INTRODUCTION

The search for knowledge on the forces that determine consumer attitudes and behaviour has led marketing researchers to investigate the causal factors that may, in one way or another, contribute towards the explanation of these phenomena. Of particular importance among the prominent areas of this line of research is the study of human values and their possible relation with the formation of opinion or attitude. The present study seeks to identify possible relationships between preponderant values in certain groups and the way in which they perceive the positioning of brands, products and utilities.

Rokeach1 (p. 132) suggests that ‘values are related to modes of conduct and the reason for being (…) and once internalized becomes, consciously or unconsciously, a criterion in guiding action, for developing and maintaining attitudes in relation to objects and relevant situations’. Values are, thus, references that people use to judge themselves and others, or even influence the values, attitudes and actions of other people, such as their children, for example. Although there are a number of definitions for the term values, practically all of them include some common characteristics: values (a) are concepts or beliefs, (b) concern behaviour or desirable ends, (c) transcend specific situations, (d) guide the choice or assessment of behaviour and events and (e) follow an order of importance.2

Attitude, like values, is based on belief, though not a unique or transcendental belief; it is, in truth, the result of an articulated set of beliefs concerning a certain object or situation, which predispose the individual to respond in one way or another.3 An attitude regarding a certain object is relatively long-lasting and persistent, formed by past experiences; for this reason it should not be confused with either momentary predisposition or behaviour. Regarding the behaviour of a person in relation to an object or situation, Rokeach1 (p. 102) suggests that it ‘will depend on both the beliefs and particular predispositions activated by the attitude, and the beliefs and predispositions activated by the situation’.

The capacity of values to influence the assessment and choices, personal or of others, suggests there is a relationship between values, attitude and behaviour,4 and this makes the topic extremely interesting for academics in marketing. According to the Means-End Chain Theory,5 values represent the desirable end-states requiring some means to be fulfilled. Therefore products and their brands may work as part of the means set used to reach such final state. The connection between values and brand is evident in Howard6 who suggests that terminal values are essential for product class decision and instrumental values determine the brand choice.

THEORETICAL FOUNDATION

The effect of variables like age, income and geography is important in the understanding of consumer behaviour and in developing marketing plans, though the analysis of individual differences, like personality, values and lifestyle can offer a better understanding of such behaviour.7

When using the term values, people probably will be trying to express something similar to the conception of a desire that influences the way in which the actions and events that involve them are selected and assessed. Rokeach8 refers to values as beliefs about life and the acceptable behaviour. Marketing professionals can focus on individual or group values. According to Blackwell et al.7 (p. 223), ‘when the importance of the values set is so widely accepted that it becomes almost a stereotype of a market segment or group, we refer to it as social values’. The present study is focused on personal values, which, according to the same authors, define the ‘normal’ behaviour of individuals, influence the way they live, what they consider right and wrong, what they buy and what they consider important for themselves, such as pleasure, honesty and ambition. Personal values, according to these same authors, assume a fundamental role in the consumer's purchase decision process, as they help us answer the question: ‘is this product of use to me?’ (p. 226).

Besides values, when we assume the role of consumers, each one of us has different attitudes in relation to products, services, advertisements, brands or ideas. Whenever they ask us whether or not we like a certain product or service, they are asking us to express our attitudes.9 Therefore, attitude is the expression of the most intimate feelings that reflect if a person is favourably or unfavourably inclined towards a certain object (eg a brand, a service, an idea, etc). Accordingly, it is important that firms understand what the consumers know and do not know regarding a product, brand, store, etc, in the same way as it is important to know what consumers like and do not like.

Behaviour, according to Rokeach,8 is formed of beliefs, attitudes and values that function as an integrative cognitive system, and if there is a change in any part of the system, the other parts are also affected, provoking changes in behaviour. According to this same author1 (p. 102), ‘ the behaviour of a person in relation to an object or situation, will depend as much on particular beliefs and predispositions activated by attitudes, as on beliefs and predispositions activated by the situation’. Regarding the situation, Blackwell et al.7 add that the simple fact of liking a certain product, brand or model, does not mean that the consumer will not consider alternatives, as it may be that he/she may not be able to pay, or may not need the preferred object. The consumer may, for example, desire a big car, but despite this chooses a small one due to the price or space available.

The concept of positioning, although implicit in the studies of market segmentation and selection since the 1960/1970s, was popularized with the work of Ries and Truot, in the 1980s,10 but their study was restricted to the intrinsic perspective of ‘manipulation of the mind’ of the target public.11 Aaker and Shansby12 made progress in detailing the various possible strategies for creating a favourable image in the mind of the consumer, while limiting themselves to recommending the use ‘of the various existing measurement techniques’ in order to assess the positioning and generate information on future positioning, without offering details on how to do it. Only recently, Blankson and Kalafatis13 sought to define a generic typology and validate a scale aimed at measuring the effectiveness of the positioning, while keeping in mind that until that moment (p. 31) ‘the effective operationalisation of the positioning had been hampered by the absence of generic consumer/customer-based empirically derived instrument to measure the perception of the consumers’. Mazanec14 addressed a methodology for the identification of a competitive structure, highlighting that the market structure might be better inferred from so-called ‘choice maps’, rather than with the use of ‘perceptual maps’ based on explicit attributes of the brands or on preference data.

The present study aims to identify possible relationships between values and the perceived positioning, using a satisfier — the Manager — as brand proxy. The pursued objectives are twofold: (a) empirically verify possible influences of respondents personal values in the choice of their future profession, and (b) test and disseminate the ‘new methodology for analyzing competitive position’— NMACP proposed by Mazanec,14, 15 in a context other than Europe. Even the business concept (values and positioning) is the main objective authors acknowledge that the technique of analysis is the most appealing contribution to be offered to marketing scientists and practitioners due to its simplicity and facility of application in the business world. To interpret marketing positioning is a rather difficult task and some sophisticated methods (ie mosaic, metaphor analysis) require lots of interpretation abilities and their results are complex to be statically validated. It is not new that interpretative methods are art and technique at the same time, and therefore different meanings can be easily attributed to the same results according to the researcher's perception. Consequently, NMACP demonstrate at least three characteristics that makes it attractive to interpret market positioning: it is relatively easy and safe to detect the ‘best’ number of cluster to be formed from data; it is not difficult to interpret consumers values and opinions; and it is possible to test significances of any variable among clusters.

METHOD

Using an instrument previously validated in the Brazilian context, the study sought to reveal the values predominating among students from several degree courses at a large Brazilian university, with the aim of verifying the existence of a possible correlation between such values and the chosen profession and, fundamentally, between those values and the perception of positioning. Rokeach's scale was chosen because it was the instrument of choice of numerous studies16 and has been also appropriated to commercial purpose.17, 18 Gastaldello17 used Rokeach's scale to assess the influence of human values in commercial negotiations among executives from Brazil and Argentina, and Marmitt18 employed the same instrument to assess the behaviour of electrical appliances consumers from two different Brazilian cities. The Rokeach Value Survey (RVS)8, 19 is a compound designed to measure two sets of values: one set is composed by 18 terminal values or desired end states of existence (eg, an exciting life, national security, happiness, etc), and the other set is composed of 18 instrumental values, or preferable modes of behaviour (eg, been ambitious, independent, clean, etc). In total, 1,609 enroled students took part from 38 different degree courses, aged between 18 and 30 years, where 50.9 per cent were women and 49.1 per cent men.

Before initiating the complete field study, it was necessary to perform an exploratory procedure to define the object whose positioning would be assessed, as well as to define the variables of interest. For this purpose the focus group technique was used, though instead of using a tangible product, it was suggested that the positioning of Manager be assessed in the context of the other professions, bearing in mind that the role played by that professional is reflected in the work of the others. Sixty-eight variables of opinion were defined and added to the 36 values of the original scale and the five control variables, which resulted in a questionnaire of 109 items. The original questionnaire is available from the authors.

For the analysis, the data were divided into two sets of variables: the first composed of the 36 values in the Rokeach Scale (previously validated), and the second containing the 68 variables of opinion, designed for the interpretation of the perceived positioning. As a first approach towards the analysis that would be carried out through neural networks, as well as to reduce the number of variables, the scale designed to assess the positioning of the Manager as a satisfying agent of the other professionals was refined with the use of exploratory factorial analysis as recommended by Hair et al.20 Five factors were identified through which the positioning is assessed: the capacities and skills of the Manager; his/her pro-social and ethical conduct; his/her knowledge and preparation for the function; his/her level of information and up-datedness; attitude in relation to authority. All the factors showed Cronbach alphas over 0.7 and a significance level <0.001 between them. The reduction in the number of items was necessary both to reduce the computational effort required and to facilitate the interpretation of the construct. The resulting scale (36 items) is available from the authors.

ANALYSIS AND RESULTS

Both the definition of the clusters extracted from each set of data and the clustering process itself were achieved using TRN-32 software, which implements the neural gas algorithm.14 The TRN model introduced under the name of Topology Representing Network by Martinetz and Schulten, in 1994,14 employs the competitive learning principle in which the prototypes rival one another in attempting to approximate the frequency distribution of empirical data; however, unlike other networks, ‘the training rule adjusts not only the winning prototype but all prototypes according to the rank of distances between data point and the first winner, second winner, etc’ (Mazanec,21p. 49).

This method of analysis, adapted by Mazanec,14 consisted of three distinct phases. In the first phase, the optimum number of clusters to be formed from the two sets of data was decided; in the second, the networks were trained to form homogenous segments, allocate the weight of the variables to the prototypes and identify the clusters of each one of the respondents; in the last phase clusters were typified according to the incident variable loads and the similarities identified between the preponderant values and opinions in each of the clusters.

The decision regarding the number of clusters was based both on statistical criteria and on interest, since the techniques for deciding the ‘correct’ number of clusters generally have their limitations.22 The statistical criterion used was the percentage of uncertainty reduction (%UR), based on the repeated quantization of the data (30 rounds), recommended as a good reference for decision.23 As a result of this process four clusters were suggested for the values and three for the positioning, as shown in Table 1.

Table 1 Statistics of the different structural alternatives
Full size table

Nevertheless, if positioning data were restricted to only three clusters, 98 per cent of the respondents would be concentrated into just two of them, which would greatly reduce the explanatory power. Thus, the option to include five clusters seemed reasonable, both in terms of the more uniform distribution of the respondents (table omitted) and for the gain in density (mean/standard deviation) obtained (0.38). Of the 36 values in the Rokeach Scale, only 14 (four terminal and 10 instrumental) had discriminant loads and for this reason were treated as distinct values. The remainder loaded (or not) indistinctly in all the clusters. The characterization of the respondents from the different groups, according to the predominant values, can be found in Table 2 (L).

Table 2 Characterization of the respondents according to the preponderant values and the ways in which they classified the Manager
Full size table

Of the 36 variables of opinion designed to interpret the positioning of the role of Manager in the mind of the respondents, only 22 are distinctly loaded in the five clusters. The characterization of the respondents of the different clusters, according to the predominant opinions about the role of Manager, can be found in Table 2 (R). Significance among clusters has been confirmed with MANOVA for all values and positioning variables, with Pillai's Trace, Wilks’ Lambda, Hotelling's Trace and Roy's Largest Root tests20 (Table 3).

Table 3 Multivariate tests of variables among clusters
Full size table

DISCUSSION

The comparison of the predominances among values and opinions made it possible to characterize at least ten areas of confluence with more than a hundred individuals in each, which corroborates the suitability of the method proposed by Mazanec14, 15 for the assessment of the positioning. There are, for example, 195 people who simultaneously belong to clusters V1 and P, that is, they are people who value a fair and pleasing, clean and stable world, while at the same time view the Manager in quite a positive light. There are, however, 128 people belonging to clusters V4 (of a independent, bold profile, though solidary), who view the Manager in quite a critical light, as bureaucrat who performs routine functions without a decision-making capacity (Cluster 2). It is not possible to describe other identified combinations due to lack of space, but they can be easily viewed in Figure 1 (Panel A).

Figure 1
figure 1

Clusters of perceived values and positioning

Full size image

Analysis of labelled data and its spatial mapping suggest the influence of values in the decision regarding the choice of profession (eg chemistry in cluster V1) (Figure 1, Panel B) and in the different perceptions of positioning that the members of the segments of the population may have of the same product. The finding that the perception the consumers have in relation to the positioning of products or brands can be related to their personal values is innovative and of great value in the area of knowledge in marketing. Furthermore, these findings are limited as the brand has been surrogated by the Manager figure, a satisfier of a professional's necessities but not of a brand itself. It seems to be very interesting to replicate this study using a well-known brand to confirm the results. Also, there is no doubt that this new method of ‘reading’ positioning, using artificial neural networks in place of conventional mapping, opens an interesting research front both in the field of marketing strategy and in the area of consumer behaviour.

The authors strongly encourage the performance of new confirmatory studies using different methods and samples, in order to check both the evidence that the values are important antecedents of perceived positioning and the greater suitability shown by neural networks, in gauging the positioning of products or brands, following the implementation of a strategy.