Printer Friendly

Cross-cultural generalizability of a scale for profiling consumers' decision-making styles.

Profiling consumers' decision-making styles has been the focus of a multitude of consumer interest studies (e.g., Bettman 1979; Sproles 1985; Thorelli, Becker, and Engeldow 1975; Westbrook and Black 1985). Consumer affairs specialists use such profiles to understand consumers' shopping behavior, while advertisers and marketing researchers use them to segment the consumers into various niches for product positioning. Until this point, most of the empirical research investigating consumer styles used U.S. samples for developing and validating the measuring instrument. However, such research may be inapplicable to other cultures, unless cross-cultural psychometric properties of the measures (i.e., dimensionality and reliability) are shown to exist (Douglas and Craig 1983; Hui and Triandis 1985). Further, if the psychometric properties of the measures of consumer decision-making styles vary widely across countries, conclusions based on the scale may actually be attributed to measure unreliability (Green and White 1976; Parameswaran and Yaprak 1987). As a result, evidence of the generalizability of consumer styles research and related instruments to other cultures is needed.

Therefore, this study examines the cross-cultural applicability of a scale for profiling consumers' decision-making styles. The study is consistent with the stream of research that addresses the cross-cultural generalizability of consumer behavior measurement scales and procedures (see Netemeyer, Durvasula, and Lichtenstein 1991).

CONSUMERS' DECISION-MAKING STYLES

Decision-making style refers to a mental orientation describing how a consumer makes choices. Extant research in this field has identified three approaches to characterize consumer styles: (a) The Consumer Typology Approach (Darden and Ashton 1974; Moschis 1976); (b) The Psychographics/Lifestyles Approach (Lastovicka 1982; Wells 1974); and (c) The Consumer Characteristics Approach (Sproles 1985; Sproles and Kendall 1986; Sproles and Sproles 1990). Integral to all these approaches is the theme that despite an element of individuality in consumers' behavior, all consumers approach shopping with certain basic decision-making styles such as rational shopping, impulsiveness, and quality consciousness.

The Consumer Characteristics Approach, however, is one of the most promising as it deals with the mental orientation of consumers in making decisions and, therefore, focuses on the cognitive and affective orientations in consumer decisionmaking. It is valuable to consumer affairs specialists because it provides a measurement system for standardized testing of consumer decision-making styles for practical applications such as counseling consumers.

The genesis of this approach was based on an exploratory study by Sproles (1985) that identified 50 items related to this mental orientation. Sproles and Kendall (1986) reworked this inventory and developed a more parsimonious scale with 40 items under the title, Consumer Style Inventory (CSI). Note that many of the original 50 items are not directly comparable to the CSI; hence reference to Sproles' (1985) findings will not be made. In the CSI, factor analysis identified eight mental characteristics of consumer decisionmaking: (1) Perfectionism or high-quality consciousness; (2) Brand consciousness; (3) Novelty-fashion consciousness; (4) Recreational, hedonistic shopping consciousness; (5) Price and "value for money" shopping consciousness; (6) Impulsiveness; (7) Confusion from over choice of brands, stores, and consumer information; and (8) Habitual, brand-loyal orientation toward consumption.

Because the reliability and validity of the CSI were established using a sample of U.S. high school students, these authors recommended validating the instrument across other populations. Therefore, the intent of this study is to validate the CSI inventory proposed by Sproles and Kendall (1986) by examining its generalizability to another country. The study responds to the criticism that models and empirical findings developed with U.S. data may not be valid in other countries, and further research is required to demonstrate their applicability (Albaum and Peterson 1984; Hui and Triandis 1985; Lee and Green 1991). Specifically, this study examines psychometric properties of the CSI and compares the findings to previous research. Though highly exploratory in nature, the study follows the recommended methodology for testing the cross-national applicability of measures (Berry 1980; Irvine and Carroll 1980; Hui and Triandis 1985). Parenthetically, the study reported here is unlike the one by Hafstrom, Chae, and Chung (1992) which uses the studies by Sproles (1985) and Sproles and Kendall (1986) to make comparisons between decision-making styles of U.S. young consumers and those in Korea; they did not examine the psychometric properties of the measures.

METHOD

Data to examine the CSI inventory were obtained from 210 undergraduate business students at a large university in New Zealand. The sample had a mean age of 20.2 years and was evenly divided by sex. The students were from diverse backgrounds ranging from urban to rural. The instrument contained 40 Likert-scaled items scored from 1 (strongly disagree) to 5 (strongly agree), and the analysis employed statistical procedures identical to those used by Sproles and Kendall (1986). Unlike the home economics high school students used by Sproles and Kendall (1986) (81 percent of whom were female), the sample of undergraduate students used in this study permits a more rigorous test of the applicability of the scale. Similarity in findings between the two samples would support the robustness of the inventory. Using a relatively more homogeneous group such as undergraduates also minimizes random error that might occur by using a heterogeneous sample such as the general public (Calder, Tybout, and Phillips 1981).

The analysis examined the psychometric properties of the CSI. First, the dimensionality of the consumer styles inventory was TABULAR DATA OMITTED assessed. This was done by examining the factor solution. Specifically, the amount of variance explained by the extracted factors (i.e., their eigenvalues) was noted. In addition, item-factor correlations (i.e., factor loadings) and other indices of model adequacy were examined. To obtain the factor solution, a principal components factor analysis was used with a varimax rotation. The purpose of factor analyzing the 40-item inventory was to determine if the factors identified by Sproles and Kendall (1986) were common to the New Zealand sample. Second, the authors computed Cronbach alpha coefficients to assess the scale reliabilities of the factors identified and to make comparisons with Sproles and Kendall's (1986) findings. In cross-cultural research, such an approach is commonly the first step in determining the generalizability of a model or scale to another culture (Irvine and Carroll 1980).

RESULTS

To compare New Zealand sample results with the U.S. sample results, an eight-factor solution was obtained for the New Zealand sample. Table 1 features factor loadings of the 40-item inventory for the New Zealand sample and those obtained for the U.S. sample by Sproles and Kendall (1986). Although the results for the New Zealand sample are not entirely equivalent to the U.S. sample, the similarities outweigh the differences. As shown in Table 1, the eight-factor model appears adequate as it explained 56 percent of the variation for the New Zealand sample. This result compares favorably with the 46 percent reported for the U.S. group. Further, all eight factors have eigenvalues greater than one, which is a rule often used to judge model adequacy. Of the eight factors, three (factors 4, 7, and 8) had substantially the same pattern of factor loadings for the two samples. For these factors, the sign and magnitude of the factor loadings were found to be similar for both samples, indicating that the factors had equal valence. An examination of the loading pattern of all 40 items reveals that the magnitude of 32 out of 40 loadings (80 percent) is similar across both samples.(1)

Although there were some differences in the pattern of certain item loadings, these were not major. In the New Zealand sample, four of the 40 items loaded on factors other than those found for the U.S. sample. Items 5 and 7, for example, showed higher loadings (0.70 and 0.42, respectively) on factor 6 labeled "impulsive and careless consumer." Intuitively, these items ("don't give purchases much thought" and "I shop quickly") seem to represent impulsiveness (factor 6) rather than high quality consciousness (factor 1) which is confirmed by the results. Similarly, items 10 ("more expensive brands are my choice") and 26 ("lower price products are my choice") had higher loadings on factor 1 than on the factors identified by Sproles and Kendall (1986). For New Zealand, these items seem to represent price cues to quality and, therefore, are appropriate measures of factor 1 (i.e., high quality consciousness). Finally, items 9, 12, 18, and 19 exhibit relatively low loadings (|is less than~ 0.4; the same criterion was used for the U.S. sample), indicating that they are relatively poor measures of the corresponding factors. Among these, item 18 had an equally high loading (0.33) on factor 6. This represents factorial complexity and suggests that this item is not uniquely associated with any one factor. In contrast, item 8, which cross-loaded on both factors 1 and 2 in the U.S. sample, exhibited a high loading only on factor 1 for the New Zealand sample. Such items are considered not to tap any single construct and, therefore, could be deleted in further scale purification processes (Gerbing and Anderson 1988).

Table 2 shows the internal consistency (i.e., Cronbach alpha) estimates of scale reliability. For comparison purposes, the same items used to compute reliability of individual subscales in the U.S. sample were also used to compute reliability estimates for the New Zealand sample. The alpha estimates are generally similar for both samples. Given that an alpha of .70 or better is desired for any measurement scale (Nunnally 1978), the scales representing "Perfectionistic," "Novelty-Fashion Conscious," and "Recreational Shopping Conscious" factors are stable and internally consistent in the two samples. Support for internal consistency exists for measures of the "Impulsive" factor only for the New Zealand sample. While the measure of "Confused by Over-Choice" approached the .70 cutoff for the New Zealand sample, its value of .55 for the U.S. data is far below this. The "Brand Conscious" scale exhibited a high coefficient alpha only in the U.S. sample, suggesting that these two scales may be affected by cultural differences. Hence, both these factors and the "Confused by Over-Choice" factor require further improvement. The scales representing "Habitual, Brand-Loyal Consumer" and "Price-Value Conscious Consumer" require further refinement as they lack acceptable levels of reliability for both U.S. and New Zealand samples.
TABLE 2
Reliability Coefficients for Eight Consumer Style
Characteristics
 Cronbach Alpha for Subscales
 U.S. New Zealand
Consumer Style Characteristics Sample Sample
1. Perfectionistic .74 .75
2. Brand Conscious .75 .59
3. Novelty-Fashion Conscious .74 .70
4. Recreational Shopping Conscious .76 .82
5. Price-Value Conscious .48 .50
6. Impulsive .48 .71
7. Confused by Over-choice .55 .66
8. Habitual, Brand-Loyal .53 .58


DISCUSSION

This study reflects the concern about the generalizability of scaling instruments to other cultures or countries. Researchers have suggested that most measures developed in consumer behavior are validated using U.S. samples and, therefore, might apply only to the United States (Green and White 1976). Perhaps for this reason, researchers have started to investigate cross-cultural dimensions of consumer behavior constructs (e.g., Andrews, Lysonski, and Durvasula 1991; Lysonski and Pollay 1990; Netemeyer, Durvasula, and Lichtenstein 1991). As the global marketplace becomes more integrated and consumer specialists develop an international focus, developing useful scales to profile consumer decision-making styles in other cultures becomes important. Hence investigation of these scales is needed.

The CSI was chosen for investigation because it can be a useful technique to alert consumers to their mental orientation toward shopping. Being informed may help consumers become more effective shoppers. Besides this approach, there appears to be none specifically designed to serve consumer interest professionals. Sproles and Kendall (1986) recommended that a person who takes this inventory be given a Profile of Consumer Style (PCS), a format that would report an individual's shopping characteristics. There are many uses for this instrument in New Zealand. For example, as the use of credit cards becomes more prevalent in New Zealand, some consumers are miring themselves in serious debt by their lack of prudence in using such credit. Use of the CSI could raise their consciousness about their approach to shopping and need for credit purchases. Obviously, if the CSI is to be used in other cultures or countries, establishing its applicability or generalizability is essential.

The retail environment in New Zealand is in sharp contrast to the United States. For example, stores close at 5:30 p.m. except for one night each week when they are open until 9:00 p.m. Stores are also closed on Sundays and Saturday afternoons. With only 3.3 million people, competition among retailers is not as intense nor are there as many competitors as one finds in the U.S. market. Hence, consumers have less choice. Furthermore, discretionary and disposable incomes are lower compared to the United States. Brand consciousness may be at a different state of development compared to the United States. In fact, a recent study by Andrews, Lysonski, and Durvasula (1991) reported that there were greater brand (i.e., user) related thoughts about advertising in general generated by subjects in the United States versus other countries, including New Zealand. Despite these structural differences, the decision-making styles are expected to have universal applicability. Much like personality traits, decision-making styles are expected to be largely independent of the culture and descriptive of a personal orientation (Sproles and Kendall 1986).

This exploratory study examines the usefulness of the consumer style inventory (CSI), developed and applied in the United States, to New Zealand--a culture located 8,000 miles away. The sample of New Zealand subjects differs from the U.S. sample for two reasons: (1) the New Zealand sample was comprised of college students with an average age of 20 years versus high school students used in the U.S. sample, and (2) the New Zealand sample was more balanced in terms of male/female representation as opposed to the U.S. sample, 80 percent of which was female. The differences in the samples provide for a stronger test of cross-cultural generalizability of the inventory.

Overall the New Zealand results compare favorably to those of the United States and provide general support for this inventory. However, not all the results were equivalent. For example, some items displayed a different pattern of loadings compared to the U.S. findings with four of the items loading on different factors. Next, two of the scales ("Impulsive" and "Brand Conscious") appear to be culture-specific as they exhibit an acceptable reliability level for one sample or the other, but not for both. Further, two other scales ("Price-Value Conscious" and "Habitual, Brand-Loyal") require refinement, no matter where they are applied. Insight into the refinement of the "Price-Value Conscious" factor can be derived from the research by Sproles (1985), where five additional items with suitable loadings were considered to measure this factor. Though a few discrepancies in results do exist between the U.S. and New Zealand samples, it is likely that sample differences coupled with the different retail environment in New Zealand might account for the variation in the findings.

In sum, searching for a Rosetta Stone that can explain consumers' decision-making styles that span population groups and cultures may be possible, but only with additional effort. A caveat, therefore, is warranted: consumer affairs specialists should not assume that instruments validated in the United States are immediately applicable to other countries. Perhaps, a more parsimonious version of the inventory with fewer scale dimensions that exhibits greater internal consistency could be developed and validated via confirmatory factor analysis. Researchers are encouraged to develop a more robust decision-making style inventory to account for the variation in findings as reported in this study.

REFERENCES

Albaum, Gerald and Robert A. Peterson (1984), "Empirical Research in International Marketing 1976-1982," Journal of International Business Studies, 15 (Spring/Summer): 161-173.

Andrews J. Craig, Steven Lysonski, and Srinivas Durvasula (1991), "Understanding Cross-Cultural Student Perceptions of Advertising in General: Implications for Advertising Educators and Practitioners," Journal of Advertising, 20(June): 15-28.

Berry, John W. (1980), "Introduction to Methodology," in The Handbook of Cross-Cultural Psychology, Vol. 2, Harry C. Triandis and John W. Berry (eds.), Boston, MA: Allyn and Bacon, Inc.: 1-29.

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley Publishing Company.

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

Carmines, Edward and John McIver (1981), "Analyzing Models with Unobservable Variables: Analysis of Covariance Structures," in Social Measurement: Current Issues, G. Bohrnstedt and E. Borgatta (eds.), Beverly Hills, CA: Sage Publications: 65-115.

Darden, William R. and Dub Ashton (1974), "Psychographic Profiles of Patronage Preference Groups," Journal of Retailing, 50(Winter): 99-112.

Douglas, Susan and C. Samuel Craig (1983), International Marketing Research, Englewood Cliffs, NJ: Prentice Hall, Inc.

Gerbing, David W. and James C. Anderson (1988), "An Updated Paradigm for Scale Development Incorporating Unidimensionality and Its Assessment," Journal of Marketing Research, 25(May): 186-192.

Green, Robert T. and Phillip D. White (1976), "Methodological Considerations in Cross-National Consumer Research," Journal of International Business Studies, 7(Fall/Winter): 81-87.

Hafstrom, Jeanne L., Jung Sook Chae, and Young Sook Chung (1992), "Consumer Decision-Making Styles: Comparison Between United States and Korean Young Consumers," The Journal of Consumer Affairs, 26(1): 114-122.

Hui, C. Harry and Harry C. Triandis (1985), "Measurement in Cross-Cultural Psychology: A Review and Comparison of Strategies," Journal of Cross-Cultural Psychology, 16(June): 131-152.

Irvine, Sid H. and William K. Carroll (1980), "Testing and Assessment Across Cultures: Issues in Methodology and Theory," in The Handbook of Cross-Cultural Psychology, Vol. 2, Harry C. Triandis and John W. Berry (eds.), Boston, MA: Allyn and Bacon, Inc.: 127-180.

Joreskog, Karl and Dog Sorbom (1984), LISREL: Analysis of Linear Structural Relationships by the Method of Maximum Likelihood, Version VI, Chicago: National Education Resources.

Lastovicka, John L. (1982), "On the Validation of Lifestyle Traits: A Review and Illustration," Journal of Marketing Research, 19(February): 126-138.

Lee, Chol and Robert T. Green (1991), "Cross-Cultural Examination of the Fishbein Behavioral Intentions Model," Journal of International Business Studies, 22: 289-305.

Lysonski, Steven and Richard W. Pollay (1990), "Advertising Sexism is Forgiven, But Not Forgotten: Historical and Cross-Cultural Differences in Criticism and Purchase Boycott Intentions," International Journal of Advertising, 9: 319-331.

Moschis, George P. (1976), "Shopping Orientations and Consumer Uses of Information," Journal of Retailing, 52(Summer): 61-70, 93.

Netemeyer, Richard G., Srinivas Durvasula, and Donald R. Lichtenstein (1991), "A Cross-National Assessment of the Reliability and Validity of the CETSCALE," Journal of Marketing Research, 28(August): 320-327.

Nunnally, Jum (1978), Psychometric Theory, New York: McGraw-Hill.

Parameswaran, Ravi and Attila Yaprak (1987), "A Cross-National Comparison of

Consumer Research Measures," Journal of International Business Studies, 18(Spring):35-49.

Sproles, Elizabeth Kendall and George B. Sproles (1990), "Consumer Decision-Making Styles as a Function of Individual Learning Styles," The Journal of Consumer Affairs, 24(Summer): 134-147.

Sproles, George B. (1985), "From Perfectionism to Fadism: Measuring Consumers' Decision-Making Styles," Proceedings, American Council on Consumer Interests, Karen P. Schnittgrund (ed.), Columbia, MO: ACCI: 79-85.

Sproles, George B. and Elizabeth L. Kendall (1986), "A Methodology for Profiling Consumers' Decision-Making Styles," The Journal of Consumer Affairs, 20(Winter): 267-279.

Thorelli, Hans B., Helmut Becker, and Jack Engeldow (1975), The Information Seekers: An International Study of Consumer Information and Advertising Image, Cambridge, MA: Ballinger.

Wells, William D. (editor) (1974), Life Style and Psychographics, Chicago: American Marketing Association.

Westbrook, Robert A. and William C. Black (1985), "A Motivation-Based Shopper Typology," The Journal of Retailing, 61(Spring): 78-103.

1 A confirmatory factor analysis, via LISREL VI (Joreskog and Sorbom 1984) was performed to assess the fit of the eight-factor model as proposed by Sproles and Kendall (1986). Results showed that the goodness-of-fit index (GFI) for the model was .71. (A GFI of one indicates perfect fit between the eight-factor model and the data, whereas zero indicates total lack of fit.) Further, the root mean squared residual was relatively small at 0.13, when considering there were still 740 degrees of freedom left after estimating all the parameters. Finally, a chi-squared to degrees-of-freedom ratio of three, two, or less has been advocated as an acceptable level of fit for confirmatory models (Carmines and McIver 1981). For the New Zealand sample, this ratio was 2.49, again indicating an acceptable level of fit for the model. These results support the findings of Sproles and Kendall (1986).

Srinivas Durvasula, Steven Lysonski, and J. Craig Andrews are Associate Professors, Marketing, Marquette University, Milwaukee, WI.

@ @
COPYRIGHT 1993 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Durvasula, Srinivas; Lysonski, Steven; Andrews, J. Craig
Publication:Journal of Consumer Affairs
Date:Jun 22, 1993
Words:3378
Previous Article:Cross-national comparison of consumer attitudes toward consumerism in four developing countries.
Next Article:Probabilities of small claims judgment satisfaction and factors influencing success.
Topics:

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters