ABSTRACT
The lifestyle research instruments developed and used by most of the larger market research firms have been criticized by academic marketing scholars mostly on five grounds (e.g., Anderson & Golden, 1984; Askegaard, 1993; Banning, 1987; Lastovicka, 1982; Roos, 1986).
There is no agreement on what lifestyle actually means. The term seems to defy definitional consensus. Anderson and Golden (1984), after perusing a large number of published lifestyle studies, conclude that in most cases the term is not defined at all, and when it is defined, the definitions range from the contradictory to the trivial.
The methods used are purely inductive and not guided by theory. Lifestyle types come about based on dimensions derived by exploratory data analysis techniques like factor analysis or correspondence analysis. These techniques are applied to sets of items, the generation of which is not theoretically guided either, but is very much based on common sense reasoning and implicit experience in carrying out market research. Although such a research procedure may be appropriate in the early phase of the life cycle of a research technique, one should hope that, based on such exploratory analysis, theory should develop, which could then guide the analysis of new and better measurement instruments. Also, many feel that consumer behavior is such a well-researched area that it should be possible to obtain some theoretical input from there that could enrich lifestyle research.
The derivation of the underlying dimensions is unclear and unsatisfactory. Because commercially marketed instruments, like VALS, RISC, or CCA, are usually proprietary, information necessary to evaluate the statistical soundness of the derived dimensional solutions is often missing. A priori, many social science researchers tend to be suspicious when several hundred variables are reduced to just two dimensions (which is the case in many instruments). On the other hand, if a large part of the variance in the data can be explained by just two dimensions, it should be possible to capture these dimensions with a much simpler instrument, which would considerably reduce the cost involved in data collection.
The explanatory value of lifestyle types or dimensions with regard to consumer choice behavior is low and not well documented. Evidence supplied has been mostly in the form of cross-tabulations of lifestyle items or types with self-reported use of or attitude towards certain products (see Wells, 1975). Also, users of a certain brand or product can be placed on the lifestyle maps based on their mean factor scores. When it has been attempted to relate purchase data and lifestyle data in such a way that the amount of variance in the former explained by the latter can be ascertained, the amount of variance explained has been very modest, sometimes even below the variance explained by demographic variables alone (Bruno & Pessemier, 1971; Wells & Tigert, 1971). As Wells put it in a review article in 1973: “Stated as correlation coefficients these relationships appear shockingly small—frequently in the .1 or .2 range, seldom higher than .3 or .4” (Wells, 1973). Newer studies do not show improvements in this respect (Aurifeille & Valette-Florence, 1992; Valette-Florence, 1989, 1991). It seems that lifestyle items are especially poor when it comes to explaining consumer behavior at brand level, while explanatory power at product category level may be a little higher (Hustad & Pessemier, 1974).
The cross-cultural validity of the international lifestyle instruments remains to be demonstrated. The larger pan-European lifestyle studies like RISC and CCA provide data that aim at identifying similar lifestyle segments across borders, and numerous other lifestyle studies have tried to identify cultural differences in lifestyle (e.g., Douglas & Urban, 1977; Hui, Joy, & Laroche, 1990; Laroche, McTavish, Johnson, Joy, Kim, & Rankine, 1990; Linton & Broadbent, 1975). Collecting data in different cultures with the aim of obtaining comparative results requires that the measurement instrument has cross-cultural validity, that is, that translation and measurement equivalence are ensured or at least tested (cf. Chandran & Wiley, 1987; Green & White, 1976; Sekaran, 1983).
