Correlates of Information Gathering and Evoked Set Size For New Automobile Purchasers in Norway and the U.S.

R. Neil Maddox, University of Missouri-St. Louis
Kjell Gronhaug, Norwegian School of Economics and Business Administration
Richard E. Homans, University of Missouri-St. Louis
Frederick E. May, University of Missouri-St. Louis
ABSTRACT - Interviews conducted with new automobile purchasers in the U.S. and Norway are analyzed in this study. Correlates of information gathering activities and the size of the evoked set in the two samples are compared. Similar patterns of relationships are found in the two samples. The Norwegian respondents generally reported higher levels of search which seems consistent with known structural differences between the two cultures.
[ to cite ]:
R. Neil Maddox, Kjell Gronhaug, Richard E. Homans, and Frederick E. May (1978) ,"Correlates of Information Gathering and Evoked Set Size For New Automobile Purchasers in Norway and the U.S.", in NA - Advances in Consumer Research Volume 05, eds. Kent Hunt, Ann Abor, MI : Association for Consumer Research, Pages: 167-170.

Advances in Consumer Research Volume 5, 1978      Pages 167-170

CORRELATES OF INFORMATION GATHERING AND EVOKED SET SIZE FOR NEW AUTOMOBILE PURCHASERS IN NORWAY AND THE U.S.

R. Neil Maddox, University of Missouri-St. Louis

Kjell Gronhaug, Norwegian School of Economics and Business Administration

Richard E. Homans, University of Missouri-St. Louis

Frederick E. May, University of Missouri-St. Louis

ABSTRACT -

Interviews conducted with new automobile purchasers in the U.S. and Norway are analyzed in this study. Correlates of information gathering activities and the size of the evoked set in the two samples are compared. Similar patterns of relationships are found in the two samples. The Norwegian respondents generally reported higher levels of search which seems consistent with known structural differences between the two cultures.

INTRODUCTION

Even though the influence of culture upon consumer behavior is widely acknowledged, cross-cultural studies of the consumer's decision process are conspicuous by their absence (Nicosia and Mayer, 1976). While there are notable exceptions (Douglas, 1976; Pfaff and Blivice, 1977), the costs and difficulties of cross-cultural research seem to have dissuaded most consumer behavior researchers from initiating investigations spanning national boundaries (Sheth, 1974).

Both theoretical and managerial concerns dictate pursuing studies in a number of different settings. From a scholarly point of view, such studies offer an invaluable arena for developing and testing theories. It is frequently held that models of consumer behavior capture fundamental human decision processes (Markin, 1974, p. 55). If so, these models should be applicable in a wide range of situations if the background variables are known. Such beliefs cannot be tested if empirical studies are only conducted within the relatively homogeneous confines of a single society. Our confidence in the generality of a theory is heightened if its descriptions of a process are found to be accurate in a wide range of settings.

More pragmatically, given the rising importance of multinational businesses, cross-cultural studies are vitally important to the marketing manager. Many foreign markets are rapidly developing from a state of scarcity to one of abundance (Sheth, 1974). Consumers in a growing number of countries have a wide range of choices among products and brands. They have, perhaps for the first time, the opportunity to make consumption decisions. Just as was the case in the United States (Engle, Kollat and Black-well, 1973, p. 46), as goods become plentiful it becomes imperative to develop an understanding of the factors influencing consumers as they make their choices.

Further, efficiency in the employment of marketing resources requires the identification of similarities as well as differences between cultures (Douglas, 1976). If the multinational firm is to succeed in truly developed markets, it must realize economies of scale in the application of its marketing efforts that are not available to its national or regional competitors. If the multinational corporation serves only an ad hoc collection of discrete national markets that must be approached independently, it will be at a disadvantage. Without some offsetting competitive advantage, the larger firm merely adds layers of overhead costs that must eventually be borne by each of the local operations. If it is to become a truly viable institution in a world of developed economies, the multinational firm must be able to serve individual markets at least as efficiently as its national counterparts. It would be desirable to identify cross-national similarities that can be used to segment markets and develop common marketing strategies.

Detailed explorations of cross-cultural similarities in consumer decision processes will entail large-scale and expensive research efforts. Before considering or recommending such an undertaking, one should wish to see some indications of the likelihood of its success. This paper represents a modest attempt at such a demonstration. Converging findings from studies of automobile decision making in two different cultures are discussed.

A SIMPLIFIED MODEL

Extensive models of the consumer decision process have been presented by a number of authors (Nicosia, 1966; Howard and Sheth, 1969; Engel, Kollat and Blackwell, 1973). These are overly elaborate for the present purposes, given the state of the art in cross-cultural studies of consumer behavior. We will constrain the discussion to the simpler model shown in Figure 1. At the beginning of any decision problem the consumer has stored in his mind a bundle of attitudes, beliefs, knowledge, evaluative criteria, etc. These result from the unique characteristics of the society within which he lives, individual differences among the members of that society (e.g., personality, intelligence, and demographics), his previous experience with the product and the making of similar decisions.

FIGURE 1

A SIMPLIFIED MODEL OF THE CONSUMER DECISION PROCESS

Societal differences are viewed as being of two sorts. First, there are the more measurable structural differences such as income levels, the general development of the economy and the availability and cost of the product and product related information. Secondly, there are the more ephemeral aspects of culture including norms, values and mores, etc.

These exogenous or background variables are fixed at the onset of the decision task. They are shown as influencing both the acquisition of information and the formation of the evoked set, that limited group of alternatives actively considered by the decision maker (Howard and Sheth, 1969, p. 26). Information gathering may vary both in terms of the overall level of activity and the consultation of specific sources. Evoked sets may be analyzed in terms of their size and the balance between familiar and unfamiliar brands as well as according to the specific brands which they contain.

Information seeking should influence the size and composition of the evoked set. Finally, a feedback relationship is shown, since the perceived satisfactoriness of the evoked set would be expected to affect the course and duration of information seeking activities.

Studies of these relationships in distinctly different cultural settings are discussed in the following sections. It should be emphasized that our concern does not lie with demonstrating the presence or absence of between-culture differences with respect to the level on any variable. Rather, we are interested in exploring cross-cultural similarities in the relationships between the different sets of variables.

THE STUDIES

Two nearly contemporaneous studies are compared and contrasted here. The United States study was conducted in St. Louis, Missouri. Data for the Norwegian study were gathered in Bergen, Norway.

Samples

Norwegian Study. Personal interviews averaging 45-60 minutes were conducted with a sampling of new car registrants during the late fall of 1969. Time and financial constraints forced limitation of the sample to Bergen, Norway's second largest town. Individuals were selected from transcripts of new car registrations. After as many as four recalls, completed questionnaires were obtained from 101 of the 138 individuals drawn for the sample. Five of the interviews could not be used because the automobiles were found to have been purchased for some commercial purpose or due to incomplete data.

United States Study. The data for this study are from personal interviews with 387 household heads residing in the St. Louis metropolitan area. A standard cluster-area probability sampling procedure was used to select a sample of dwelling units in the metropolitan area. Interviews were conducted by the Public Opinion Survey Unit of the University of Missouri in the Spring of 1966. Up to four interviewer calls were made at each selected dwelling unit. The response rate of 77 percent of attempted interviews appears reasonable for a large metropolitan area and for a high proportion of nighttime interviews. Respondents were asked to recall and describe all cars owned and previously owned by themselves and their family. Next, each respondent was asked to describe the decision leading to his or her most recent purchase. The analysis reported here is based upon data from those 132 household heads whose most recent purchase was a new car, who owned at least one car before the most recent purchase, whose brand choice was not restricted by special deals, and whose answers to the questions used were ascertained.

Measures and Analysis

Each study gathered a large amount of information about the individual, his automobile ownership history and his most current auto purchase decision. A number of questions reflecting on the relationships shown in Figure 1 were common to both studies. These are reproduced in Appendix A.

It seemed desirable to make minimal assumptions about the level of measurement. Even though some variables were measured on an interval scale, Kendall's Tau, which requires only ordinal measures, was used throughout the analysis unless another statistic is indicated.

FINDINGS AND DISCUSSION

The patterns of findings shown in Tables 1-4 are supportive of the notion that similar processes operate in the two cultural settings. Where differences are noted at least some explanation can be offered in terms of the effects of structural differences.

Demographic Variables and Evoked Set Size

The relationships between demographic variables and the size of the evoked set are shown in Table 1. The similarities between the two studies are striking. Age and education are significantly related to evoked set size in both studies; income is found to be lacking in explanatory power. The direction of the significant relationships is as expected. As the individual ages, he accumulates experience with the product and product related decisions and he develops a more completely formed preference ordering (Howard and Sheth, 1969, pp. 26-29). The elimination of marginally interesting brands, coupled with a finite upper limit on the number of brands of automobiles available, would lead to smaller evoked sets. Similarly, a positive relationship between evoked set sizes and education was anticipated. If the educational system is successful, it should develop an individual with heightened capabilities and inclinations to retain and manipulate information. An educated person would be expected to discover and consider more alternatives than one not so favored.

TABLE 1

RELATIONSHIPS BETWEEN EVOKED SET SIZE AND DEMOGRAPHIC VARIABLESA

Demographic Variables and Information Seeking

These relationships are shown in Table 2. The pattern is much the same as that noted in Table 1. Age shows significant negative correlations with all measures of information seeking. Similar arguments to those advanced in the preceding paragraphs apply here; as the individual develops greater total experience he has less need to gather information specifically for the present decision.

With one nonsignificant exception, the relationships with education are, predictably, positive. The fact that a far greater number of significant correlations are found in the Norwegian study does not admit to simple explanations. Perhaps this reflects the emphasis upon consumer education found in Scandinavian educational systems (Thorelli, 1977).

The findings as to the effects of income seem confusing. The Norwegian data show consistently negative signs. This would support a hypothesis that the higher the income the lower the perceived risk and, hence, the need to acquire information as a risk reducing mechanism (Engel, Kollat and Blackwell, 1973, p. 587). This hypothesis would not explain the United States data where the significant relationships are positive. In this instance availability may be the key. The significance of advice results from the tendency of higher income persons to consult with individuals having technical or professional expertise; income is negatively related to discussing the purchase with friends, relatives and co-workers (data not shown). Higher income persons may have easier access to those with professional knowledge of automobiles. A similar situation may prevail with brochures. A substantial number of those reporting they had read brochures indicated that these had been mailed to them by the seller without any initiative on the buyer's part. Since they are attractive prospects, this may reflect the fact that higher income consumers are likely to find themselves on a large number of mailing lists. The interesting thing is that these unsolicited brochures were not only received, but also read and remembered.

TABLE 2

RELATIONSHIPS BETWEEN DEMOGRAPHIC VARIABLES AND USE OF INFORMATION

The (nonsignificant) negative sign found for test driving in the United States data may be reflecting the influence of compliance rather than education. Paraphrasing one respondent, "The salesman is supposed to get you to test it, so you drive it a few blocks so he doesn't get in trouble." Perhaps less educated respondents are more easily cowed by auto salesmen.

Information Seeking and Evoked Set Size

It was expected that the level of information seeking would be positively related to the size of the evoked set. Two considerations gave rise to this prediction. First, both the level of information seeking and the evoked set are believed to be associated with the complexity of the decision process. Small evoked sets (approaching one) and minimal information gathering are characteristic of routine response behavior (Howard and Sheth, 1969, p. 528). Problem solving (limited or extended) is believed to be associated with larger evoked sets and more active efforts to acquire information.

As shown in Table 3, this expectation is supported by both sets of findings: larger evoked sets are positively related to more intense information gathering. With the exception of advice, stronger relationships are found for the Norwegian respondents. This can, perhaps, be explained by structural differences between the two societies. At the time of the studies automobiles were two or three times as expensive in Norway as in the United States and incomes were roughly 30-40% lower. As a re-suit, a far smaller proportion of the households owned a car. Since the Norwegian buyer was committing a larger proportion of his income to a product with which he was less familiar than his United States counterpart, he would be expected to engage in more extensive problem solving.

TABLE 3

RELATIONSHIPS BETWEEN EVOKED SET SIZE AND USE OF INFORMATION

SUMMARY

The case for cross-national studies of consumer behavior was argued in the introduction. Such studies are needed both for theory development and for the managerially useful insights which may be gained. A simplified model was then presented. Findings from studies of Norwegian and American automobile purchasers on the relationships contained in this model were then compared. Differences in the level or intensity of information search activities were noted. It seemed that these differences could be attributed to structural differences between the two cultures. However, the most notable findings were the similarities in the nature of the relationships between variables. There seemed to be strong indications that similar underlying processes were operating in these quite different societies.

APPENDIX

QUESTIONS ASKED OF RESPONDENTS AND THEIR CODING

REFERENCES

Susan P. Douglas, "Cross National Comparisons and Consumer Stereo Types: A Case Study of Working and Non-Working Wives in the U.S. and France," Journal of Consumer Research, 3(June, 1976), 12-20.

James F. Engel, D. T. Kollat and R. D. Blackwell, Consumer Behavior, Rev. Ed. (Homewood, Ill.: Dryden Press, 1973).

John A. Howard and J. N. Sheth, The Theory of Buyer Behavior (New York: John Wiley & Sons, Inc., 1969).

Rom J. Markin, Jr., Consumer Behavior: A Cognitive Orientation (New York: MacMillan, 1974).

F. M. Nicosia, Consumer Decision Processes: Marketing and Advertising Implications (Englewood Cliffs, NJ: Prentice-Hall, 1969).

Francesco M. Nicosia and R. N. Mayer, "Toward a Sociology of Consumption," Journal of Consumer Research, 3(September, 1976), 65-75.

N. H. Nie, C. H. Hull, J. G. Jenkins, K. Steinbrenner and D. H. Bent, Statistical Package for the Social Sciences, 2nd Ed. (New York: McGraw-Hill Book Company, 1975).

Martin Pfaff and S. Blivice, "Socio-Economic Correlates of Consumer and Citizen Dissatisfaction and Activism," Paper presented to The Research Symposium on Consumer Satisfaction/Dissatisfaction and Complaining Behavior, Indiana University, Bloomington, Ind., April 21-22, 1977.

Jagdish N. Sheth, "The Next Decade of Buyer Behavior Theory and Research," in J. N. Sheth, Ed., Models of Buyer Behavior: Conceptual, Quantitative and Empirical (New York: Harper and Row, 1974).

Hans G. Thorelli and Y. R. Puri, "On Complaining in Norway and the Role of Information Seekers," Paper presented to The Research Symposium on Consumer Satisfaction/Dissatisfaction and Complaining Behavior, Indiana University, Bloomington, Ind.: April 21-22, 1977.

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