4 ) -coefficient. Cohen (1988:109-111) gives the following guidelines
for the interpretation of such results:
small effect: w S 0,l (practically insignificant)

medium effect: 0.3 S w S 0,5
D large effect: w 2 0,5 (practically significant)
d) Correspondence analysis
Correspondence analysis is an exploratory technique for analysing cross-
classifications of two or more categorical variables (multi-way frequency
tables) by representing it geometrically. To determine how row and column
categories from a multi-way frequency table interact with one another, they
can be plotted on the same diagram (biplots).

Row and column categories that have similar profiles will appear in close
proximity on the plot. A value close to zero on the plot indicates no
interaction (association) between the row and column categories. If, for
example, row category 1 is closest to column category 2 on the plot,
nothing can be determined about their magnitude of interaction in an
absolute sense. It can only be interpreted in relative terms, that is, row
category 1 is relatively more likely to interact with column category 2
(Bartholomew et ab, 2002:90).
4.2.4 Reliability of the Study
As discussed, Cronbach's coefficient alpha was used to test the reliability of the
SU-LSM-variables. As Table 4-2 indicates, the correlation between each pair of
variables is more than 0.7, and the average alpha coeff~cientis almost 0.8.
From the aforementioned information, it can be said that this testing method -
the 29 variables of the SU-LSM - is highly significant. These variables therefore
have a strong correlation with one another, and make the study in general
reliable.
Once the importance of market segmentation and how the LSM operates is
understood, the surveying results can be interpreted to determine the objectives
of the study. In this section, the information gathered from the questionnaires
will be analysed.
Firstly, the general information on the respondents, which includes area of

residence, gender, population group, education, age and monthly income, is
studied. Secondly, to successfully classlfy the respondents, their living
standards are analysed according to the SU-LSM-scorecard (see Appendix 6).
Chapter 4: Empirical Methodology and ~ x p s i t i o n 90

Table 4-2 Cronbach's coefficient alpha of variables
. ~ ~.
.
1. Hot running water 0,757061

2. Washing machine 0,748991
3. Fridge IFreezer (combination) 0.769619
4. No domestic in household 0,816123
5. 0 or 1 radio set in household 0,803924
6. Flush toilet in- Ioutside house 0,769679
7. Video cassette recorder 0,755129
8. Motor vehicle(s) in household 0,747043
9. Personal computer 0,760241
10. No cellphone in household 0,798243
11. Electric stove 0,767345
12. Tumble dryer 0,750239
13. Home telephone 0,756157
14. Traditional hut 0,806820
15. Vacuum cleaner IFloor polisher 0,750055
16. Television set(s) 0,767887
17. Microwave oven 0,754440
18. M-Net IDStv subscription 0.757377
19. Rural rest (exdudii Western Cape 8 Gauteng Rnal) 0,787921
!O. Hi-fi IMusic centre 0,766544
!I. Built-in kichen sink 0,756438
!2. Home security service 0,758913
!3. Deep freezer 0,756054
!4. Water in home or on stand 0,773483
!5. Sewing machine 0,761019
!6. Dishwasher 0,760278
!7. Electricity 0,775206
!8. Live in Gauteng 0,776105
!9. Live in Western Cape 0,776766
berage alpha coefficient 0,775998

4.3.1 Demographic Information
To test the accuracy of the SU-LSM market-segmentationtool on the customers

of the selected stores, it was important to gather the customers' general
information. The following tables represent the analysis of the respondents'
demographics, including the total for each store (Pick 'n Pay and Shoprite
Checkers) as well as the overall total.
As shown in Table 4-3, more than 80% of the total amount of respondents
resides in the North-West Province. The aforementioned is probable, since the
study was only conducted at Shoprite Checkers and Pick 'n Pay in
Potchefstroom (North-West Province). However, 18% of the total amount of
respondents resides outside of the North-West Province. This is likely, since
Potchefstroom is home to the North-West University (Potchefstroom campus),
which hosts a substantial amount of students from around the world.
Nevertheless, as can be seen from the w-value, there is practically no similarity
between Shoprite Checkers and Pick 'n Pay's respondents and their province of
residence (w-value = 0,1709).
Table 4-3 Province of residence distribution for respondents
PROVINCE SHOPRITE
CHECKERS
1 PICK 'N PAY I TOTAL
Westem Cape 1 3 2
Northem Cape 1 1 1
Free State 3 3 3
Eastern Cape 0 0 0
KwaZulu-Natal 1 2 2
Mpumalanga 1 3 2
Limpopo 0 2 1
Gauteng 5 9 7
North-West 88 77 82
W-value = 0,1709 (practically insignificant- no similarity)

Table 4-4 indicates that males are becoming more involved in the shopping
process. Although the majortty of respondents (59%) were female, they were
not much more than the male respondents (41%). Therefore, the similartty of
the gender distribution between Shoprite Checkers and Pick 'n Pay's
respondents are practically insignificant (w-value = -0,0663), even though
Shopriie Checkers had slightly more female respondents.
GENDER
Male
Female
C W-value = -0,0663 (practically insignificant - no similarity)
Even though both stores are located in the CBD of Potchefstroom and are only
three blocks apart, the majority of the total respondents were White (see Table
4-5). However, it is clear that Shoprite Checkers had significantly more Black
respondents (50%) in relation to the other population groups. Pick 'n Pay, on the
contrary, had mainly White respondents (72%). The Coloured and Indian
population groups were not notably represented at either of the stores. As the
w-value indicates, the similarity between the respondents' population group and
the selected stores are practically significant (w-value = 0,4059).
- - - -
Chapter 4: Empirical Methodology and l&osi&on 93

The aforementioned relationship between the respondents' population
distribution and the respective stores is clearly illustrated in the correspondence
analysis in Figure 4-1.

respondents' populatlon group and choice of store
As indicated in Table 4-6, 38% of the respondents have matric, with 39% post
matric. In the case of Shoprite Checkers, 33% have less than matric, while 13%
of Pick 'n Pay shoppers have less than matric. Pick 'n Pay's respondents overall
have higher education levels than the respondents of Shopriie Checkers, with
87% having matric or higher. Statistically, the relationship between level of
education and choice of store are practically insignificant (w-value = 0,2743).
The respondents' ages varied from young to old. However, as Table 4-7
signifies, more than 60% of the total amount of respondents was between the
ages of 16 - 34 years. The aforementioned is understandable, since
Potchefstroom is home to a considerable amount of students. However,
Shoprite Checkers and Pick 'n Pay's respondents can - as seen from the w-
value - not be distinguished by means of age distribution.

Table 4-6 Education distribution of respondents
No schooling 2 1 1
Some primary schooling 2 0 1
Primary schooling 4 0 2
Some high schooling 25 12 19
Matric 35 40 38
Technikon I 6( 8 1 7
Diploma or Certificate I 15 1 17 1 16
University degree 11 I 22 1 16
W-value = 0.2743 (small ~racticallvsionificance effect)
Table 4-7 Age distribution of respondents
The variation in the respondents' monthly income is shown clearly in Table 4-8.
The largest percentage of Shoprite Checkers's respondents (37%) fell into the
lowest income category, earning less than R500 per month. Twenty-one percent
of Pick 'n Pay's respondents earn between R4 000 and R6 999 per month, with
18% earning less than R500 per month. The similarity of monthly income
between the respondents of Shoprite Checkers and Pick 'n Pay can be accepted
to be practically significant (w-value = 0,2987).

Table 4-8 Monthly income of respondents
Figure 4-2 clearly illustrates the relationship between the respective stores and
monthly income. The numbers used in Figure 4-2 represent the various
categories of monthly income as allocated in Table 4-8 (for example: 1 = "Less
than R500").

respondents' monthly income and choice of store
0 Retailer Ll Income group

4.3.2 Living Standards Information
To determine the LSM-groups according to which the customers of the related

stores can be classified, their responses on the LSMquestions had to be
analysed. After calculating the respondents' LSM-scores - by means of the
process discussed in Chapter 3 (paragraph 3.4.2) - their respective SU-LSM-
groups could be determined.
As can be derived from Table 4-9, in total more respondents fall into SU-LSM-
group 6 than in any of the other groups (see Table 4-9). SU-LSM-groups 1 to 5
can almost be eliminated, since nearly none of the respondents complied with
these five groups. Although the majonty (50%) of Shoprite Checkers's
respondents fall into SU-LSM-group 6, the second highest group is SU-LSM-
group 10 (17%). It would therefore be reasonable to conclude that Shoprite
Checkers's respondents (target market) in Potchefstroom range from SU-LSM-
group 6 to 10.
Table 4-9 Contingency: The similarity between the respondents of

Pick 'n Pay and Shoprite Checkers's SU-LSMgroups
SHOPRITE PICK'N PAY

CHECKERS
1 All values are expressed as percentages

W-value = 0,2824 (medium practically significant effect)

As for Pick 'n Pay, it is apparent that its respondents are distributed over SU-
LSM-groups 6 to 10. Although Pick 'n Pay has slightly more respondents in SU-
LSM-group 10, the w-value indicates only a medium practically-significant effect.
Thus, meaning that the type of store might have some connection to the various
SU-LSM-groups. Since none of the respondents could be classified into SU-
LSM-group 4 or 5, these related groups will be left out for the remainder of the
study.
From Figure 4-3 it is apparent that Shoprite Checkers's respondents are mainly
classified into SU-LSM-group 6, and Pick 'n Pay's respondents into SU-LSM-
groups 7 to 10. It is also clear that SU-LSM-groups 1 to 3 are standing alone,
since almost none of the respondents were classified into these groups.

respondents of Pick 'n Pay and Shoprite Checkers's SU-
a6
Q4 ~
a2
S ao
-
~
-a4
8
-a&
-a8
-1.0 I;&
3
-12
0 Retailer 0 LSM group

It is clear from Table 4-10 that most of the SU-LSM-groups reside in the North-
West Province. The aforementioned is understandable, since the study was
only conducted on the customers of Shoprite Checkers and Pick 'n Pay in
Potchefstroom (North-West).
The w-value furthermore indicates a medium practically-significant effect and it

could therefore be accepted that there is a similarity between the respondents'
province of residence and their SU-LSM-groups.

SU-LSMgroups and their province of residence
0 100 2
0 93 74 76 83 72
Gender, on the other hand, cannot be used as a determinant of SU-LSM-groups

(see Table 4-1 1). Even though females seem to be dominating the lower SU-
LSM-groups, the w-value indicates that this relationship is practically
insignificant.

SU-LSMgroups and their gender
SU-LSM-GROUPS
GENDER
1 2 3 6 7 8 9 1 0
Male 33 20 0 36 47 47 47 42
Female 67 80 100 64 53 53 53 58
1 All values are expressed as percentages 1
Although population group (race) is not included as a variable of the LSM index,
it was tested to determine its relation with the living standards of the
,
respondents.
As shown in Table 4-12, the Black population correlates strongly with the lower
SU-LSM-groups (groups 1 to 6) and, however low in numbers, it also correlates
with some of the higher groups. The White population group, on the contrary,
correlates stronger with the higher SU-LSM-groups (groups 7 to 10). Most of the
Coloured population (60%) falls into SU-LSM-group 2, with the small Indian
population being represented in the higher SU-LSM-groups.

SU-LSMgroups and their population group
I W-value = 0,6795 (highly significant)

The w-value indicates that the correlation between the respondents' SU-LSM-
group and population group is practically significant (large effect). Thus, even
though race is not a denominator of SU-LSM, the respondents of this study still
indicate the reality.
As illustrated in Figure 44. it is apparent that the Black and Coloured population
correlates strongly with SU-LSM-group 6. The Indian and White population, on
the other hand, strongly correlate with SU-LSM-groups 9 and 10. Population
group 5, which represents any other population group, cannot be related to any
specific SU-LSM-group.
The numbers used in Figure 4 4 represent the various population groups as

allocated in Table 4-12 (for example: 1 = 'Black?.
Figure 44 Correspondence Analysis: The similarity between the

respondents' SU-LSMgroups and their population group
Dinrrsicn1; 6 g 6 d u e ,42828 (8e7FRbdI&@

0 LSM group
0 Population group

Table 4-13 indicates that SU-LSM-groups 1 and 2 have the lowest education
levels. The higher SU-LSM-groups (groups 7 to lo),on the other hand, have
higher education levels. The average level of education is having some high
schooling or having matric. However, the majority of the SU-LSM-groups have
some high school or higher education levels. Education levels have a strong
correlation with the various SU-LSM-groups (w-value = 0,6535).

SU-LSMgroups and their education levels
1 All values are expressed as percentages 1
The graphic illustration in Figure 4-5 reviews the abovementioned correlation

between respondents' SU-LSM-groups and their education levels. It is clear that
the lower SU-LSM-groups correlate stronger with the lower education levels, and
the higher SU-LSM-groups with the higher education levels. The numbers used
in Figure 4-5 represent the various education levels as allocated in Table 4-13
(for example: 1 = "No schooling").

respondents' SU-LSMgroups and their education levels
C
Urrmginl; fipndusr.~aO(787l%d1rsrti@
0 SU-LSM group
Education level
c A.
Age and the SU-LSM-groups of respondents have a medium significance effect

(see Table 4-14). SU-LSM-groups 1 to 3 are relatively young, falling in the "16
to 34 years old" age group. However, the higher the SU-LSM-groups, the more
distributed the age groups become.

SU-LSMgroups and their age
W-value = 0.2532 (medium ~racticallvsianificant effect)

As indicated in Table 4-15, the lower SU-LSM-groups (1 to 3) tend to have a low
monthly income. SU-LSM-groups 6 to 10 are more distributed over the various
income groups, with the majority of group 10 (65%) earning more than R4 000
per month. The significance effect of the correlation between the respondents'
SU-LSM-groups and their monthly income is relatively large (w-value = 0,6254).

SU-LSMgroups and their monthly income
7 1 R7000-R11999 1 0 1 01 01 11 21 121 211 26

8 I R12 000 or more I 01 01 01 11 01 2 ) 81 20
1 W-value = 0,6254 (practically significant) 1
As illustrated in Figure 4-6, SU-LSM-group 6 relates strongly with the lower
income groups (groups 1. 2 and 4). SU-LSM-groups 7 and 8 correlate with
income group 5, SU-LSM-group 9 correlates with income group 6, and SU-LSM-
group 10 correlates with the two highest income groups (groups 7 and 8). The
numbers used in Figure 4-6 represent the various income groups as allocated in
Table 4-15 (for example: 1 = "Less than R500").
Thus, even though income is not a SU-LSM-variable or determinant, it is clear

that income and SU-LSM do relate strongly with one another. The higher the
SU-LSM-group, the higher the income levels.

respondents' SU-LSMgroups and their monthly income
Dinrrsim1; m u a .3ilEB(79.73%dlrrrtiap
0 LSMgroup
Income arouD
In this chapter, the research method was discussed and interpretations were
made on the results of the empirical study. These findings should be considered
by Shoprite Checkers and Pick 'n Pay in order to develop effective, efficient and
sustainable marketing strategies for their respective businesses. The next
chapter will include recommendations made to assist these stores in selecting
target markets and developing successful marketing strategies for these target
markets.
Chapter 4: Empirical Methodology and Expasition 105

CHAPTER 5
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
5.1 SUMMARY
AND OVERVIEW OF THE STUDY
This study focussed on the process of market segmentation and two relatively
unknown market segmentation tools, namely the Living Standards Measure
(LSM) and Media Groups Measure (MGM). Emphasis was placed on
segmenting the customers of Shoprite Checkers and Pick 'n Pay in
Potchefstroom.
The primary goal of this study was to determine the actual target markets of
selected stores by means of the LSM-tool, and to furthermore recommend
improved marketing strategies. To achieve this goal, it was essential to examine
market segmentation, LSM and MGM comprehensively.
In chapter one the key terms used throughout the study were defined. The
reasons and need for the study were motivated, and objectives were formulated
to describe the means by which the aim of the study would be achieved. The
population and sample for the study was identified. Finally, the research design
and procedure were clarified, the statistical analysing methods were explained,
and the limitations of the study were listed.
To achieve the first objective -to describe market segmentation by emphasising

its importance and discussing the market segmentation process -a detailed
study of market segmentation was conducted in chapter two. This chapter
commenced by providing a background on and emphasising the importance of
market segmentation. Then, the entire market segmentation process was
discussed in detail.
Chapter 5: Summary, Conclusions and Recommendations 106
-
Since customers form the core of a business, it is essential for the business to
identify its customers and their needs. Even though customers differ in terms of
their wants, needs and behaviours, they can be grouped together according to
some similarities in their character. The aforementioned grouping of
homogenous customers is known as market segmentation. The most important
objective of market segmentation is to improve the marketing performance of a
business.
To conduct successful market segmentation, a business should identify the

market(s) of interest. A market can consist of either consumers or businesses
that want or need a certain product offering, and that are able and willing to pay
for the required product offering. Since a business wants a homogeneous group
of customers that are distinct from other segments, segmentation bases can be
combined to effectively segment the selected market.
As discussed, successful market segmentation requires comprehensive

evaluation, and has to comply with certain criteria, characteristics and rules.
After the segments have been evaluated, the business should select the
segment that complies best with the objectives and activities that it wishes to
target. The business should determine a strategy to target the selected segment
and should identify ways to distinguish itself in the competitive market
environment.
Finally, in order to develop an effective marketing strategy, the business should

develop an effective marketing mix, customised to tit the business and marketing
objectives. The four marketing mix elements - product, price, place and
promotion - are essential to achieve the business' marketing objectives and to
maintain a competitive advantage in the market environment.
The second objective - to explain the application of the LSM- and MGM-tools in
identifying target markets and selecting media to reach these target markets -
was achieved in chapter three of this study. This chapter mainly discussed the
- -

nature of the South African Advertising Research Foundation (SAARF), the
development and application of the LSM- and MGM-tools.
SAARF is a non-profit organisation that developed both the LSM- and MGM-
tools. SAARF is managed by a joint industry body that consists of the
advertising, marketing and media industries in South Africa. This joint body
assists SAARF in achieving its main objective, namely to provide a common
research currency to aid in the buying and selling of advertising time and space.
The LSM is South Africa's first multivariate market-segmentation tool, used for
the first time in 1989. This tool is unique since it was developed specifically for
the diverse South African market. Furthermore, the LSM differentiates people
according to their living standards, instead of their income or race. The LSM
developed substantially from what it was in 1989, to what it is now.
Currently the LSM-index, now known as the SAARF Universal LSM (SU-LSM),
comprises of 29 household variables. The SU-LSM-index furthermore ranges
from SU-LSM-group 1 (lowest living standards) to SU-LSM-group 10 (highest
living standards). As indicated in chapter two, the SU-LSM clearly differentiates
better between some socioeconomic factors than others.
The MGM is also a market segmentation tool, with the main purpose of
optimising the selection of media in reaching certain target markets. The MGM
obtains consumers' media consumption information, by examining the amount
and cornbinations of the different media consumed.
The types of media examined include cinema, newspapers, magazines, Internet,

radio, television and outdoor media. The MGM-scale furthermore consists of
eight groups - from MGM-group 1 (one) with the lowest media exposure levels,
to MGM-group 8 (eight) with the highest media exposure levels. The MGM -
just as the SU-LSM - differentiates stronger between some socio-economic
factors than others.

Shoprite Checkers currently makes use of a mass-market approach, by targeting
the broad-middle to lower market, which includes SU-LSM-groups 3 to 8. Pick 'n
Pay, in contrast, targets customers with medium-high living standards. This
store follows a more concentrated marketing approach, by targeting only SU-
LSM-groups 6 to 8.
The third objective -to conduct an empirical investigation into the application of
the LSM-tool in the selected stores - was reached in chapter four. This chapter
explained the method by which the empirical study was conducted and
interpreted the empirical results. Interpretations of respondents' answers were
demonstrated by means of tables and figures.
The customers' responses to the questionnaires were analysed to determine

their demographic status, as well as their living standards. The respondents of
Shoprite Checkers and Pick 'n Pay were compared in terms of their SU-LSM-
group classification. The questionnaires were therefore used to determine both
Shoprite Checkers and Pick 'n Pay's actual target markets, from which
recommendations regarding their marketing activities can be made.
The overall results. SU-LSM-groups, and the results of Pick 'n Pay and Shoprite
Checkers's customers can accordingly be concluded.
5.2.1 Overview of Results
General conclusions regarding the study entail:
By using the formula of Steyn et a/. (see Chapter I),

the target response
rate was determined at 400 respondents for each store, thus 800
respondents in total. However, 490 respondents were surveyed at each

store amounting to 980 respondents in total. The aforementioned
therefore resulted in a more viable study.
A diverse amount of customers were included in the study. Respondents

varied substantially in gender, age, population group, education levels
and income groups.
Most of the respondents (82%) reside in the North-West Province, since

the study was only conducted in Potchefstroom in the North-West
Province. The remainder of the respondents that reside outside of the
North-West Province, amount for students coming from other parts of the
country.
Gender does not differentiate shoppers significantly. The amount of

female respondents (59%) was slightly more than the male respondents
(41%). but it is clear that males are increasingly participating in the
shopping process.
There were slightly more Whiie (53%) respondents than Black (34%)
ones. The Coloured and Indian population groups, however, were not
significantly representative in the study.
Respondents ranged from the youngest to the oldest age groups.

However, 63% of the respondents were between the ages 16 to 34 years.
Due to the university located in Potchefstroom, this city can be considered
as a "student town". It is thus reasonable that the "16 to 34 yearsn-age
group had the highest amount of respondents.
The SU-LSM-tool was tested in practice by determining the SU-LSM-

classifications of the customers of Shoprite Checkers and Pick 'n Pay in
Chapter 5: Summary, Conclwions and Recommendations 110

Potchefstroom. By means of Cronbach's coefficient alpha, the SU-LSM-index's
variables were proven reliable.
The following conclusions regarding the application of the SU-LSM-tool in the

study can be made:
Almost all of the SU-LSM-groups reside in the North-West Province. This

is, as mentioned, because the study was only conducted in
Potchefstroom in the North-West Province.
There is no significant correlation between the respondents' gender and

their SU-LSM-groups.
The lower SU-LSM-groups (groups 1 to 3) mainly consist of young people,

varying between the ages 16 to 34 years. The higher SU-LSM-groups,
however, vary from young to old.
The population groups correlate strongly with the SU-LSMclassifications.

SU-LSM-groups 1 to 6 mainly consisted of Black and Coloured people,
whereas SU-LSM-groups 7 to 10 mainly comprised of White and Indian
people. Black people are visible in the higher SU-LSM-groups, however
not that strongly. Thus, Black and Coloured people still have lower living
standards than White or Indian people.
SU-LSM-groups 1 to 6 clearly have lower education levels than the higher

SU-LSM-groups. The highest education for the majority of SU-LSM-
groups 1 to 6 is people with matric. SU-LSMgroups 7 to 10, on the other
hand, have education levels that range from "some high schooling" up to
having a university degree. Thus, lower SU-LSM-groups have low
education levels, and higher SU-LSM-groups have higher education
levels.

There is a strong correlation bebeen the respondents' monthly income
and their respective SU-LSM-groups. SU-LSM-groups 1 to 3 have
considerably lower income levels - ranging between less than R 500 to
R1 399 per month -than the higher SU-LSM-groups. The income levels
of SU-LSM-groups 6 and 7 are distributed over the different income
categories. However, the higher the SU-LSM-groups the higher the
income levels become.
The abovementioned conclusions can be compared with the conclusions SAARF

made from the SU-LSM-tool in practice - see Appendix D.
5.2.3 Pick 'n Pay
The following conclusions regarding the customers of Pick 'n Pay in

Potchefstroom can be made:
Pick 'n Pay's customers were mainly White (72%). educated people, with
87% of the respondents having matric or higher education.
The highest income category for Pick 'n Pay's respondents is the 'R4 000
to R6 999 per month" category (21%). Although Pick 'n Pay's customers
tend to be in the higher income groups, 18% of its respondents fall in the
"Less than R500 per month" category. The largest age group of Pick 'n
Pay ranges from 16 to 24 years old (34%). which indicates that these
respondents were mainly students that do not earn a monthly income.
Pick 'n Pay's customers were classified - according to their living

standards - into SU-LSMgroups 6 to 10. The majority of Pick 'n Pay's
respondents were divided between SU-LSM-group 6 (25%) and 10 (27%).
Thus, Pick 'n Pay's customers have medium-high living standards.
- - -

5.2.4 Shoprite Checkers
The responses of the customers of Shoprite Checkers, Potchefstroom, revealed

the following conclusions:
Shoprite Checkers's customers included both Black (50%) and White

(34%) people, with relatively low education levels. Only a third of this
store's respondents have up to some high schooling.
Shoprite Checkers's respondents are generally in the lower income

groups, with 37% earning less than R500 per month.
The majority (50%) of Shoprite Checkers's respondents were classified

into SU-LSM-group 6, with the remaining respondents ranging from SU-
LSM-groups 7 to 10. Thus, Shoprite Checkers's customers also have
medium-high living standards.
As the abovementioned conclusions indicate, the target SU-LSM-groups for both

Shoprite Checkers and Pick 'n Pay include SU-LSM-groups 6 to 10. These SU-
LSM-groups correlate strongly with MGM-groups 6 to 8 - see Table 3-10. Thus,
the type of media used for MGM-groups 6 to 8, can be applied on SU-LSM-
groups 6 to 10.
As mentioned in Chapter 3, the target market of Shoprite Checkers includes SU-

LSM-groups 3 to 8 (see paragraph 3.6.1), and Pick 'n Pay's target market
ranges from SU-LSM-group 6 to 8 (seeparagraph 3.6.2). These stores' target
markets are determined by their respective headquarters, and are therefore
similar countrywide.
-
From the results, it is clear that Shoprite Checkers and Pick 'n Pay's target SU-
LSM-groups in Potchefstroom are different from those assigned to them by their
respective headquarters. As the empirical results indicated, the target market of
Potchefstroom's Shoprite Checkers mainly falls in SU-LSM-group 6, with the rest
of its target market distributed over SU-LSM-groups 7 to 10. Pick 'n Pay in
Potchefstroom's target market ranges from SU-LSM-groups 6 to 10.
In order for these stores to improve their overall performance, the following
recommendations can be made regarding their target markets:
The headquarters of both stores should revise their strategies, by

focussing on provinces rather than countrywide. The headquarters
should therefore consider developing differentiated marketing strategies
for stores in different locations.
Shoprite Checkers and Pick 'n Pay (in Potchefstroom) should rather
target SU-LSM-groups 6 to 10.
For Shoprite Checkers and Pick 'n Pay to target their market of SU-LSM-groups
6 to 10, the stores should concentrate on:
All age groups, with slightly more focus on the 16 to 34 year olds.
Medium to high income categories (R2 500 to R15 000+ per month)
People with 'some high schooling" or higher education levels.
Shoprite Checkers should focus on males in the Black and White
community, whereas
Pick 'n Pay should focus mainly on females in the Whiie community.
As mentioned in Chapter 3 (paragraph 3.4.4), race is not a SU-LSM-

determinant. However, since race was tested in the study to determine its
relation to the SU-LSM-index, conclusions can be made according to which race
the stores in Potchefstroom should target.

The identiid SU-LSM-groups (groups 6 to 10) for Shoprite Checkers and Pick
'n Pay as target market, is a considerably large segment. However large, this
segment's general household possessions or durables - as concluded from
SAARF (2004) - can be identified as in Table 5 1 .
Table 5-1 General household possessions for SU-LSMgroups 6 to 10
As concluded from SAARF (2004), the media listed in Table 5-2 can be applied
to effectively target SU-LSM-groups 6 to 10. In addition, the criteria on MGM-
groups 6 to 8 were used to determine media that can reach these SU-LSM-
groups effectively.

Table 5-2 Media for SU-LSM groups 6 to 10
I 5fm
Jacaranda 94.2fm
Edgars Club Magazine
True Love
Bona
Huisaenoot
"~ ~
You
I
Sunday Times Magazine
Sowetan
Daily The Star
Rapport
I Sunday Tribune
Billboards
I Stores
Bus shelters
Trailers
Taxis IMinibuses
Trains
Buses
Trucks

The abovementioned media can serve as a guideline for Shoprite Checkers and
Pick 'n Pay when selecting media for their target markets. The following
recommendations can therefore be made regarding media selection for Shoprite
Checkers and Pick 'n Pay in Potchefstroom:
TV: Advertisements can be placed on SABC 1 , 2 , 3 and e.tv. M-Net and

DStv can also be used, however, these stations are not suitable for local
advertising. Thus meaning, that Shoprite Checkers and Pick 'n Pay in
Potchefstroom will not be able to effectively reach only the local
community.
Radio: Since Shoprite Checkers's customers were mainly Black people,

Shoprite Checkers can advertise on local "Black" radio stations, METRO
FM and Kfm. Pick 'n Pay can advertise on RSG, Highveld Stereo and
5fm. since their customers were mainly Whiie people.
Magazines: Bona and True Love would be ideal magazines for Shoprite
Checkers to advertise in, since these magazines are mainly directed
towards Black people. Pick 'n Pay can place advertisements in
magazines such as the You or Huisgenoot, which have mainly White
readers.
= Daily newspapers: The Sowetan and Beeld newspapers can be ideal

media for daily coverage.
Weekly newspapers: Sunday Times, Soccer Laduma and Rapport will

be good for provincial coverage. However, local newspapers (in
Potchefstroom) such as the Herald and Gazette, will have the largest
reach of the Potchefstroom community.
Outdoor: Since Potchefstroom does not make as much use of billboards

as in metropolitan areas, advertising on this type of medium might not be

as effective. However, advertisements on taxis (minibuses) might be
extremely effective, especially for Shoprite Checkers that should mainly
target the Black community.
Internet: Internet advertising will be more successful for the higher SU-
LSM-groups, which have the luxury of a personal computer and lntemet.
Pick 'n Pay with its White target market, can therefore consider
advertising on the lntemet.
Freesheets or flyers: Inserts in local papers and flyers that are

distributed directly to the market, can be a successful means of reaching
these stores' target markets. Freesheets are usually delivered directly to
the wmmun~ty'shouses. Thus meaning that Shoprite Checkers can
distribute freesheets directly to the Black community, and Pick 'n Pay to
the White community.
As the study indicates, it can be noted that the strategies developed by the
respective stores' headquarters do not always suit the respective store. Thus,
stores should rather determine their own target markets, as applicable to the
community they are located in. The stores should furthermore develop their own
marketing strategies, using as much local media as possible in these strategies.
From the abovementioned discussion, it can be assumed that all the objectives
of the study were achieved, and that the aim - to determine the actual target
markets of the selected stores by means of the LSM-tool, and to recommend
improved marketing strategies based on the MGM-tool -was accomplished.

ALLEN, G. 1999. Markets. [Web:] httD://ollie.dcccd.edulmrkt2370~ok/mrkt
book.htm [Date of access: 4 May 20041.
ANON. 2002. New SAARF Universal LSMs. [Web:] httD:l/www.fastmovina.m.

[Date of access: 15 July 20041.
za/detail.is~?~roviderNo=1959
ANON. 2004a. Cronbach's coefficient alpha. [Web:] b:/Iieff-lab.aueensu.ca/

stat/saslsasman/sashtmi/~roc/zom~meth.htm[Date of access: 14 September
20041.
ANON. 2004b. Customer needs and wants. [Web:] httD:l/www.tutoRu.net/

glossarvladvanced.as~?a=4&=941[Date of access: 28 May 20041.
ANON. 2004c. How to segment your market. [Web:] hb:lEwww.toolkit.cch.

comltextlP03 1 0 3 6 . a ~[Date of access: 28 May 20041.
ASSAEL, H. 1992. Consumer behaviour and marketing action. 4* ed. Boston

: PWS-Kent. 748 p.
BAGOZZI, R.P., ROSA, J.A., CELLY, K.S. 8 CORONEL, F. 1998. Marketing

management. Upper Saddle River, New Jersey : Prentice-Hall. 718 p.
BAKER, M.J. 2000. Marketing strategy and management. 3" ed. Basingstoke
: Macmillan. 560 p.
BARTHOLOMEW, D.J., STEELE, F., MOUSTAKI, 1.8 GALBRAITH, J.I. 2002.

The analysis and interpretation of multivariate data for social scientists. London
: CRC Press. 263 p.
References 119
BELCH, G.E. 8 BELCH, M.A. 2001. Advertising and promotion: an integrated
marketing communications perspective. 5med. New York : McGraw-Hill. 795 p.
BERENSON, M.L. 8 LEVINE, D.M. 1996. Basic business statistics: concepts

and applications. 6* ed. Englewood C l i . NJ : Prentice Hall. 904 p.
BEVAN-DYE, A.L 2002. Internet malketing in the South African banking

sector. Potchefstroom : PU for CHE. (Thesis - MBA). 118 p.
BLANKENSHIP, A.B., BREEN, G.E. 8 DUTKA, A. 1998. State of the art

marketing research. 2* ed. Lincolnwood, Illinois : NTC Business Books. 454 p:
BRINK, A. 1998. Market segmentation, market targeting and product

positioning. Strydom, J.W., a.Introduction to marketing. Kenwyn : Juta. p.
61-90.)
BULL, N.H. 8 PASSEWITZ, G.R 2003. Finding customers: market

segmentation. web:] www.ohioline.aa.ohi0-state.edu [Date of access: 28 May
20041.
BUSINESS BUREAU. 2002. Market segmentation. web:] htt~:lhnrww.

businessbureau-uk.co.uklsalesmarketina/seamentation.htm [Date of access:
28 May 20041.
COHEN, J. 1988. Statistical power analysis for the behavioural sciences. 2""
ed. Hillsdale. NJ : Erlbaum. p. 109-111.
COMPETITION TRIBUNAL. 2002. In the large merger between: Pick 'n Pay
Retailers (Pty) Ltd and Boxer Holdings (Pty) Ltd. [Web:] httD:lW.com~trib.co.
zaldecidedcases/docl52LMJUL02.doc [Date of access: 21 July 20041.
- - - - -
References 120
CORDER, C. 2001. Media Groups Measure. web:] htt~:lEwww.saarf.co.zal
mediaarou~smeasure.htm[Date of access: 21 July 20041.
COSENZA, R.M. 2004. Positioning. web:] httD://home.olemiss.edul-

i t . ~ of
rcosenzal ~ ~ t s / ~ o s [Date ~ access:
t 30 June 20041.
DETERMAN, M. 1999. Positioning differences. web:] httD:lIvaratek.com/

launchsite/wsition.htrn~mentation[Date of access: 25 June 20041.
DEVEAU, J. 2004. Mass marketing. web:] httD://HMMI.marketin!a~rofs.com/

ealast auestion.as~?astlD=l765[Date of access: 25 June 20041.
DSS RESEARCH. 2004. Market segmentation. [Web:] httD:llwww.dss

.asp [Date of access: 3 May 20041.
research.com/tml~tlresource/~a~edSROl
DUNNE, P.M., LUSCH, R.F. 8 GRIFFITH, D.A. 2002. Retailing. 4" ed.
Mason, Ohio : Thomson Learning. 594 p.
DU PLOOY, G.M. 2001. Communication research: techniques, methods and

applications. Lansdowne : Juta. 384 p.
FILL, C. 1999. Marketing communications: contexts, contents and strategies.

London : Prentice Hall. 656 p.
GILBERT, D. 1999. Retail marketing management. London : Financial Times

Management. 356 p.
GILLIGAN, C 8 WILSON, R.M.S. 2003. Strategic marketing planning. Oxford :

Butterworth-Heinemann. 625 p.
References
HAUPT, P. 2001a. The SAARF Universal Living Standards Measure (SU-LSM)
- 12 years of continuous development. web:] www.saarf.co.za [Date of
access: 2 March 20041.
HAUPT, P. 2001b. How to use LSMs to best advantage. Saturday Star: 13,
Mar. 31.
HAUPT, P. 2002. SAARF - in short. [Web:] httD:lhww.saarf.co.za/annual

re~ort.htm[Date of access: 2 June 20041.
HAUPT, P., SMIT, P. 8 NCUBE, M. 2002. Industry presentation on the SAARF

Universal LSM. web:] http://www.saarf.co.z%~t-htmRsm~resentationFinal
fileslframe.htm [Date of access: 25 June 20041.
HSRC (HUMAN SCIENCES RESEARCH COUNCIL). 2002. Public attitudes in

contemporary South Africa. Cape Town : compress. 182 p.
KOTLER, P. 8 ARMSTRONG, G. 2004. Principles of marketing. 10" ed.

Upper Saddle River, New Jersey : Prentice-Hall. 661 p.
LAMB, C.W., HAIR, J.F. 8 McDANlEL C. 2000. Marketing. 5" ed. Cincinnati,
Ohio : South-Westem College. 773 p.
LANCASTER, G. 8 REYNOLDS, P. 1998. Marketing. Hampshire, London :

Macrnillan. 327 p.
LEDUC, B. 2000. Target marketing strategy: find your own niche market.
[Web:] httD:llwww.soho.o~arketinoArticlesAamet marketino strate4v.htm
[Date of access: 28 May 20041.
LEVY, S.J. 1999. Brands, consumers, symbols & research. California : Sage.
590 p.
References 122
McCARTAN-QUINN, D., DURKIN, M. & STEWART, K. 2004. The marketing
mix and segmentation. [Web]: htt~:IEwww.busmat.ulst.ac.uklmanaaelmark
notes 4.html [Date of access: 29 June 20041.
McDONALD, M. 2000. Market segmentation. Cranfield School of

Management Marketing management: a relationship marketing perspective.
Hampshire. London : Maunillan. p. 77-95.)
MONASH UNIVERSITY. 2004. Market defined. [Web:] htt~:llwww.buseco.

m0nash.edu.a~ [Date of access: 23 June 20041.
NEWMAN, A.J. & CULLEN, P. 2002. Retailing: environment & operations.

London : Thomson Learning. 496 p.
NICKELS, W.G. & WOOD, M.B. 1997. Marketing: relationships, quality, value.
New York : Worth. 633 p.
PERREAULT, W.D. 8 McCARTHY, E.J. 1999. Basic marketing: a global-

~ Boston : Irwin McGraw-Hill. 791 p.
managerial approach. 1 3 ed.
RECKLIES, D. 2001. Why segmentation? [Web:] htt~:lEwww.themanaaer.oml

Marketina/Seamentation.htm [Date of access: 5 March 20041.
RIX, P. 2004. Marketing: a practical approach. 5" ed. Boston : McGraw-Hill.

558 p.
SAARF (SOUTH AFRICAN ADVERTISING RESEARCH FOUNDATION).

2001a. Living Standards Measure. web:] www.saarf.co.za [Date of access: 2
March 20041.
References 123
2001b. All about SAARF and AMPS. web:] www.saarf.co.za [Date of access:
15 June 20041.

2002a. SAARF universal LSM: living standards measure: all you need to know
about LSMs. Bryanston. 66 p.

2002b. The SAARF Media Groups Measure. web:] htt~://www.saarf.co.za/~~t
-htmlMGM~resentation25Nov02fileslframe.htm [Date of access: 21 July
20041.

2003. SAARF - pointing the way in media research. web:] htt~:lhnrww.saarf.
~0.~a/DDt/2002BamD~.~df
[Date of access: 21 June 20041.

2004. Segmentation handbook. Bryanston. [CD-ROM.]
SAS Institute lnc., SASISTAT User's Guide, Version 6, Fourth Edition, Volume
1, Cary, NC: SAS Institute lnc., 1989.943~~.
SHOPRITE. 2004. Shopper profile. [Web:] httD:/lwww.sho~rite.co.~Iesl

19204235/HTMLfiles/2003FinaVindex.htrnl [Date of access: 29 July 20041.
SHOPRITE HOLDINGS LIMITED. 2003. Our business at a glance. (Web:]

htt~:llw w w . s h o ~ r i t e . c o . z a l f i l e s / l 9 2 0 4 2 3 5 / H T M L f i . h t m l [Date
of access: 29 July 20041.
STEYN, A.G.W., SMIT, C.F., DU TOIT, S.H.C. 8 STRASHEIM, C. 1998.

Moderne statistiek vir die praktyk. 6" ed. Pretoria : Van Schaik. 760 p.
References 124
STEYN, H.S. (jr.) 1999. Praktiese beduidenheid: die gebruik van effekgroottes.
Potchefstroom : PU vir CHO. 28 p.
STEYN, H.S. (jr.) 2002. Practically signiiicant relationships between two

variables. SA Journal of Industrial Psychology, 28(3):10-15.
STRUWIG, F.W. 8 STEAD, G.B. 2001. Planning, designing and reporting

research. Cape Town : Masker Miller Longman. 279 p.
TERBLANCH~,N. 1998. Retail management. H a h y House : International

Thomson. 404 p.
THOMPSON. K.N. 2000. Market segmentation & target marketing. web:]

httD://courses.unt.edu/kt36507IsId001.htm [Date of access: 3 May 20041.
TOWNLEY, J. 2001. Consumer markets and buying behaviour. web:] httD://

www.bicvclesb2b.corn/htrnleditorAavslarticlesll~Iv2OOl.htm[Date of access: 25
June 20041.
UTD (THE UNIVERSITY OF TEXAS AT DALLAS). 2004. Criteria for target

market. web:] httD:/lwww.utdallas.edu/-tskimlLecture%20Ne%206.f [Date
of access: 24 June 20041.
WEBBER, S. 2001. The marketing mix. web:] hit~:lldis.shef.ac.uklsheilal

marketina/mix.htm [Date of access: 29 June 20041.
WESTGROUP RESEARCH. 2004. Market segmentation analysis. w : ] hlt~xJ

www.westarou~research.comlresearcWseamentation.html [Date of access: 3
May 20041.
WORTLEY, B 8 MASERAME, T. 2002. Segmentation by Living Standards
Measure. [Web:] h~:I~.acis.aov.za/docsl~0rt~0mlO2seamentsub.
html
[Date of access: 15 July 20041.
YANKELOVICH, D. 1999. New criteria for market segmentation. [Web:]

htt~://www.dan~ankelovich.comlnewcriteria.html
[Date of access: 4 May 20041.
References 126
APPENDIX A
QUESTIONNAIRE DISTRIBUTION SCHEDULE
Monday 1 08:OO - 10:OO 1 16:OO - 18:OO ( 13:OO - 15:OO

Tuesday 1 09:30 - 11:30 ( 13:30 - 15:30 ( 12:OO - 14:OO
Wednesdav 1 12:30 - 14:30 1 14:30 - 16:30 1 13:30 - 15:30
!Thursday 1 OW30 - 11:30 1 15:OO - 17:OO 1 14:OO - 16:OO
Friday 1 12:OO - 14:OO 1 16:OO - 18:OO 1 15:OO - 17:OO
Saturdav 1 09:OO - 11:OO 1 11:OO - 13:OO 1 13:OO - 15:OO
ISunday 1 0900 - 11:OO 1 1O:OO - 12:OO 1 11:OO - 13:OO
Appendix A: Questionnaire Distribution Schedule 127

APPENDIX B
QUESTIONNAIRE
Demografiese Inligting 1 Demographic Information
1) Supermark I Supermarket 5) Opleiding I Education

Shoprite Checkers 1 Ongeskoold I No schooling
- -
Pick 'n Pay 2 Gedeettelik laerskoolI Some
Laerskool 1Primary schooling

2) Provinsie I Province Gedeeltelik hdrskool I Some
high schooling
Matriek I Matric 5
Tegnikon I Technikon 6
Diploma of Setiikaat I
I 7
0 0 s Kaap I Eastern Cape 1 4 Diploma or Certificate
I
I KwaZulu-Natal 1 5 Universlteiigraad I University
degree
6) Ouderdom I Age
-
16 24 jaar I years
25 - 34 jaar I years
35 - 49 jaar I years
3) Geslag I Gender SO+ jaar I years
Manlik I Male 1
Vroulik I Female 2 7) Maandelikse Besteebare Inkomsb
I Monthly Disposable Income
4) Populasie Groep I Population 1 Minder as 1Less than R500 I 1 1

Group
ISwart I Black 1 1 1
Kleurling I Coloured
lndigr I Indian
Blank I White
Ander I Other 5
Appendix B: Quesiionnaire 128

Lewenstandaarde I Living Standards
8) Is die volgende van toepassing op u? IDoes the following apply to you?
1 Lopende warm water IHot running water

2 Wasmasjien I Washing machine
3 Yskas of vfieskas (komblnasie)IFridge or freerer (combination)
4 Bediende in huishouding IDomestic in household
Meer as 1 radio in huishoudingIMore than 1 radio set in
household
1
6 Spoebilet binne of buite huis IFlush toilet in- or outside house I I
1
7 Videomasjien IVideo cassette recorder I
1
26 Skottelgoedwasser IDishwasher I I
27 ( ElektrisiteitI Electricity
28 ( Woon in Gauteng IReside in Gauteng I
1
29 Woon in Wes Kaap IReside in Western Cape I
Appendix B: Questionnaire
APPENDIX C
SUMMARY OF THE SU-LSM-GROUP CLASSIFICATION
SU-LSM-GROUP 1
Gender: Female (58%)

Age: 50 years older (29%)
Schooling: No formal schooling (22%)
Live: Rural (99%)
I
- - - -
Dwelling: Tradlional hut (67%) I
Minimal access to services: electricity (1 1%). water (4%)

Minimal ownership of durables, except radio sets (69%)
I
SU-LSM-GROUP 2
Gender: Female (543%)

Age: 16 - 24 years (33%)
Schooling: Some high school (41%)
Live: Rural (88%)
-.
Electricity (56%), water on plot (30%)

Minimal ownershio of durables. exceDt radio sets 180%) and stoves
Appendix C: Summary of the SU-LSM Classification

SU-LSM-GROUP 3
- -- - - - -
Age: 16 - 49 years (82%)

Schooling: Some high school
Live: Rural (62%)
Earn: R1 318 oer month
I
I
Radio (92%): African Lanauaae stations I

TV: SABC 1 (93%) 1
Outdoor (53%) 1
Electricity (85%), water on plot (66%)

Minimal ownership of durables, except radio sets (82%) and stoves
-
SU-LSM-GROUP 4
1 14% of adult oooulation 1
1Aae: 25 - 34 vears (30%) I

1 Schoolina: Some hiah school or matric 121%) 1
1 Live: Rural 137%) and urban 130%) I
1 Earn: R1 724 m r month I
1 Radio (93%): African Lanauaae stations 1

1 TV (90%): SABC 1 and SABC 2 1
1 Outdoor 156%) 1
1 Electricity (97%), water on plot (85%), flush toilet (59%) I

1 TV sets (78%). hi-fi (52%)l radio (85%), stove (79%), fridge (69%) I

SU-LSM-GROUP 5
Age: 16 - 34 years (55%)

Schooling: Some high school or matric (47%)
Live: Urban 144%)
Earn: R2 421 oer month 1
, Radio (94%): African Language stations

TV: SABC 1 (96%). SABC 2 (72%). SABC 3 (48%). e.tv (69%) I
Weekly newspapers (49%), magazines (40%)
1 Outdoor
I1
Electricity (99%), water (97%). flush toilet (85%)
Ownership of W s (91%), hi-fi (72%) Iradio set (88%), stove (72%),
SU-LSM-GROUP 6
1 12.5% of adult ~ooulation 1
1Aae: 25 - 34 vears (28%) I

1 Schoolina: UDto oost-matric. not universitv (37%) I
1 Live: Urban (57%) I
1 Earn: R3 897 oer month I
1 Wide ranae of commercial 1 communitv radio 194%) I
1 TV 197%): All SABC stations. e.tv I
Daily Iweekly newspapers (66%), magazines (50%)
Cinema (7%). outdoor I
I Electricitv. hot runnina water 162%). flush toilet I
1 Ownership of a number of durables, plus cell phones (33%) I

SU-LSM-GROUP 7
IAae: 25 - 49 vears 153%) 1

1 Schoolina: Matric and hiaher 141%) 1
I Live: Urban 160%) I
1 Earn: R5 859 ~ emonth
r 1
IWide ranae of commercial Icommunitv radio 193%) I

1 TV 198%): All SABC stations. e.tv and M-Net 1
Daily I weekly newspapers (77%). magazines (63%)
Accessed Internet
1 Cinema 11 2%) and outdoor
I
1
I Full access to sewices I

Increased ownership of durables, plus motor vehicles (33%) I
SU-LSM-GROUP 8

Schooling: Matric and higher
Live Urban 195%)
Wide range of commercial Icommunity radio (91%)

TV (98%): All SABC stations, e.tv and M-Net I
Accessed Internet I
) Full access to services
1Full ownership of durables, including PC (22%) and satellite dish (10%) 1I
Appendix C: Summary of the SUlSM Classification 133

SU-LSM-GROUP 9

Schooling: Matric and higher (73%)
Live: Urban
1 Earn: R10 732 oer month 1
Wide range of commercial 1 community radio (91%)

TV (98%): SABC 2. SABC 3. M-Net and DStv I
Daily 1 weekly newspapers (84%. magazines (77%)
Accessed lntemet
1 Cinema (21%) and outdoor 169%)
4I
I Full access to services I
1 Full ownership of durables, including PC (47%) and satellite dish (21%) 1
SU-LSM-GRGOUP 10
Age: 35 - 49 years (35%)

Schooling: Matric and higher (43%)
Live: Urban (81%)
Earn: R15 931 per month
I
Wide range of commercial 1 community radio (90%)
I
TV (99%): SABC 2, SABC 3, e.tv, M-Net and DStv
Accessed Internet
Cinema (29%) and outdoor
Full access to services

Full ownership of durables, including PC (80%) and satellite dish (45%)

APPENDIX D
SAARF SU-LSM CONCLUSIONS
Only the SU-LSMgroups related to the study - SU-LSMgroups 1 to 3, and 6 to

10 -are referred to in the following concluding tables of SAARF (2004).
PROVINCE
GENDER
POPULATION
GROUP
White I 01 91 281 51 1
661 84
All values are exmegeed as Dercentaaes
Appendix D: SAARF SU-LSM Conclusions
- -
ONTHLY INCOME
I All values are expressed as percentages 1)

Note: The values referred to in the abovementioned tables are the latest
available values (20028).
Appendix D: SAARF S U l S M Conclusions 136

LSM Segmentation

Uploaded by

Copyright:

Available Formats

LSM Segmentation

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

LSM Segmentation

Uploaded by

Copyright:

Available Formats

THE LIVING STANDARDS MEASURE AS A MARKET

SEGMENTATION TOOL FOR SELECTED RETAILERS

(B.COM., HONS. B.COM)

Dissertation submitted in partial fulfillment of the requirements for the degree of

ENTREPRENEURSHIP, MARKETING AND TOURlSlul MANAGEMENT,

POTCHEFSTROOM CAMPUS. NORTH-WEST UNIVERSITY

SUPERVISOR: Prof. L.R. Jansen van Rensburg

My best friend, Heleen, for her faith in me and her continuous

Professor van Rensburg, my supervisor for his expertise, guidance,

Christine, my assistant supervisor for her friendship, encouragement and

The School of Entrepreneurship, Marketing and Tourism Management,

LIVING STANDARDS MEASURE AS A MARKET SEGMENTATION TOOL

The objectives of the study were to describe market segmentation by

A market segmentation strategy concentrates the business' marketing efforts on

Market segmentation furthermore assists in the development of an effective

Some of the most important recommendations include:

- The headquarters of both retailers should consider developing

"LIVING STANDARDS MEASURE" AS 'N

'n Marksegmenteringstrategiefokus die ondememing se bemarkingspogings op

Die LSM en MGM is slegs twee van 'n verskeidenheid segmenteringsinstrumente

Sommige van die belangrikste aanbevelings sluit in:

Die hoofkantore van albei handelaars moet dit oorweeg om

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix

LIST OF KEY TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xi

CHAPTER 1: INTRODUCTION. MOTIVATION. GOAL AND METHOD ....1

CHAPTER 2: MARKET SEGMENTATION . . . . . . . . . . . . . . . . . . . . 16

Table of Contents iii

CHAPTER 5: SUMMARY. CONCLUSIONS AND RECOMMENDATIONS 106 ..

CHAPTER 1: INTRODUCTION. MOTIVATION. GOAL AND METHOD ... I

CHAPTER 2: MARKET SEGMENTATION . . . . . . . . . . . . . . . . . . . . 16

CHAPTER 3: LIVING STANDARDS MEASURE AND MEDIA GROUPS

Table of Contents vii

CHAPTER 4: EMPIRICAL METHODOLOGY AND EXPOSITION . . . . . . 85

Table of Contents viii

CHAPTER 2: MARKET SEGMENTATION . . . . . . . . . . . . . . . . . . . . 16

CHAPTER 3: LIVING STANDARDS MEASURE AND MEDIA GROUPS

CHAPTER 4: EMPIRICAL METHODOLOGY AND EXPOSITION . . . . . . 85

CHAPTER 5: SUMMARY, CONCLUSIONS AND RECOMMENDATIONS. .I06

The study of market segmentation is an intricate one, frequently raising many

Marketers soon realised that it would be practically impossible to satisfy every

Chapter 1: Introduction, Motivation, Goal and Method 1

A market can be described as a group of customers with specific needs or

1.2.2 Target Market

A target market consists of a group of individuals (or businesses) with similar

1.2.3 Market Segmentation

1.2.4 Living Standards Measure

The Living Standards Measure (LSM) is a market segmentatin tool that

1.2.5 Media Groups Measure

The Media Groups Measure (MGM) is a tool designed to obtain media

Chapter 1: Introduction, Motivation, Goal and Method 2

The aforementioned formed a basis on which could be determined whether

Market segmentation, according to Rix (2004:93), is the process of dividing a

The SAARF Universal LSM (SU-LSM) is a market segmentation tool, assisting

Chap6er 1: Introduction, Motivation, Goal and Method 3

1.3.1 Reasons for Selecting the Topic

As mentioned, the LSM categorises people according to their living standards

Market segmentation was selected as a study field, since it is evident that it is

Chapter 1: Introduction, Motivation, Goal and Method