The use of personal values in living standards measures L.M. Ungerer & J.P.R. Joubert 6A B S T R A C T 14The Living Standards Measure (LSM), a South African marketing segmentation method, is a multivariate wealth measure based on standard of living. This article reports on whether a rationale can be found for the inclusion of psychological variables, particularly personal values, in this type of multivariate segmentation. Schwartz’s (1992; 2006) values model was used to operationalise personal values. 15Data were collected by means of a survey from a nationally representative sample (N = 2 566) of purchase decision-makers. The Portrait Values Questionnaire (PVQ) (Schwartz, Melech, Lehmann, Burgess, Harris & Owens 2001) was used to measure personal values. Centred value scores were used to statistically control for variation among respondents in rating themselves on the PVQ. The 10 LSM groups were collapsed into four LSM super groups to provide a more rigorous analysis of measured personal value diff erences between LSM groups. The statistical analyses included descriptive and inferential statistics. 16The fi ndings in general supported Schwartz’s theory of basic human values. Respective values could be allocated among diff erent wealth- based consumer segments, which explained variances between these groups. It would be advisable to further investigate how these diff erences can be applied in marketing. Although the LSM segmentation approach has been adapted somewhat, the principal fi nding of this research remains applicable. 17Key words: personal values, market segmentation, multivariate segmentation, Portrait Values Questionnaire, Living Standards Measure, purchase decision-makers Dr L.M. Ungerer is in the Department of Industrial and Organisational Psychology, University of South Africa; Prof. J.P.R. Joubert in the Bureau of Market Research, University of South Africa. E-mail: L.M. Ungerer & J.P.R. Joubert Introduction 1In the 53 years since the pioneering article by Wendell Smith (1956), market segmentation can be considered one of the most dominant concepts in marketing literature and practice. Market segmentation research involves obtaining behavioural information that is useful in guiding managerial choices for strategic action. The underlying logic is that customers differ in terms of their product preferences and buying behaviour, and groups of customers with similar characteristics will tend to respond similarly to marketing programmes (Dibb & Simkin 2008). According to Nwanko, Aiyeku and Ogbuehi (2006), constantly changing modern marketing environments, fierce competition in many consumer goods markets and the need to target consumers more effectively force marketers to continually review methods of segmenting markets in which they are interested. Although socio-economic segmentation, for example, provides a broad picture of a particular market, Mead (in Nwanko et al. 2006) points out that socio-economic categories are increasingly being replaced by other more sophisticated segmenting techniques, which probe the psychological aspects of consumer decision-making. Conceptualisation 1In this section, market segmentation, the role of values in market segmentation, Schwartz’s values theory, and the South African Advertising Research Foundation’s Universal Living Standards Measure (LSM) are discussed. Market segmentation 1Higgs (2008) points out that South Africa is ideally suited to market research because of its cultural diversity and its high Gini co-efficient (a measure of the degree of inequality in people’s incomes). A marketplace as diverse as in South Africa, composed of many different people, with different backgrounds, interests, needs and wants, especially lends itself to market segmentation – the practice of dividing a market into smaller specific segments sharing similar characteristics (Tranter, Stuart-Hill & Parker 2009: 42). Market segmentation is usually the first step in a three-phase marketing strategy. After segmenting the market into homogeneous clusters, one or more segments are selected for targeting. To accomplish this, the marketer must decide on a specific product, price, channel and/or promotional appeal for each distinct segment. Finally, the product is positioned in such a way that it is perceived by each target market as satisfying that market’s needs better than competitors’ products. Segmentation 98 SABVIEW15_2_finale proofs.indd 98 2011/08/26 08:13:31 AM The use of personal values in living standards measures studies can also guide the redesign or repositioning of a product, where the product targeting is extended to a new consumer segment. Furthermore, segmentation research may assist marketers in identifying the most appropriate media in which to place advertisements (Schiffman & Kanuk 2007). Selecting the most appropriate base (or bases) on which to segment the market forms the first step in developing a segmentation strategy. Michman, Mazze and Greco (2003) point out that segmentation should not be limited to a single variable; instead multiple dimensions should ideally be considered. A segmentation scheme based on multiple dimensions is often more valuable and flexible in planning marketing strategy than a single base. A combination of several segmentation bases is consequently used in hybrid (or multivariate) segmentation to create comprehensive profiles of particular consumer segments. The role of values in consumer behaviour 1Kluckhohn (in Berry, Poortinga, Segall & Dasen 2002: 59) defines a value as a conception, explicit or implicit, distinctive of an individual or characteristic of a group, of what is desirable, which influences the selection from available modes, means and ends of action. Rose and Shoham (2000) regard values as a learned set of desired outcomes and beliefs that guide attitudes and behaviour, while Rokeach (in Rose & Shoham 2000) considers values as enduring beliefs that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state. According to Schwartz (2006), values are concepts or beliefs about desirable end-states or behaviours that transcend specific situations, guide selection or evaluation of behaviour and events and are ordered by relative importance. Five important features of values, identified by Grunert and Scherhorn (in Christiansen & Hansen 2001), are evident from the foregoing. Values: • are beliefs or concepts, • about desirable behaviour or end-states, • go beyond particular situations, • guide the choice or evaluation of events and behaviour, and • are arranged in a hierarchy. According to Schwartz (1992), scholars in the areas of psychology, anthropology and sociology all agree on two core principles. First, values serve as guiding principles in peoples’ daily lives because they are enduring beliefs about desirable end-states. Second, although values are utilised at the personal level, values are socially 99 SABVIEW15_2_finale proofs.indd 99 2011/08/26 08:13:31 AM L.M. Ungerer & J.P.R. Joubert constructed and inherently cultural. People are thus not born with values; they learn values through socialisation. According to Solomon, Bamossy, Askegaard and Hogg (2006), a person’s set of values plays a very important role in his or her consumption activities, because people tend to purchase many products and services, believing that they will help them attain a value-related goal. A value system is a learned organisation of principles and rules to help one choose between alternatives, resolve conflicts and make decisions (Rokeach, in De Mooij 2004: 77). The extent to which people share a value system is a function of individual, social and cultural forces, as was evident in the preceding discussion. Values serve as guiding principles in consumers’ everyday life (Kahle, Rose & Shoham 2000). Because of differences in culture and socio-economic conditions, certain types of values may be regarded as more important to consumers in one country than another and may affect their attitudes and purchase decisions more significantly. It is therefore important to understand consumer behaviour in a particular cultural setting, especially consumers’ social values. This would be indicative of the macro-approach to understanding the role of values in consumer behaviour, as will be evident in the next section. Consumers’ values serve as a focal point in many cognitive tasks, such as attitude formation or decision-making about a brand purchase. They further serve to organise the meaning representations for products and brands stored in consumers’ knowledge structures (Peter & Olson 2008). According to Cleaver and Muller (2002), marketers are increasingly acknowledging the role that human values play in consumers’ decisions. Durgee, O’Connor and Veryzer (in Lages & Fernandes 2005) also point out that one of the most powerful ways to understand and reach consumers is to understand their values and value systems. Reflecting firmly established values in advertising and product offerings is an essential ingredient in achieving brand awareness, consumers’ trial of products and their subsequent loyalty to a product. The application of the values perspective to the marketing of consumer products is based on two theoretically grounded perspectives, namely a macro-perspective representing sociology and a micro-perspective representing psychology (Reynolds, in Joubert & Mabunda 2007). In the macro-approach underlying this research, standard survey research methodology is combined with a classification scheme to categorise respondents into predetermined clusters or groups. In this way, a classification system or taxonomy is developed, which can be used to segment consumers into groups based on their value orientations. The psychological perspective offered by the micro-approach, based on Means- End Theory, specifically focuses on the links between the attributes that exist in 100 SABVIEW15_2_finale proofs.indd 100 2011/08/26 08:13:31 AM The use of personal values in living standards measures products (the means), the consequences for the consumer provided by the attributes, and the personal values (the ends) that the consequences reinforce (Veludo-de- Oliveira, Ikeda & Campomar 2006). Since human values serve as guiding principles in the life of a person or social entity, information regarding the human values that are important to the target market would be valuable in the development of marketing strategies. Related research should therefore focus on values, on what they mean, how they evolve and how they are manifested in products as perceived by target consumers (Allen 2001). Schwartz’s values theory 1Several theoretical perspectives have been used to understand how values develop and how they impact on behaviour. This research was based mainly on Schwartz’s (1992; 2006) values theory. Schwartz (2006) presents a theory of the basic values that people in all cultures recognise. It identifies ten motivationally distinct value orientations and specifies the dynamics of conflict and congruence among these values. Schwartz and Bilsky (in Schwartz 2006) furthermore propose a universal structure that is particularly useful in the macro-approach to segmentation underlying this research. They identify different groups of values according to the following three criteria: • according to the objective (instrumental or terminal), • according to interest (individualist, collectivist or both), and • according to ten motivational domains (power, achievement, hedonism, stimulation, self-direction, universalism, benevolence, conformity, tradition and security). Values can thus be considered as people’s adherence to objectives (terminal or instrumental), which are conducive to satisfying certain interests (individual, collective or both). These values belong to ten motivational areas, which are of varying importance in a person’s daily life. These lead to motivationally distinct value types that are likely to be recognised within and across cultures and are used to form value priorities. In addition, Schwartz (2006) conceptualises the complete pattern of relations of conflict and congruity among values in a circular structure, as evident in Figure 1. The closer any two values are in either direction around the circle, the more their underlying motivations correspond. The more distant values are, the more opposite their underlying motivations are. Some pairs of values therefore compete, while 101 SABVIEW15_2_finale proofs.indd 101 2011/08/26 08:13:31 AM L.M. Ungerer & J.P.R. Joubert others are complementary. Benevolence and power, for instance, are contradictory, while others, such as conformity and security, are compatible. Tradition and conformity are located in the same wedge because they share the same broad motivational goal. Conformity lies more towards the centre and tradition more towards the periphery. This signifies that tradition values conflict more strongly with the values opposing them. The expectations linked to tradition values are more abstract and absolute than the interaction-based expectations of conformity values. Tradition values therefore demand a stronger, explicit rejection of opposing values. OPENNESS SELF- TO TRANCENDENCE CHANGE Self-direction Universalism Stimulation Benevolence Hedonism Conformity Tradition Achievement Power Security SELF- ENHANCEMENT CONSERVATION Source: Schwartz (1994: 24) Figure 1: Theoretical model of relations among motivational types of values, higher order value types and bipolar value dimensions Two basic dimensions organise value systems into an integrated motivational structure with consistent value conflicts and compatibilities (Schwartz 2006). As Figure 1 shows, one dimension refers to ‘openness to change’ versus ‘conservation’ values. This dimension captures the conflict of an emphasis on one’s own independent thought and action and favouring change (self-direction and stimulation values), versus submissive self-restriction, preservation of traditional practices and protection of stability (security, conformity and tradition). The second dimension deals with 102 SABVIEW15_2_finale proofs.indd 102 2011/08/26 08:13:31 AM The use of personal values in living standards measures ‘self-enhancement’ versus ‘self-transcendence’ values. This dimension captures the conflict of an emphasis on acceptance of others as equal and concern for their welfare (universalism and benevolence) versus pursuing one’s own relative success and dominance over others (power and achievement). Hedonism contains elements of both openness to change and self-enhancement. The Schwartz (2006) theory of the basic values consequently distinguishes ten values that form a continuum of related motivations. This continuum explains the circular structure where adjacent values share motivational emphases. Power and achievement, for instance, share the emphases of social superiority and esteem. Regarding values as being organised in a circular motivational structure has an important implication for their relations to other variables – it implies that each value in the whole set of ten relates to any other variable in an integrated manner. According to Schwartz (2006), associations of value priorities with other variables, such as attitudes, behaviour and demographics, tend to reflect this structure. If any variable associates most positively with one value type (for instance, age with conformity), its associations with the other tends to be progressively less positive as one moves around the circle, in either direction, towards the diametrically opposing type. Support for the distinctiveness of the ten values and the prototypical circular structure of relations among them, as presented in Figure 1, was found in 95% of the samples in research from 63 nations in which the SVS was used. Schwartz et al. (2001) point out that it is evident that, although people differ substantially in the importance they attribute to values, most literate adults’ values across cultures are organised according to the common structure of motivational oppositions and congruities, as proposed by Schwartz’s theory. South African Advertising Research Foundation’s Universal Living Standards Measure (LSM) 1According to Haupt (2006), a requirement for an effective market segmentation tool is that it should create an index that will differentiate between people with different behaviour patterns and group together those people with similar behaviour. One such index, the Living Standards Measure (LSM), developed by the South African Advertising Research Foundation (SAARF), is one of the most widely used marketing segmentation tools in South Africa. The target population for the purpose of the SAARF surveys is defined as the adult population (16 years and older) of South Africa, who reside in any type of private household, including live-in domestic workers, hostel dwellers and residents of informal settlements. 103 SABVIEW15_2_finale proofs.indd 103 2011/08/26 08:13:31 AM L.M. Ungerer & J.P.R. Joubert The SAARF Universal LSM measure is calculated by summing individual scores on 29 variables. These variables reflect access to services and durables, and geographic indicators as determinants of standard of living. The weighted scores are used to divide the South African population into ten LSM groups. The LSM groupings are currently based on quantifiable, descriptive variables – particularly consumers’ demographic characteristics and the assets they own. Haupt (2006) points out that the LSM can consequently not be used as a psychographic or attitudinal measure. Although Haupt (2006) does not refer to psychological variables as such, it appears that LSM-based segmentation can be enhanced by incorporating these types of variables. Psychological characteristics refer to the inner or intrinsic qualities of the individual consumer (Schiffman & Kanuk 2007). According to Yeshin (2006), identifying consumers’ underlying psychological characteristics provides a much richer texture from which to discriminate sufficiently among them. If certain psychological variables are found to distinguish among consumer segments, this information can be applied to further augment this type of segmentation. It may also be useful to position products more effectively in advertising to multivariate segments, which, in turn, would lead to more effective allocation of funds. In summary, the LSM is a wealth measure, based on standard of living. This research investigates the inclusion of values as psychological variables in this type of multivariate segmentation. Research aim 1The research problem investigated in this research was whether consumer values as measured by Schwartz’s (1992) values model can be used as additional psychological segmentation variables in the widely accepted multivariate LSM segmentation approach used in South Africa, and whether certain value priorities are more prevalent within broad groupings identified in this type of segmentation. The primary aim of this research was therefore to investigate the extent to which values, as measured by Schwartz’s (1992) model, differ across major groupings as defined by the LSM segmentation approach. Therefore, the general aim was to investigate whether values can be used to further discriminate among multivariate consumer segments. 104 SABVIEW15_2_finale proofs.indd 104 2011/08/26 08:13:31 AM The use of personal values in living standards measures Research method 1In light of the fact that creating a distinct value profile for each of the LSM groups is a fairly complex task involving large data tables, and that some of the LSM groups share common features, it was considered appropriate to reduce the number of LSM groups in this research before creating a distinct value profile for them. The collapsing of ordinal and higher level taxonomies into larger groups is standard practice in analytical empirical research in order to enhance the statistical interpretability of the research data. In deciding which groups to combine for further analysis, general market practice was taken into consideration, and the following super groups were created: LSM super group 1 (purchase decision-makers from LSM 1–3), LSM super group 2 (purchase decision-makers from LSM 4–5), LSM super group 3 (purchase decision-makers from LSM 6–7) and LSM super group 4 (purchase decision-makers from LSM 8–10). Four LSM super groups were therefore created to investigate differences in values among them. Figure 2 presents the LSM super groups and their corresponding percentages as represented in the sample. 150 145 140 135 133 130 126 124 1% 125 120 16 15 10 1 5 1 0 1LSM1-3 1LSM4-5 1LSM6-7 1LSM8-10 Figure 2: LSM super groups (N = 2 566) This research can be classified as non-experimental exploratory research because no attempt was made at manipulating any variables – certain phenomena (consumers’ values) were only measured at a particular time. This research can be further 105 SABVIEW15_2_finale proofs.indd 105 2011/08/26 08:13:31 AM L.M. Ungerer & J.P.R. Joubert classified as quantitative research, because a structured questionnaire was used to collect numeric data, which were analysed by means of statistical procedures. The specific quantitative research method that was used entailed the use of structured questionnaires that were completed during personal interviews in a national survey, as will become evident. Participants 1The data were collected by a reputable marketing research company based in Johannesburg. The national (both metro and non-metro) surveys were conducted in October 2005. It is argued that the contribution or central tenet espoused in this research is not bound by a specific year or time period. The value of this research is grounded primarily in the enhancement of multivariate segmentation approaches by including psycho-social variables such as values as descriptors of particular wealth groups. The population for this survey consisted of all South African adults. A stratified probability sample of 3 500 adults was selected for the survey. The primary stratifica- tion variable was geographical region, using the nine provinces as a basis, and the second stratification variable was community size, with cities, large towns, small towns, villages and rural areas as categories. The metro sample was a stratified probability sample consisting of 2 000 adults (16 years and older) living in metropolitan areas. The non-metro sample also consisted of a stratified probability sample of 1 500 adults (16 years and older), but living in non-metropolitan areas. This sampling process resulted in 3 500 face-to-face, in-house interviews being undertaken with respondents 16 years and older in both metro and non-metro areas. In order to determine the appropriate respondents for this research, the questionnaire included a screening question requiring respondents to indicate whether they were the main purchasers for a particular household. This exclusion criterion resulted in a final sample of 2 566 respondents for the research. The final sample consequently consisted of adult South African purchase decision-makers. The questionnaire was translated into isiZulu, isiXhosa, Setswana, Sesotho, Sepedi and Afrikaans. Interviews were conducted in respondents’ preferred language. All interviewers had a Senior Certificate qualification and were trained and thoroughly briefed before they conducted any interviews. The interviewers were representative of South African society. All black interviewers were required to be proficient in four languages, while white, coloured and Indian interviewers were required to have command of at least two languages. Although all interviews were conducted under the constant supervision of trained and experienced supervisors, a 106 SABVIEW15_2_finale proofs.indd 106 2011/08/26 08:13:31 AM