Providing Information For the Consumer Search Process

ABSTRACT - Each of the three papers presented in this session focuses on issues related to the transfer of information. The Hirschman paper focuses on information transfer within three religious groups. Specifically, novelty seeking about consumption areas and the transfer of information to others are shown to differ across religious groups. Granzin and Schjelderup examine the anticipated satisfaction that comes from information transferred to the decision maker. The decision maker's anticipated satisfaction was shown to vary across situations given the level of self confidence, perceived risk, anxiety, and expected benefits. Robin, et. al., examined the impact of information transfer that was intentionally targeted toward members of a lower socio-economic group.

Citation:

Scott M. Smith (1982) ,"Providing Information For the Consumer Search Process", in NA - Advances in Consumer Research Volume 09, eds. Andrew Mitchell, Ann Abor, MI : Association for Consumer Research, Pages: 244-246.

Advances in Consumer Research Volume 9, 1982      Pages 244-246

PROVIDING INFORMATION FOR THE CONSUMER SEARCH PROCESS

Scott M. Smith, Brigham Young University

ABSTRACT -

Each of the three papers presented in this session focuses on issues related to the transfer of information. The Hirschman paper focuses on information transfer within three religious groups. Specifically, novelty seeking about consumption areas and the transfer of information to others are shown to differ across religious groups. Granzin and Schjelderup examine the anticipated satisfaction that comes from information transferred to the decision maker. The decision maker's anticipated satisfaction was shown to vary across situations given the level of self confidence, perceived risk, anxiety, and expected benefits. Robin, et. al., examined the impact of information transfer that was intentionally targeted toward members of a lower socio-economic group.

GAINING A PERSPECTIVE ON INFORMATION TRANSFER

Information transfer either from others to oneself (defined by Hirschman as novelty seeking), or from ourselves or from an institution to others occurs as a function of the desire for information. The consumer's desire for information may, in turn, be conceptualized to activate the search process. Each of the papers discussed operationalizes at least one construct related to the process whereby the transfer of information occurs. Little theoretical explanation of this process is, however offered in any of the three papers.

Numerous variables have been theoretically linked to information search. Many of these variables have been measured, including product related, situation related, and person related dimensions (see Table 1).

The Hirschman paper examines novelty seeking and the transfer of information to normative groups. Neither of these activities are rigorously defined in terms of the nature of or direct antecedents of the information search process.

The Granzin paper examines the dimensionality of information about the product and the respondent as correlates of anticipated satisfaction. Anticipated satisfaction is judged to be analogous to expectation as defined by Oliver (1980). Although the dimensionality of "information" was related to expected satisfaction, causal linkages were not established.

TABLE 1

INFORMATION TRANSFER IN A SEARCH-SATISFACTION CONTEXT

The Robin paper measured the receipt of information using a knowledge index, given exposure to selected media and product advertisements. Although information was perceived, and knowledge gained, the dimensionality of this knowledge was not investigated.

In an attempt to integrate the viewpoints and objectives, and assess the theoretical contributions of the various papers, information search or transfer must be identified as a construct that is conceptually linked and even mediated by involvement (Newman and Staelin, 1972; Huston and Rothschild, 1977). Several forms of the involvement-search relationship may exist, with the level of mediation varying as a function of characteristics of the product, situation, or respondent.

Product variables (Bettman and Park, 1980) that influence information search and information transfer may include price, guarantee, quality, performance, credit, and service (Granzin). Situation variables (Graham, 1981) to be considered might include urgency (Granzin), financial pressures, and special buying situations. Respondent variables (Kassarjian, 1981), include personality, enjoyment of shopping, anxiety, self confidence, and demographics such as religion (Hirschman), income (Robin, et. al.), and occupation (Robin, et. al.).

The absence of a conceptual framework identifying the dimensionality of information and relating it to information search or transfer is the major deficiency of the papers reviewed. The identification of these conceptual units and linkages is of primary importance if the conceptual and measurement rigor of this research stream is to be increased.

Churchill (1979) provides guidelines for increasing measurement rigor. Of prime importance is the specification of the domain of the construct, i.e., those variables that directly and indirectly influence the information search-transfer process. The specification of the domain of a construct requires that a sample of items be generated. The dimensionality of the measures of the construct are next reduced or "purified" using alpha factor analysis (and coefficient alpha). Finally, validation of the measure is achieved by applying the multi-trait/multi-method approach to the collected data. Further analyses replicating the measures across situations, products, or patronage groups will lead to norms for expected results. It is with this foundation in mind that the information search-transfer papers are considered.

RELIGIOUS DIFFERENCES IN COGNITIONS REGARDING NOVELTY SEEKING AND INFORMATION TRANSFER

The Hirschman paper, through a brief review of interdisciplinary sources, makes it clear that religious differences impact on consumption. This relationship may be either direct, as when consumption of specific products varies as a function of the tenets held by the religious group, or indirectly, as a function of differences in fertility rates, political orientation, or even psychological constructs such as personality, self confidence, or aggressiveness.

Three religions - Catholicism, Protestantism, and Judaism - were investigated as cognitive systems. The term "cognitive systems" refers to beliefs, values, and expectations held and shared by the religious groups. The study design evaluated these systems while controlling for known differences in socio-economic status that existed across the religious groups. Strength of religious affiliation was similarly measured and assumed constant across religious groups. Unfortunately, this latter assumption may be questioned in that the large confidence interval associated with the Scheffe technique would tend to find no differences between the religious groups. A Student's t-test is suggested as the correct statistic for this analysis.

It is clear that the express purpose of the study is not to establish the dimensionality of information search or transfer. However, given the measures of novelty seeking and the transfer of information, a significant need for this conceptualization is present. Multiple measures, or a summative measure of the willingness to search for or share information would increase the validity of the conclusions.

The crux of the Hirschman study is a factor analysis of the 15 product areas, where the factors produced are indicative of product areas where a similar willingness to "try" or share exists. Although it is clear that factors should be labeled, the interpretation of these factors would seem to be overly strong. Because factors labeled "sensory experiential consumption," or "mass media consumption" differ across groups, differences in overall set or MMC are claimed to exist across religious groups. It should be recalled that the products included in the analysis were selected by judgmental sample. Confirmatory factor analysis using multiple measures of the resulting factors is suggested as an avenue for further research.

As a final methodological note, it is more expedient to conduct a three group discriminant analysis in lieu of comparing the three distinct factor structures, each of which was produced in a separate analysis. The discriminant analysis would provide a direct test of differences between the religious groups and avoid the problem of questionable inferences leading from the interpretation of the individual factors.

SITUATIONS AS AN INFLUENCE ON ANTICIPATED SATISFACTION

The Granzin and Schjelderup paper is an attempt to model satisfaction as a function of a variety of situation and enduring personal variables. Single measures of anxiety and self confidence, as well as multiple measures of product benefits. risks, and information search were included in the model and analysis.

The inclusion of multiple measures is a noteworthy methodological strength of the paper. This strength would be enhanced further, as would the model by incorporating multiple measures of the anxiety and self confidence constructs. Such a modification would provide for treating the multiple measures as random variables so that an alpha factor analysis could be conducted to purify the measures (Churchill, 1979). As a second benefit of the multiple measures approach, the dimensionality of the expected satisfaction could be established using path analysis.

The major question that can be raised about the study is a definitional question of the dependent variable: anticipated satisfaction. "Anticipated Satisfaction" measures the expected satisfaction resulting from the selection of one of eight types of car repair facilities given a particular car repair problem and situation.

Oliver (1980) would argue that the dependent measure is not a measure of anticipated satisfaction, but rather expectation. It is clear that the dependent measure is not a direct measure of satisfaction, but rather a measure of expectancy, and is defined as originating in expectancy theory. The difference would therefore appear to be one of semantics rather than one of substance, with "expected satisfaction" capturing the essence of the dependent measure.

As one final note. the results of the study reported as correlations between anticipated satisfaction and the situation influenced intervening variables are in many cases weak, but statistically significant. Individual correlations rarely exceed .20, explaining less than 4 percent of the variance. In conclusion, the strength of these findings may be increased by including alternative measures of the intervening variables. Alternatively, the intervening variables, if "purified," would provide single score or measure for the construct under investigation. This construct would necessarily provide greater explanation of the satisfaction variable. Finally, increasing the rigor of the definition of the dependent variable (an operationalization using actual satisfaction) is suggested.

ATTACKING THE KNOWLEDGE GAP PHENOMENON

The paper by Robin, et. al., is an applications paper demonstrating that differential acquisition of advertising information occur across social class groups. The increase in knowledge of three bank services was measured using telephone interviews of respondents from the respective-social classes. As a preliminary issue, one must question why social class is investigated as an intervening variable in the dissemination of information. Similarly, the importance of the upper-lower social class must be questioned as an alternative to the lower-lower, or any other social class.

Market analysts would normally forego the ad hoc identification of any market group in favor of a market profile of patrons of each bank service. Unless members of the upper lower social class are an integral part of the bank services markets, the dissemination of information that occurs to the social classes is of little consequence.

The three papers show commonality when considered within the framework of using measures of information search. All too often studies are completed without consideration of the theoretical linkages preceding or mediating the information transfer process being investigated. The methods of measurement refinement and path analysis provide useful insights for dealing with these problems.

Researchers and practitioners are urged to consider information search theory and measurement refinement techniques in dealing with the information transfer problem. Without search, there can be no effective transfer of information.

REFERENCES

Bettman, J. R., and Park, C. W. (1980), "Effects of Prior Knowledge and Experience and Phase of the Choice Process on Consumer Decision Processes: A Protocol Analysis," Journal of Consumer Research, 7, pp. 234-247.

Churchill, G. A., Jr. (1979), "A Paradigm for Developing Better Measure of Marketing Constructs," Journal of Marketing Research, 16, pp. 64-73.

Graham, R. P. (1981), "The Role of Perception of Time in Consumer Research," Journal of Consumer Research 7 pp. 335-342.

Huston, M. J., and Rothschild, M. L. (1978), "A Paradigm for Research on Consumer Involvement," Unpublished Working P per, University of Wisconsin.

Kassarjian, H. H. (1980), "Low Involvement: A Second Look," in Advances in Consumer Research, Vol. VII, . Rent B. Monroe, (ed.), Ann Arbor: Association for Consumer Research.

Newman, J. W., and Staelin, R. (1972), "Prepurchase Information Seeking for New Cars and Major Household Appliances," Journal of Marketing Research, 9, pp. 249-257.

Oliver, R. L. (1980), "A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions," Journal of Marketing Research, 17, pp. 460-469.

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