Consumerism and Related Information Processing Issues: Some Observations

Cynthia J. Frey, Boston College
[ to cite ]:
Cynthia J. Frey (1984) ,"Consumerism and Related Information Processing Issues: Some Observations", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 385-386.

Advances in Consumer Research Volume 11, 1984      Pages 385-386


Cynthia J. Frey, Boston College


This paper presents a discussion of three peripherally related studies. "Is Consumerism Dead or Alive? Some New Evidence" by Smith and Bloom deals with the evolution of the consumerism movement and prospects for its future. The remaining papers, "The Effect of Nutrition Information Format on Cognitive Response, Product Evaluation, and Choice" by Rudell and "Probing the Locus of Causation in the Miscomprehension of Remedial Advertising Statements" by Jacoby, Nelson, Hoyer and Guental, are applications of information processing theory to consumer information. Due to the substantive and methodological distinctions among these three papers, each will be considered separately in this commentary.


Attempting to evaluate the future of the consumerism movement by assessing the attitudes and convictions of some of the participants has considerable merit. Consumers, consumer advocates, and public policy makers in government have different objectives and certainly different standards for measuring success.

The Smith and Bloom paper builds on the Bloom and Greyser Harvard Business Review paper that proposed the application of the product life cycle concept to the study of consumerism as a social movement. By attempting to quantify some of the relationships discussed in the earlier paper, the authors have accepted a challenge that many others would have declined as being too broad, too complicated, and too unmanageable.

The paper does, however, have its shortcomings. A fundamental problem is that the relatively small samples or policy makers and advocate organizations make quantitative analysis difficult. While there is some evidence to support the authors' contentions, the lack of availability of statistical tests due to small cell size in the data means their findings are open to broader speculation.

In spite of the weaknesses contained in this paper there is real potential to build on what has been accomplished so far.

If one turns to the literature on the evolution of social movements, the following six conditions have been found to exist and be applicable to the study of consumerism (Smelser 1963, Kotler 1972): (1) structural conduciveness -as seen by advancing incomes, education, and technical complexity of products; (2) structural strains - evidenced by general discontent with the economy and social atmosphere, ecological concern, and dissatisfaction with the political climate and business practices. These factors lead to (3) the growth of a generalized belief or popular interpretation of the problem evidenced by outspoken social advocates. It then requires (4) some precipitating factor to ignite the discontent whereby (5) individuals mobilize for action through new or existing advocate groups. The movement can be fueled or diffused by the (6) attitudes of those with social control. In this case the resistance to change or indifference evidenced by business, legislature, and the executive office.

Seen as an iterative process, as legislation or regulatory action is precipitated, the obvious agitation tends to subside. Deregulation action and perceived advocate and government inactivity in recent years has enabled consumer dissatisfaction to intensify and be rechanneled. Hence, the growth of more single issue, local advocate groups and the hard times for traditional activist organizations that have grown to the point where they may have lost their identity.

The Smith and Bloom paper provides evidence on several of these conditions contributing to the development of a social movement. Consumers' interpretation of the problem or changes in the generalized belief about consumerism plus evidence on mobilization through consumer advocate groups and indifference prevalent among regulators who maintain social control are present in this study.

By collecting additional measures and incorporating some secondary data on structural strains and conduciveness, the authors could expand their effort to include examination of the interrelationships among these factors - a more elaborate modeling approach might be considered to study the end results of consumerism. Study of the phenomenon would be of interest to marketers, public policy makers and sociologists and could lead to improved responsiveness to consumer discontent and offer feedback on the long run success of programs and policies.


The study of information acquisition style and its complications for cognitive processing and evaluation is a very complex research area. As Rudell points out in her literature review, individual differences and situational factors seem to play a determining role in the outcome of much of the experimental work with regard to consumer education information.

Rudell's findings that information presented in an attribute format results in more recipient modified protocol statements whereas brand format presentation results in more recipient generated statements is an interesting one. Since there were no differences in post treatment measures of nutrition evaluation or brand preference, however, there is only speculation as to the meaning of this finding.

Evidence indicates that type of processing during acquisition affects information organization in memory and ease of information retrieval (Biehal and Chakravarti 1982). The logical next step is to ascertain the implications of recipient modified, recipient generated, and externally generated cognitions for cognitive integration by examining memory related aspects with delayed measures.

On a related topic, the use and interpretation of nutrition information needs further examination. The meaning of nutrition quality to consumers is deserving of further study. The trade-offs consumers make between calcium and vitamin A in a forced choice experiment and the enduring nature of those cognitions will have meaning for future cognitive research as well as public policy. If consumers cannot make judgments based on the content of the information then perhaps there should be less concern with information acquisition strategies and more effort placed on study of the interpretation of information content.

Similarity of Advertising Statements

The Jacoby et al research on mi comprehension has highlighted an important point for the consideration of policy makers. Indeed, remedial advertising statements may be as confusing and misleading to the consumer as the advertising messages they are designed to counteract. Toward that end, the paper's effort to substantiate the intended similarity of three proposed remedial advertising statements should be commended.

This paper offers two potential contributions to the field: (1) an application of an elegant multidimensional scaling algorithm, COSPAR, and (2) substantive evidence as to the perceived similarity - dissimilarity of three proposed remedial advertising statements for Excedrin.

Methodology. The authors imply that COSPAR is a unique multidimensional scaling algorithm because it provides for empirical testing of goodness of fit. Indeed, the program does compute two indices (the common space index which relates to the dimensionality of the solution and the diagonality index which focuses on the geometric directions of those dimensions). These indices are then compared to norms computed for random data and tested at specific alpha levels providing an empirical goodness of fit test.

COSPAR, however, cannot be considered as the only multidimensional scaling method with this capability. For the iterative MDS algorithms such as MDSCAL, MINISSA, KYST, TORSCA, etc. (COSPAR provides an algebraic solution rather than an iterative solution) the most appropriate, although not the only, goodness of fit measure is the stress coefficient. Like the indices generated by COSPAR, stress coefficients for particular solutions can be compared to norms generated by Monte Carlo simulation techniques and tested at various alpha levels (Spence and Graef 1974). Relatively the same empirical testing capability as COSPAR is available for a variety of MDS programs

Analysis. Regardless of the particular multidimensional scaling approach that is applied, there seems to be considerable potential to shed light on the perceived similarity of specific messages. Examination of the two-dimensional solution leads the authors to conclude that the proposed remedial test statements are similar due to the overall pattern of the clustering. More careful inspection of the entire configuration, however, suggests the possibility of a degenerate solution.

According to Kruskal and Wish (1978), when the dissimilarities between objects in different clusters are all or almost all larger than the dissimilarities within each cluster, the points in each cluster will tend to converge to a single location. The three or four divergent clusters will then tend to form an equilateral triangle or tetrahedron. This condition occurs in conjunction with strong goodness of fit measures. This phenomenon is thought to be the result of a violation of the assumption that there is a smooth relationship between distance and proximity of the factors being analyzed. In the context of this data, it would appear that points 7 and 4, and 5 and 10 are conceptually so far removed from the remaining points that the large cluster (see Figures 1 and 2) is forced to converge. The suggested course of action when degeneracy is suspected is the reanalysis of each cluster separately. This would mean subsetting the similarity matrix in this case to include only statements, 1,2,3,6,8 and 9. A reanalysis might lead the authors to a different conclusion.

Based on the relative position of the proposed remedial statements, the authors suggest statements 6 and 9 are preferable to the FTC generated statements (1,2 and 3) because 6 and 9 appear less punitive and more neutral relative to the two axes of the configuration. The authors acknowledge that there is no empirical test of distance appropriate to this application.

MDS configurations like the original data are subject to substantial-variability. The authors should be cautioned about drawing conclusions with such far reaching policy implications from this configuration. As a guideline for examining relative position in MDS plots, Kruskal and Wish (1978) suggest the following "rule of thumb": "inferences should not be drawn that would change if several points were relocated by about 10% of the diameter of the overall configuration." Certainly in the present configuration none of the points 1,2,3,6,8 and 9 can be considered any more or less punitive than others within the set. If the data could be reanalyzed by cluster the distance among these remedial statements might be far more conducive to meaningful interpretation. The approach taken by the authors offers a great deal, but reanalysis of the data is in order before this contribution can be realized.


Biehal, Gabriel and Chakravarti, Dipankar (1982), "Information-Presentation Format and Learning Goals as Determinants of Consumers' Memory Retrieval and Choice Processes," Journal of Consumer Research, 8 (March), 431-441.

Kotler, Philip (1972), "What Consumerism Means for Marketers," Harvard Business Review, 50 (May-June), 48-57.

Kruskal, Joseph B. and Wish, Myron (1978), Multidimensional Scaling, Beverly Hills, CA: Sage Publications.

Smelser, Neil J. (1963), Theory of Collective Behavior, New York; The Free Press.

Spence, Ian and Graef, Jed (1974), "The Determination of the Underlying Dimensionality of an Empirically Obtained Matrix of Proximities," Multivariate Behavioral Research, 9, 331-342.