Empirical Relationships Between Cognitive Style and Lov: Implications For Values and Value Systems

Roger P. McIntyre, East Carolina University
Reid P. Claxton, East Carolina University
David B. Jones, La Salle University
[ to cite ]:
Roger P. McIntyre, Reid P. Claxton, and David B. Jones (1994) ,"Empirical Relationships Between Cognitive Style and Lov: Implications For Values and Value Systems", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 141-146.

Advances in Consumer Research Volume 21, 1994      Pages 141-146

EMPIRICAL RELATIONSHIPS BETWEEN COGNITIVE STYLE AND LOV: IMPLICATIONS FOR VALUES AND VALUE SYSTEMS

Roger P. McIntyre, East Carolina University

Reid P. Claxton, East Carolina University

David B. Jones, La Salle University

INTRODUCTION

Human values serve as vehicles by which culture is handed from one generation to another and as guideposts by which the paths of individual lives are charted. Values have been variously defined as enduring beliefs about preferred conduct or end states of being (Rokeach 1973); as conceptions of desirable influences on the ways people select behavior and evaluate their world (Schwartz and Bilsky 1987); and as society's goals represented within the individual (Beatty, Homer, and Kahle 1988). More narrowly, the study of values offers a richer, fuller picture of both the consumer and consumer motivation. Accordingly, values often provide the basis by which markets are segmented.

Cognitive style refers to the Jungian (1971) psychological type theory which most often classifies people according their modes of information intake on one hand, and their modes of information processing and decision making on the other. This paper contends that cognitive style is an antecedent condition that affects the formation of human values. Because cognitive style defines individual differences in the perception and processing of information and in resulting patterns of decision making, the overall inference is that cognitive style is an immediate and stable influence on human values and value systems. Empirical relationships between cognitive style and items from the List of Values (LOV) scale (Kahle 1983; Veroff, Douvan and Kulka 1981) will be offered in support.

Responding to the call for more complex measures of values that include provision for the effects of fundamental differences among individuals (Richins and Dawson 1992), this paper suggests that cognitive style is a major individual difference variable that may contribute to improved measures of values and value systems. In hierarchical terms, cognitive style is believed to affect value systems and values, which in turn affect beliefs, attitudes, and behaviors.

VALUES AND THEIR MEASUREMENT

Values are central to individual human lives and to society as a whole (Rokeach 1973). Values are cognitive representations of the biological, interpersonal, and social demands placed upon each of us as human beings (Schwartz and Bilsky 1987). People use culturally learned values to create and assess conditions of morality and competence, to guide social interplay, and to help rationalize beliefs, attitudes and behaviors (Rokeach 1973).

In response to the subjectivity of earlier value scales (e. g., Allport, Vernon and Lindzey 1960), Milton Rokeach in 1973 introduced the Rokeach Value Survey (RVS). The RVS measures 18 instrumental and 18 terminal values. Instrumental values pertain to ideal modes of behavior. Terminal values pertain to ideal end states of existence. In the RVS, all 18 items in each set are ranked according to their importance to the respondent. Difficulty arises, however, in the ranking of such a large number of values. People rank middle values less reliably than more extreme ones. RVS also ignores a number of widely-held values, and questions have arisen about the superiority of interval- and ratio scaling to the RVS method of rank ordering (Beatty, Kahle, Homer and Misra 1985).

An alternative to the RVS was subsequently developed based on the work of Feather (1975), Maslow (1954) and Rokeach (1973). The List of Values (LOV), developed by Kahle (1983) and Veroff, Douvan and Kulka (1981), modifies the terminal values of RVS into a smaller nine-item subset. LOV asks respondents to identify their two most important values from the nine-value list.

The LOV scale has exhibited validity comparable to RVS while offering greater parsimony. LOV has also been found to provide a higher percentage of items that respondents said influenced their lives, although both LOV and RVS are susceptible to social desirability bias. Finally, the LOV scale offers the advantage of greater ease of administration and completion (Beatty et al. 1985; Kahle, Beatty, and Homer 1986).

The Values and Life Style (VALS) methodology was developed at SRI International by Mitchell (1983). Based on the need hierarchy of Maslow (1954) and the social character concept of Riesman, Glazer and Denney (1950), VALS incorporates 32-36 questions pertaining to general and specific attitudes and demographics. Respondents are ultimately classified into one of nine lifestyle groups. In 1989, SRI International introduced VALS 2 to replace VALS. While the VALS system has had a dramatic impact on the business community, its proprietary nature has largely precluded scholarly and scientific attention (Kahle et al. 1986).

All three of the instruments, RVS, LOV, and VALS, show relationships to various forms of consumer behavior. RVS values were found to predict mass media usage, for example (Becker and Conner 1981). Another study found LOV to offer greater utility in predicting consumer trends than VALS (Kahle et al. 1986).

An inherent difficulty arises in the body of work just cited. Simply stated, people are almost never guided by just one value (Micken 1992). While consumers are often segmented using their top-ranked LOV value, for example, such an approach conflicts with Rokeach's concept of an overall hierarchy of individual, ordered values (Kamakura and Novak 1992). A number of researchers have supported the investigation of value systems in preference to single values. Pitts and Woodside (1983, p. 38) indicate that "single values are salient only in the context of the entire value system." Similarly, Schwartz and Bilsky (1987) argue that research should focus on value domains as opposed to individual values.

Instead of relying on directly observable value rankings, Kamakura and Novak (1992) report creating value system segments from latent discriminal values. The authors construct value system segments described as having greater abstraction than traditional single values. The greater abstraction of the value system segments purportedly reflects the multiple values affecting a person's behavior. The authors ultimately produce a segmentation map delineated by the nine LOV values which identifies the dimensions underlying the value systems of each segment. Each dimension is prioritized by the worth/utility assigned by segment members. The resulting value system segments and map are reported to have face validity consonant with the psychological structure of values advanced by Schwartz and Bilsky (1987).

Unlike segmentation based on a LOV top-ranked value, Kamakura and Novak (1992) indicate that more abstract values defined by the latent system provide "a better understanding of the motivations that drive the beliefs, attitudes, and behavior of each segment" (p. 120). They recognize that consumer values are "also affected by many other more immediate (but also less stable) environmental influences, such as price, sales promotions, exposure to advertising messages" (p. 130), and state that such influences must also be considered. It is the contention of this paper that cognitive style may be both an immediate and stable influence on human values. Cognitive style may therefore affect behavior to the extent that it may affect formation of value systems and values, which in turn affect beliefs, attitudes and behaviors.

Having been led to the study of values by a search for measures of materialism, Richins and Dawson (1992) similarly conclude that values are complex phenomena that require the use of complex measures to complement traditional ranking procedures:

It is no more reasonable to measure the value of "warm relationships with others" with a single item than it is to measure attitudes toward religion with a single question on a survey (p. 314).

Richins and Dawson recommend that, as an alternative to revising the Rokeach scales, it may be time to develop entirely different measures of values; measures that are "more suited to the usual application of individual difference variables in consumer behavior" (p. 314). One of the most fundamental individual difference variables is cognitive style. As discussed in the next section, the theory of cognitive style offers a logical and compelling starting point for developing new measures of human values.

COGNITIVE STYLE

The Jungian (1971) psychological type theory known as cognitive style is based on three bi-polar axes: extroversion/ introversion, sensing/intuiting, and thinking/feeling. The instrument developed to operationalize cognitive style, the Myers-Briggs Type Indicator (MBTI), adds a fourth axis: judging/perceiving (Myers and McCaulley 1985). The four axes produce a somewhat unwieldy 16 psychological types, thus compelling most researchers to use just two axes: sensing/ intuiting and thinking/feeling. Placed at right angles, these axes produce the quadrants that delineate cognitive style (Frisbie 1988; Mitroff 1981). An individual's cognitive style results from the combination of favored information intake mode (sensing/intuiting) with the preferred decision making mode (thinking/feeling). The four resulting cognitive styles include: sensory-thinking (ST), intuitive-thinking (NT), intuitive-feeling (NF) and sensory-feeling (SF). Table 1 presents the generalized characteristics of the four cognitive styles.

Sensing/intuiting is the information intake axis. It refers to how individuals learn about their environment. People typed by the MBTI as sensors (S) tend to learn from detailed sensory input such as data and hard facts. Sensors have been found to perceive reality as being concrete and divisible. Not surprisingly, they are pragmatic and oriented toward numbers, details, and other tangibles. In contrast, people typed as intuitors (N) perceive realities that are holistic and often beyond direct sensation. Intuitors tend to rely on conceptualization and imagination, frequently constructing realities based on possibilities and ideals. Discerners of patterns and trends, Ns have been found to be quite comfortable with intangibles (Gould 1991; Mokwa and Evans 1984).

Thinking/feeling is the axis of information processing and decision making. Thinkers (T) tend to rely on conventional deductive logic. Individuals typed as thinkers tend to prefer explanations that are formal and general, the types in which human qualities play a relatively minor role. Tending to see situations in black and white, Ts are relatively more analytical, impersonal, and objective. Classical conceptions of logic and science provide an excellent portrait of typical T traits. On the other hand, people typed as Feelers (F) place relatively more emphasis on human qualities. They often search for justification of circumstances and expression of feeling, and so operate with greater questioning and individuation. Fs have been found to be empathizing, personal, subjective, and accepting of ambiguity. The disciplines of aesthetics, ethics, religion, and politics are rich in F characteristics (Mokwa and Evans 1984).

The tendency toward a particular cognitive style appears to be innate, presumably due to the endurance of genetic influences (Johnson 1992; Myers and McCaulley 1985). The two poles that predominate a particular cognitive style heavily shape the individual's perceptions of the world. The two non-dominant poles play lesser roles in perception shaping. The cognitive styles of greatest contrast (i. e., those which are "opposites") occur in the quadrants diagonal to each other in Table 1. Two people having opposite cognitive styles may interpret the same phenomenon in markedly different ways.

Cognitive style has been shown to influence problem solving (Hellriegel and Slocum 1975), conflict management (Percival, Smitheram, and Kelly 1992), and information processing (Smith and Urban 1978). A number of cases of cognitive style-based differential behavior are reported in the areas of communicating, learning, and decision making (e. g., McCaulley and Natter 1974). With regard to buying behavior, consumers have been found to prefer interacting with salepeople who share the consumer's cognitive style (McIntyre 1991). Cognitive style may also affect other facets of consumer-salesperson interaction (Singh 1990).

PROPOSED RELATIONSHIPS BETWEEN COGNITIVE STYLE AND LOV

Preliminary theoretical support exists for a relationship between cognitive style and values. Kamakura and Novak (1992) state that value systems are important tools for decision making and conflict resolution. The additional likelihood must be considered that values and value systems occur as the result of decision making in many instances. Partly because decision making is subsumed by the thinking/feeling axis of cognitive style theory, the major inference is here advanced that cognitive style is often antecedent to the formation of human values.

It is important to remember that LOV modifies the 18 terminal values of RVS into a smaller nine-item subset and asks respondents to identify their two most important values therein. Terminal values deal with the major undertakings of life, i. e., marriage, parenthood, and other avenues of contentment and realization of self such as daily consumption, work and leisure (Kahle et al. 1986). Therefore, terminal values are perhaps the most subjective of all values in a person's life. The relatively subjective nature of terminal values may make them especially sensitive to polarized responses, depending on a respondent's type from the thinking/feeling axis of cognitive style. Said another way, terminal values refer to many personal and interpersonal states. Reactions to such complex and socially oriented areas may well be expected to vary depending on whether an individual's preferred way of processing information and making decisions is more affective (feeling) or cognitive (thinking) in nature. For these reasons, the eight hypotheses which follow are framed in terms of the thinking/feeling axis of cognitive style theory. Accordingly:

H1: Subjects typed as Feelers (F) will have a significantly greater response to "Warm relationships with others" than those typed as Thinkers (T).

H2: Subjects typed as Feelers (F) will have a significantly greater response to "A sense of belonging" than those typed as Thinkers (T).

TABLE 1

CHARACTERISTICS OF THE FOUR COGNITIVE STYLES

Because few LOV respondents have traditionally selected "excitement" as their top value choice, the category is usually collapsed into "fun and enjoyment" (Kahle and Kennedy 1988). Based on the perceptions that fun, enjoyment, excitement and security are predominantly affective states, the third hypothesis emerges:

H3: Subjects typed as Feelers (F) will have a significantly greater response to "Fun and enjoyment in life/excitement" than those typed as Thinkers (T).

For the individual, the condition of living is ephemeral. Rationally, all life has the same, "non-secure" eventual outcome. As a result, the value of security is similarly perceived as an affective state:

H4: Subjects typed as Feelers (F) will have a significantly greater response to "Security" than those typed as Thinkers (T).

Many areas of life, however, offer relatively greater opportunity for objective (cognitive) evaluation (e. g., successfully mowing a lawn, having a baby, earning a promotion). Theoretically, individuals who prefer to process information and make decisions from a cognitive orientation (Ts) may have a greater preference for values that are more readily demonstrable. Propositon five reflects the many social indicators of accomplishment:

H5: Subjects typed as Thinkers (T) will have a significantly greater response to "A sense of accomplishment" than those typed as Feelers (F).

In social terms, a corollary to Hypothesis 5 may be:

H6: Subjects typed as Thinkers (T) will have a significantly greater response to "Being well-respected" than those typed as Feelers (F).

While contemporary living offers many value-states that may be judged relatively more affective or cognitive in nature, some values strike at the absolute core of human life. Such values, it would seem, may tend to be desirable regardless of one's preferred style of information processing and decision making. Accordingly:

H7: Subjects typed as Feelers (F) will not have a significantly greater response to "Self-fulfillment" than those typed as Thinkers (T).

H8: Subjects typed as Feelers (F) will not have a significantly greater response to "Self respect" than those typed as Thinkers (T).

TABLE 2

COMPARISONS OF LOV DIMENSIONS AND COGNITIVE STYLES DOMINATED BY THINKING (T) VERSUS FEELING (F) (N=218)

RESULTS

Methodology

Sample. Because interest focused more on theory testing than application, a student sample proved appropriate (Calder, Phillips, and Tybout 1981). Responses were analyzed from 218 business students at a southeastern university. Some 110 females and 92 males participated. Average age was 22.6 years (range: 20 to 47).

Instruments. The LOV scale and the sensing/intuiting and thinking/feeling portions of the MBTI were administered to the sample. Following the suggestions of Kahle and Kennedy (1988), the excitement value of LOV was collapsed into the fun and enjoyment value, and responses were measured by a 7-point, Likert-type scale where 1 was "Not Important" and 7 was "Very Important." Regarding the MBTI, its validity and reliability as an indicator of cognitive style have been extensively critiqued (e. g., Druckman and Bjork 1991; Johnson 1992).

Results

The goal was to test each of the hypotheses with regard to the thinking/feeling dimension of the MBTI. Accordingly, t-tests were conducted on the LOV means for each hypothesis. To maintain an experiment-wide alpha of 0.05, a Bonferroni-adjusted alpha of 0.006 was necessary for each of the eight comparisons (Hayes 1988, p. 411).

As summarized in Table 2, five of the eight predictions were fully supported and one was partially supported. As predicted, Feelers valued the LOV categories of "warm relationships with others," "sense of belonging," and "security" more than Thinkers. Also as predicted, no significant differences in the valuation of "self-fulfillment" and "self-respect" were found between Feelers and Thinkers. Similarly, Thinkers were found to value "sense of accomplishment" more than Feelers, although this finding did not meet the required alpha of 0.006. Not supported were the predictions that Feelers would value the collapsed category of "fun and enjoyment"/"excitement" more than Thinkers, and that Thinkers would value "being well-respected" more than Feelers.

IMPLICATIONS, CAVEATS, AND FUTURE RESEARCH

Empirical evidence has been presented of a relationship between the thinking/feeling axis of cognitive style theory, as measured by the MBTI, and the values measured by LOV. Four differences in value preference were predicted and either supported or partially supported between subjects typed as either Thinkers or Feelers. In addition, two values were shown not to be statistically different in their valuation by Ts and Fs, as predicted. Two predictions of differential emphasis on values between Ts and Fs were not supported.

Values are culturally learned (Kamakura and Novak 1992; Rokeach 1973). On the other hand, as previously noted, the tendency toward a particular cognitive style appears to be innate. Given the empirical relationships demonstrated by this paper, a resulting implication is that the presumed genetic influence of cognitive style merits careful consideration as a potential antecedent to the formation of values and value systems in many instances. More specifically, the individual's preferred mode of decision making (thinking or feeling) may not only be antecedent to value formation, it may also closely influence the manner of adoption of values and their ultimate prioritization within value systems. Thus, the introduction of cognitive style into research on values and value systems may offer an additional way of building upon the impressive body of work that already exists in this area.

A further implication is the possibility that MBTI may subsume LOV. The thinking/feeling axis of MBTI measures an indi-vidual's preferred information processing and decision making mode. If the innate (and therefore antecedent) preference for a decision making mode does influence the adoption and priori-tization of values, MBTI may indeed subsume LOV. Deductively, if an influential antecedent to values (i. e., preferred decision making mode) is measured by an instrument (the MBTI), then the possibility exists that the instrument which measures values' antecedents may subsume other instruments which measure values themselves (LOV). Clearly, a considerable amount of additional research would be necessary to support this possibility. It should also be noted that in conducting a 30-month test-retest reliability study on the MBTI, Johnson (1992) found that most changes in cognitive style designation occur on the thinking/ feeling axis. Although 77% of respondents had retained their original T-F preference upon retest, the author indicates that "the problems with stability of type lie chiefly with the T-F preference" (p. 57).

Lastly, it should also be noted that responding to items on the LOV scale is clearly a cognitive activity, the result of processing the content of each item and deciding on an appropriate response to it. Therefore, this paper's hypotheses were confined to potential influences on LOV of the information processing and decision making component of cognitive style (thinking/feeling). Future investigation will need to determine any effects the information intake component (sensing/intuiting) may have on responses to LOV, the adoption values, and their prioritization within value systems.

REFERENCES

Allport, G., P. Vernon, and G. Lindzey (1960), Study of Values, Boston: Houghton Mifflin Co.

Beatty, Sharon E., Lynn R. Kahle, Pamela Homer and Shekhar Misra (1985), "Alternative Measurement Approaches to Consumer Values: The List of Values and the Rokeach Value Survey," Psychology & Marketing, Vol 2 (No. 3, Fall), 181-200.

Beatty, Sharon E., Pamela M. Homer and Lynn R. Kahle (1988), "Problems With VALS in International Marketing Research: An Example From An Application of the Empirical Mirror Technique," in Advances in Consumer Research, 15, Michael J. Houston, ed., Provo, UT: Assn. for Consumer Research, 375-380.

Becker, B. W. and P. E. Conner (1981), "Personal Values of the Heavy User of Mass Media," J. of Adv. Rsrch (21) 37-43.

Calder, Bobby, Lynn W. Phillips, and Alice Tybout (1981), "Designing Research for Application," Journal of Consumer Research, 8 (Sept.), 197-207.

Druckman, D., and R. A. Bjork (1991), In the Mind's Eye: Enhancing Human Performance (Chapter 5), Committee on Techniques for the Enhancement of Human Performance, National Research Council, Washington, D. C.: National Academy Press.

Feather, Norman T. (1975), Values in Education and Society, New York: Free Press.

Frisbie, George R. (1988), "Cognitive Styles: An Alternative to Keirsey's Temperaments," J. of Psychological Type, 16, 13-21.

Gould, Stephen J. (1991), "Jungian Analysis and Psychological Types: An Interpretive Approach to Consumer Choice Behavior," in Advances in Consumer Research, 28, Rebecca H. Holman and Michael R. Solomon, eds., Provo, UT: ACR, 743-8.

Hayes, William L. (1988), Statistics, Orlando, FL: Holt, Rinehart and Winston, 411.

Hellriegel, Don and John W. Slocum, Jr., (1975), "Managerial Problem-Solving Styles," Business Horizons, (Dec.), 29-37.

Johnson, D. A. (1992), "Test-Retest Reliabilities of the Myers-Briggs Type Indicator and the Type Differentiation Indicator Over a 30-Month Period," J. of Psychological Type, 24, 54-58.

Jung, Carl G. (1971), Psychological Types, (H. G. Baynes, trans., rev. R. F. C. Hull), Princeton NJ: Princeton U. Press.

Kahle, Lynn R., ed., (1983), Social Values and Social Change: Adaptation to Life in America, New York: Praeger.

Jung, Carl G., and Particia Kennedy (1988), "Using the List of Values (LOV) to Understand Consumers," Journal of Services Marketing, Vol. 2 (Fall), 49-56.

Jung, Carl G., Sharon E. Beatty and Pamela Homer (1986), "Alternative Measurement Approaches to Consumer Values: The List of Values (LOV) and Values and Life Style (VALS)," Journal of Consumer Research, Vol. 13 (December), 405-409.

Kamakura, Wagner A. and Thomas P. Novak (1992), "Value-System Segmentation: Exploring the Meaning of LOV, Journal of Consumer Research, Vol. 19 (June), 119-132.

Maslow, A. H. (1954), Motivation and Personality, NY: Harper.

McCaulley, Mary H., Gerald P. Macdaid, and Richard I. Kainz (1985), "Estimated Frequencies of the MBTI Types," Journal of Psychological Type, 9, 3-9.

McCaulley, Mary H. and F. L. Natter (1974), "Psychological (Myers-Briggs) Type Differences in Education," in The Governor's Task Force on Disruptive Youth: Phase II Report, Tallahassee, FL: Office of the Governor.

McIntyre, Roger P. (1991), "The Impact of Jungian Cognitive Style on Marketing Interactions," unpublished doctoral dissertation, Arizona State University.

Micken, Kathleen S. (1992), "Materialism Research: Suggestions for New Directions," Proceedings of the Research Workshop on Materialism and Other Comsumption Orientations, eds. Floyd Rudmin and Marsha Richins, ACR, Kingston, Ont., pp. 110-112.

Mitchell, A. (1983), The Nine American Life Styles, NY: Warner.

Mitroff, I. I. (1981), "Is a Periodic Table of the Elements for Org. Behavior Possible?", Human Systs. Mgmt, 2, 3, 168-76.

Mokwa, Michael P. and Kenneth R. Evans (1984), "Knowledge and Marketing: Exploring the Foundations of Inquiry," in Marketing Theory: Distinguished Contributions, Stephen W. Brown and Raymond P. Fisk, eds., New York: Wiley, 170-9.

Myers, Isabel Briggs and Mary H. McCaulley (1985), Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator, Palo Alto, CA: Consulting Psychologists Press.

Novak, Thomas P. and Bruce MacEvoy (1990), "On Comparing Alternative Segmentation Schemes: The List of Values (LOV) and Values and Life Styles (VALS), Journal of Consumer Research, Vol. 17 (June), 105-109.

Percival, Terrance Q., Verner Smitheram, and Margaret Kelly (1992), "Myers-Briggs Type Indicator and Conflict-Handling Intention: An Interactive Approach," Journal of Psychological Type, 23, 10-16.

Pitts, Robert E. and Arch G. Woodside (1983), "Personal Value Influences on Consumer Product Class and Brand Preferences," Journal of Social Psychology, 119, 37-53.

Reisman, David, Nathan Glazer, and Revel Denney (1950), The Lonely Crowd, New Haven, CT: Yale University Press.

Richins, Marsha L. and Scott Dawson (1992), "A Consumer Values Orientation for Materialism and Its Measurement: Scale Development and Validation," Journal of Consumer Research, Vol. 19 (Dec.), 303-317.

Rokeach, M. (1973), The Nature of Human Values, NY: Free Press.

Schwartz, Shalom H. and Wolfgang Bilsky (1987), "Toward a Universal Psychological Structure of Human Values," Journal of Personality and Social Psychology, Vol. 53 (3), 550-562.

Singh, Jagdip (1990), "A Typology of Consumer Dissatisfaction Response Styles," Journal of Retailing, Vol. 66 (1), 86-87.

Smith, August W. and Thomas F. Urban (1978), "Myers-Briggs Personality Orientations and Information Processing Styles: Implications for Management Education and Development," paper presented at the National Academy of Management, MED Session, San Francisco, August 10.

Synder, M. (1976), "The Self-Monitoring of Expressive Behavior," J. of Personality & Social Psychology, 30 (Oct.), 526-37.

Veroff, Joseph, Elizabeth Douvan, and Richard A. Kulka (1981), The Inner American, New York: Basic Books.

----------------------------------------