Australian Versus American Consumer Decision Making Strategies: a Comparative Analysis

ABSTRACT - In an exploratory study, consumer decision making strategies are compared between American and Australian subjects on a recall task for high and low involvement products. Results indicate significant differences between which decision making strategies American consumers believe they use and strategies Australian consumers believe they use for 10 of the 20 products tested. Although both American and Australian subjects report using non-compensatory decision strategies most frequently for most products tested, Australians report using Affect Referral more frequently for High Involvement products compared to Americans. A 2 x 2 x 6 factorial design is analysed using MANOVA and reveals significant main effects for all 3 independent variables, as well as significant interactions for all variables except country and involvement. These results suggest the generalizability of consumer involvement in the decision making process cross-culturally. However, specific decision strategies used by American consumers significantly differ from those used by Australians for both high and low involvement products. Possible explanations for the results are examined from a cultural perspective.

Citation:

Paula M. Tidwell and William Marks (1994) ,"Australian Versus American Consumer Decision Making Strategies: a Comparative Analysis", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 148-152.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 148-152

AUSTRALIAN VERSUS AMERICAN CONSUMER DECISION MAKING STRATEGIES: A COMPARATIVE ANALYSIS

Paula M. Tidwell, Charles Sturt University-Mitchell

William Marks, Memphis State University

[Author's note: The American data used in this paper was analyzed in a different way and presented at a small conference in Australia, This paper has not been previously published.]

ABSTRACT -

In an exploratory study, consumer decision making strategies are compared between American and Australian subjects on a recall task for high and low involvement products. Results indicate significant differences between which decision making strategies American consumers believe they use and strategies Australian consumers believe they use for 10 of the 20 products tested. Although both American and Australian subjects report using non-compensatory decision strategies most frequently for most products tested, Australians report using Affect Referral more frequently for High Involvement products compared to Americans. A 2 x 2 x 6 factorial design is analysed using MANOVA and reveals significant main effects for all 3 independent variables, as well as significant interactions for all variables except country and involvement. These results suggest the generalizability of consumer involvement in the decision making process cross-culturally. However, specific decision strategies used by American consumers significantly differ from those used by Australians for both high and low involvement products. Possible explanations for the results are examined from a cultural perspective.

INTRODUCTION

An American ethnocentric viewpoint toward consumers and their decision making strategies would predict that all consumers throughout the world choose to use decision strategies in the same way as American consumers. Although American textbooks are used to teach consumer behaviour in Australia, it has become increasingly obvious that Australian consumer decision making does not follow the ideas presented in those books. This study compares consumer decision strategies Americans believe they use to Australian beliefs.

This study examines consumer beliefs about what choice strategies are used with products in two cultural contexts. What strategies consumers actually use has been thoroughly investigated (Engel, Blackwell and Miniard, 1993; Howard, 1989; Bettman, 1979) and is therefore not included in this study. While in some contexts a large discrepancy exists between beliefs and actual behaviour (Nesbitt and Wilson, 1977; Nesbitt and Ross, 1980), with consumer choice and strategies this discrepancy has been proven to be negligible (Weitz and Wright, 1979; Wright and Rip, 1980, Wright and Kriewall, 1980). While Nesbitt and Ross (1980) report that self-reports of unconscious and automated processes are highly inaccurate, consumer choice strategies are neither unconscious nor automated for the majority of product purchases are can therefore be examined with retrospective self-reports with a high degree of accuracy (Weitz and Wright, 1979). Therefore, retrospective self-reports are used in this study to collect data on consumers' use of choice strategies when purchasing a variety of products.

Furthermore, marketers are equally concerned about what consumers believe to be the case and what actually is the case. Marketing decisions rely heavily on consumer perceptions and not just empirical data about consumer behaviour. What is important for marketers to understand is what the consumer believes to be the case because future purchase behaviour relies heavily on those beliefs, and is often measured using strength of attitude scores in multi-attribute models, based solely on retrospective self-reports (Fishbien, 1963; Fishbien and Ajzen, 1975). Marketers understand that consumer attitudes may or may not reflect the true state of the world, but they are real in that they direct and influence consumer behaviour, which ultimately determine product purchases. This study reveals which choice strategies American and Australian consumers believe they are using when purchasing products.

The Two Cultures

Although literature comparing the American and Australian cultures is scarce, the few studies and articles that exist provide sufficient background information for this exploratory study. Several articles provide descriptions of the Australian culture, but with very little empirical evidence to support their claims. Although many of the authors qualify their statements as opinions, this seems to be a common problem in studying culture, as expressed by the few academic papers available in this body of literature.

Although the American and Australian cultures have at times been positioned as very similar for political and military rhetoric, the two cultures have more differences than similarities (Phillips, 1987). The apparently common language of English is used and expressed in different ways, indicating differences between the two cultures (Wierzbicka, 1986; Phillips, 1987). In addition, managerial styles have been examined and shown to have more differences than similarities (Dewhirst, 1989).

Although there are many differences that could be discussed, there are only two major differences that in our opinion have any relevance to this study. One is the populism of the Australian culture, that is not as prevalent in the American culture. Populism is defined as a culture's intrinsic value in familiar practices and institutions (Patience, 1991). The other is the concept of time, with associated features of modernism, technology, impersonalism and violence (Patience, 1991; Lattas, 1992). Although these issues are addressed by different authors using different perspectives, these two issues seem to be diametrically opposed to one another in the two cultures (Dewhirst, 1989; Patience, 1991; Lattas, 1992). A continuum with varying degrees of populism and modernism seems to exist, with Australian culture embracing the height of populism and American culture embracing modernism (Patience, 1991; Lattas, 1992).

Sydney is seen as a duplication of American metropolitan culture, and therefore has the same degree of modernism and the same conceptualisation of time (Lattas, 1992). However, Australians do not worry as much about whether things will be accomplished in a certain amount of time, compared to Americans who worry all the time, about time (Dewhirst, 1989; Lattas, 1992).

The laid-back attitude of the Australians accompanied by a strong work-ethic confuses many Americans who believe that if you are not worried you must be lazy (Dewhirst, 1989). Their "no worries" attitude is symptomatic of an underlying confidence that all things will be accomplished in due course (Dewhirst, 1989). The Australian laid-back attitude is found in shopping behaviour and is very different compared to the American culture, where shopping is seen predominantly as a complex decision-making task that must be mastered. Shopping in general is more important in the American culture compared to the Australian culture (Polonsky and Jarratt, 1991).

Consumer Choice Strategies

A few of the main types of choice strategies will be briefly described here , however, for a complete review of choice strategies, see Bettman (1979). There are two broad categories of choice strategies, under which the majority of subtypes fall - compensatory and non-compensatory.

Non-compensatory strategies are characterised by the fact that a weakness in one attribute cannot be offset by a strength in another attribute (Bettman, 1979). Consumers are able to process decisions in two different ways. One is by brand, called Processing By Brand (PBB) another is by attribute, called Processing By Attribute (PBA). In PBB information is required one brand at a time. PBA, on the other hand, involves the acquisition of information about a particular attribute of the various brands being considered. There are three main types of non-compensatory choice strategies used by consumers: Lexicographic, Sequential Elimination, and Conjunctive. Lexicographic involves PBA since consumers compare brands on the most important attribute, one attribute at a time. The one brand that is perceived as being superior on that one attribute is then selected. Similar to the lexicographic process, with Sequential Elimination, consumers compare brands on the most important attribute, but then impose cut-offs. If the brand meets the cut-offs for all attributes, then it is chosen. Failure on any one attribute leads to rejection of that brand. Conjunctive choice strategies involve PBB. Cut-offs are first established by the consumer for each attribute. Each brand is then compared, one at a time, against the set of imposed cut-offs. If the brand meets the cut-offs for all attributes, then it is chosen. Failure on any one attribute leads to rejection of that brand.

Compensatory strategies are characterised by the fact that a weakness in one attribute can be offset by another attribute (Bettman, 1979). There are two main types of compensatory choice strategies - simple additive and weighted additive. With simple additive the consumer adds the number of times each alternative is favourably judged in terms of the evaluative criteria, and the alternative having the most positive attributes is chosen. On the other hand, weighted additive is characterised by consumers taking into account the salience of relevant evaluative criteria in addition to whether an alternative is favourable or not on that criteria. This strategy is very similar to multi-attribute attitude models, like Fishbein's model (Fishbein, 1963; Fishbein & Ajzen, 1975; Ajzen & Fishbein, 1980).

Other consumer choice strategies include Phased Decision Strategies, which are used by consumers as a means of coping with a large number of choice strategies. Consumers will sequentially use at least two different strategies in this type of choice strategy. Affect Referral is yet another type of choice strategy employed by consumers. This is a special type of choice strategy that assumes the consumer has previously formed overall evaluations of each choice alternative. Instead of using evaluative criteria, the consumer simply retrieves these global evaluations from memory.

Involvement

Involvement has been operationally defined in the literature in many different ways (for a complete review see Zaichkowsky, 1986). The three most prevalent forms of involvement are message, product/brand, and purchase decision. Although the message form of involvement has been popularized in the literature by Petty and Cacioppo's Elaboration Likelihood Model (ELM) (1986), this study does not deal with the message form of involvement. Involvement is operationally defined in this study using the other two forms of involvement - product/brand and purchase decision. The design of this study using level of involvement as an independent variable is discussed in the next section.

This study has a 2 x 2 x 6 factorial design. One of the three independent variables is involvement, with two levels, high and low. There are 10 products in each of two levels of involvement. The second independent variable is country, with two levels, American and Australian consumers. Type of decision strategy is used as the dependent variable for the first series of statistical analyses and is then transformed into the third independent variable for the final analyses. For each of the 20 product categories, there are 337 data points on a 7-point scale representing the seven different choice strategies subjects could select from. The null hypotheses is: There is no significant difference between the seven different consumer choice strategies used for any of the 20 products tested for American compared to Australian subjects.

First chi-square analyses were conducted, then a t-test was conducted on each the 20 products to determine if decision strategy usage reported by Australians differed significantly from Americans. On those found to be significant a frequency analysis was conducted on the data to determine which strategies were used more often by which group of subjects. The frequencies were then used as data in a repeated measures MANOVA, with likelihood of strategy selection as the dependent variable.

METHODS

Subjects

Both American and Australian subjects were undergraduate students at a four year university. Students earned credit toward their final grade in their respective subjects in turn for their participation in the research. The age of the subjects ranged between 17 to 70 years of age. Income of the students also had a wide range.

Materials

All data was collected using questionnaires.

Design

The independent variable is country, with two levels, American and Australian. The dependant variable is the consumer choice strategy. Thus for each of the 20 product categories, each subject assigned a number, on a scale of one to seven, representing the consumer choice strategy used for that particular product.

Procedure

The questionnaire was developed in two stages. First the product categories were developed, then the choice strategies were determined. The 20 products selected, based on their level of involvement, from this questionnaire, were used in the final experimental manipulation. Each of these stages will now be discussed in detail.

Developing the Product Categories. Names of product categories were collected from subjects in a free-generation recall task. Those product categories listed by five or more subjects were used in the final questionnaire. When a single category had several different names, the name most frequently used was selected for use in the questionnaire. In case of a tie, two independent judges were consulted to break the tie. Agreement between the two judges on the product category names determined the name. Twenty of these product categories were selected for use in the experimental manipulation.

Consumer Choice Strategies. Seven consumer choice strategies found in the literature were used. General descriptions of each strategy were developed from several different sources (Engel, Blackwell, and Miniard, 1990; Howard, 1989; Bettman, 1979). The choice strategies were listed on the questionnaire as follows:

1. I compare brands based on the most important attribute. If two brands tie on the most important attribute, then I compare those brands on the second most important attribute. The process continues until the tie is broken.

2. I compare brands based on the most important attribute. Then I use a cut-off (ie. "must be under $2" or "must be nutritious") and select the brand that meets the cut-off criteria.

3. I establish cut-offs for each important product attribute. Then I compare brands one at a time, against the set of cut-offs. If a brand meets all cut-offs for all attributes, it is selected.

4. I add the number of time search brand is judged favourably in terms of a set of evaluation criteria. Then, the brand having the largest number of positive attributes is selected.

5. I assign different values (or weights) to each attribute, based on how important it is to me. Then, I determine which brands have each attribute, add the values (or weights) and select the brand with the highest score.

6. I use one of the rules listed above (1-5) to narrow down the alternatives. Then I use another one of the rules listed above to compared a few brands and make my final selection.

7. I retrieve a global evaluation of all brands form memory that is based on previously formed overall evaluations of each brand.

The Questionnaire. The 337 subjects in both the United States and Australia were asked to indicate which of the seven consumer choice strategies they use when purchasing a product for each of the product categories developed earlier. The instructions were "The following is a list of different rules consumers use to decide which brand of a product they should buy: " After strategy number six, the instruction was "Please write down which 2 rules you use and the order you use them, in parentheses after 6...i.e. cookies 6(3,1)". After the rules were listed, the following instructions were given:

For each of the above products listed on the next 2 pages, please tell me WHICH ONE of the above decision rules BEST describes what you actually do when buying a product. Put one rule number on the line provided for each product. Please leave the space blank if you have never, or will never purchase a product on the list.

Thus, subjects were giving retrospective reports about the choice strategies used for the past product purchases.

Determining Product Involvement. Another 172 subjects rated how many of five decision steps they are likely to engage in when purchasing a particular product. It is assumed that the level of product involvement is highly correlated with the number of steps on the consumer's purchase process (Beatty & Smith, 1987; Zaichowsky, 1985). Therefore, high involvement products require more steps prior to purchase compared to low involvement products, which require fewer steps prior to purchase. The description of the five steps were based on several consumer behaviour texts (Assael, 1983; Engel, Blackwell and Miniard, 1990). Instructions on the questionnaire were:

Generally, there are steps in the decision making process a consumer goes through to purchase a product. These steps are:

1. Recognising the need or problem

2. Searching for information internally (from your own memory)

3. Searching for information externally (going to or telephoning different stores, dealers, etc.)

4. Evaluating the alternative choices

5. Choosing what to buy

For each of the products on the next page, please list which of the above steps you take when buying that particular product.

For example, when I buy a refrigerator, I go through all of the above steps, so I would write 'all' next to refrigerator. However, when I buy deodorant, I only do steps 1 and 5, so I would write '1,5' next to deodorant.

If you do not buy or have never bought a particular product on the next page, please leave the space blank.

The steps listed by the subjects were coded in the same way as in Marks, Tidwell, and Spence (1991), based on the three levels of a consumer problem solving (EPS, LPS, or RPS). When a subject listed all steps, the product was coded a 5. When all steps except external search were listed, the product code was a 4. When a subject's list for a product excluded both internal and external search, the product received a 3. If the list was only 1, 4 and 5, the product was coded a 2. If only 1 and 5 are given, the product received a 1. The 10 products with the lowest coded scores were selected as the low involvement products and those with the 10 highest were selected as the high involvement products for the experimental manipulation. Those products selected for use in this study can be found in Table 1.

RESULTS

The decision strategies American consumers believe they use were significantly different for 10 of the 20 products tested when compared to those Australian consumers on all three methods of analyses used: chi-square, t-test, and multi-variate analysis of variance.

A chi-square test was conducted for each product category for American and Australian data separately, to see whether subjects preferred one decision strategy over another for any product tested. The null hypothesis was: there is no difference between any of the seven different choice strategies tested for any product. The null hypothesis was rejected for the American data, revealing significant chi-square results (p=.000) for all 20 product categories tested. However, the null hypothesis could only be rejected for certain product categories for the Australian data, as the chi-square tests showed 7 of the 10 high involvement products were not significant. Table 2 shows the products with non-significant results for the Australian data.

Table 4 shows the differences in decision strategy use between Americans and Australians. Although compensatory decision strategies were believed to be used most frequently with the products tested, Australian consumers believe they use Affect Referral more often than Americans with 8 of the 10 significantly different products. Furthermore, all 8 product categories are high involvement products. For example, although Americans believe they use the Lexicographic decision strategy most often when purchasing beer, Australians believe they use Affect Referral next most frequently, compared to Americans who report Sequential Elimination.

TABLE 1

HIGH AND LOW INVOLVEMENT PRODUCTS

TABLE 2

TABLE 3

Americans and Australians report frequent use of non-compensatory decision strategies when purchasing both high and low involvement products. Non-compensatory strategy use with low involvement products supports the Engel-Kollatt-Blackwell (EKB) model of consumer behaviour and contradicts Howard's model (Engel, et al, 1993; Howard, 1989). Consumers in this study report little or no use of compensatory strategies with high involvement products, which contradicts both the EKB and Howard models of consumer behaviour.

Parametric inferential statistics were computed to test for differences between high and low involvement products. To determine whether the three main effects of involvement, strategy, and country (and their interactions) were significant, a repeated measures analysis of variance was performed on the dependent variable, likelihood of strategy selection. Phased Decision Strategy was combined with subject non-response (those products not purchased by consumers) and was selected because it was a combination of the other strategies. This procedure also provided a single degree of freedom for the purpose of analysing the seven strategies.

The analysis of variance was conducted with products as the unit of analysis. The design was a two (country) by two (involvement) by six (strategy) factorial design with strategy as the within subject variable. There was a significant main effect of country, F(5,90)=5.72, p<.05, MS error=4.75. The main effect of involvement was also significant F(1,18)=4.66, p<.05, MS error=13.27. The main effect of strategy was also significant F(5,90)=54.24, p<.01, MS error=29.25. The interaction between involvement and strategy was significant F(5,90), p<.01, MS error=29.25, as was the interaction between country and strategy F(5,90)=17.64, p<.01, MS error 10.29. However, the interaction between country and involvement was not significant. The three-way interaction was significant F(5,90)=4.11, p<.01, MS error 42.28.

DISCUSSION

For all consumers, in both countries, different strategies are used for high compared to low involvement products. Americans believe they use choice strategies that are significantly different from those Australians believe they use for the same product categories. Although the interaction between involvement and country was not significant, this means that, although Australians differ from Americans overall, they still use the same strategies for high compared to low involvement products. These results indicate that the concept of involvement, as operationally defined in this study is generalizable across the two cultures examined in this study, and may therefore be a cross-cultural construct inherent in the perceived risk associated with particular products, similar to affordance inherent in objects in perceptual processes.

TABLE 4

PERCENTAGE OF DECISION STRATEGY PREFERENCE FOR SIGNIFICANTLY DIFFERENT CATEGORIES

The tendency for Australians to use Affect Referral decision strategies more frequently than Americans can be explained by the differences between the two cultures regarding the importance shopping has in the cultures, as described in the literature review. One would expect the Americans to use more analytical strategies based on the information overload in the culture, symptomatic of the prevalent modernism. The desire to have the latest everything typifies American culture, unlike Australians, who revel in their populism. The laid back attitudes in Australian culture explain the preference for Affect Referral over other compensatory and non-compensatory strategies.

BIBLIOGRAPHY

Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Engelwood Cliffs, NJ: Prentice Hall.

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

Dewhirst, H. D. (1989). Management in Australia: 'No Worries'. Survey of Business, 25(2), Fall 1989, 3-6.

Engel, J.F., Blackwell, R.D., and Miniard, P.W. (1993). Consumer Behavior. Chicago: The Dryden Press.

Fishbein, M. (1963). An investigation of the relationships between beliefs about an object and the attitude toward that object. Human Relations, 16, August, 233-240.

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.

Howard, J.A. (1989). Consumer Behavior in Marketing Strategy. Englewood Cliffs, NJ: Prentice Hall, Inc.

Lattas, A. (1992). Primitivism, nationalism and individualism in Australian popular culture. Journal of Australian Studies, 35, November 1992, 45-58.

Marks, Tidwell and Spence (1991). Psychology of Marketing, December 1 issue.

Patience, A. (1991). Softening the hard culture. Mental Health in Australia, 3(3), December, 29-35.

Petty, R. E. & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of Persuasion. In Leonard Berkowitz (ed) Advances in Experimental Social Psychology, 19, New York: Academic Press.

Phillips, D. (1987). Ambivalent allies: myth and reality in the Australian/American relationship. Australasian Journal of American Studies, 6(2), December, 1-8.

Polonsky, M. J. and Jarratt, D. G. (1991). A comparative study of psychographic profiles of rural shoppers. Journal of International Marketing and Marketing Research, 16(2), 85-106.

Nisbett, R.E. and Wilson, T.D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231-259.

Wierzbicka, A. (1986). Does language reflect culture? Evidence from Australian English. Language in Society, 15(3), 349-373.

Weitz, B. and Wright, P. (1979). Retrospective self-insight on factors considered in product evaluation. Journal of Consumer Research, 6(3), 280-294.

Wright, P. and Rip, P. (1980). Retrospective reports on consumer decision processes. In J. Olson (Ed.) Advances in Consumer Research, 7, 146-147.

Wright, P. and Kriewall, M.A. (1980). State-of-mind effects on the accuracy with which utility functions predict marketplace choice. Journal of Marketing Research, 17, 277-293.

Zaichkowsky, J. L. (1986). Conceptualising involvement. Journal of Advertising, 15(6).

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