An Experimental Examination of Deception in Labeling: Consumer Research and Public Policymaking

ABSTRACT - This study expands knowledge about potential sources of marketplace deception in two ways. First, conceptual and methodological insights evolving from research in deceptive advertising are applied to the empirical study of deception in labeling. Second, since existing U.S. Department of Agriculture labeling regulations affecting actual consumer meat products are examined in the study, consumer research input to public policymaking processes are illustrated.

Citation:

Philip G. Kuehl and Robert F. Dyer (1978) ,"An Experimental Examination of Deception in Labeling: Consumer Research and Public Policymaking", in NA - Advances in Consumer Research Volume 05, eds. Kent Hunt, Ann Abor, MI : Association for Consumer Research, Pages: 206-212.

Advances in Consumer Research Volume 5, 1978      Pages 206-212

AN EXPERIMENTAL EXAMINATION OF DECEPTION IN LABELING: CONSUMER RESEARCH AND PUBLIC POLICYMAKING

Philip G. Kuehl, University of Maryland

Robert F. Dyer, George Washington University

ABSTRACT -

This study expands knowledge about potential sources of marketplace deception in two ways. First, conceptual and methodological insights evolving from research in deceptive advertising are applied to the empirical study of deception in labeling. Second, since existing U.S. Department of Agriculture labeling regulations affecting actual consumer meat products are examined in the study, consumer research input to public policymaking processes are illustrated.

INTRODUCTION

The subject of deception in advertising, packaging, and labeling has emerged as a highly visible public policy-related issue among consumer researchers in recent years. Three major factors have contributed to such widespread interest in examining marketplace practices which exhibit the potential to deceive consumers.

First, the so-called consumer movement has focused much of its recent emphasis on encouraging business firms and public sector institutions to implement policies (i.e., the "right to be informed") which strengthen the availability of consumer information in the marketplace. Second, public sector institutions like the Federal Trade Commission (FTC) and Food and Drug Administration (FDA) have pursued policies to reduce deceptive practices in the marketplace. Similarly, agencies like FDA and the U.S. Department of Agriculture (USDA) have funded large-scale consumer research studies to examine a variety of product packaging and labeling policy alternatives (e.g., open-dating provisions; product storage and usage instructions; disclosure of ingredients statements; nutritional labeling; "package inserts" with consumer information; and "safety" caps and injury prevention packaging in poisons, medicines, etc.). Third, many companies have initiated activities in support of consumers' rights to product information in the marketplace by sponsoring innovative consumer education programs. In the Washington, D.C. metropolitan area, for example, Giant Foods, Inc. has embarked on a series of such consumer education activities (under the leadership of Esther Peterson who is now Special Assistant to the President for Consumer Affairs) which focus on improving the consumers' ability to make intelligent purchase decisions.

OBJECTIVES

While the preceding factors highlight general concern about alleged deceptive practices in the marketplace, conceptual and empirical studies published by consumer researchers to date have focused primarily on the issue of deception in advertising. As a result, there appears to be a need for research which expands this existing research tradition to include other sources of potential deception. The present study helps to fulfill this need by providing insights on two major objectives.

1. To determine if conceptual and methodological knowledge about deception in advertising is applicable to measuring deception in consumer product labeling

2. To provide consumer research input to public policymaking processes by examining whether or not existing USDA-approved labeling is likely to stimulate falsely based consumer beliefs associated with the attributes of two alternative meat products

Since the conceptual objective of the study relates to over 25 major articles dealing with defining deception in advertising, the following discussion summarizes this body of knowledge and assesses its applicability to measuring deception in labeling. Next, the public policy related objective of the study--which involves actual USDA labeling regulations for "ham" and "turkey-ham" products--is discussed. The study methodology, findings, and conclusions are presented in the remaining three major sections of the paper.

CONCEPTUAL AND METHODOLOGICAL FRAMEWORK

The framework for the conceptual and methodological approaches used in this study evolves from published literature on deceptive advertising. This academic research tradition can be classified into three major types of contributions.

Preliminary Methodological and Conceptual Insights

Results from four studies in the literature (Hunt, 1972 and 1973; Dyer and Kuehl, 1974; and Kassarjian, Carlson, and Rosin, 1975) provided two basic insights on conceptualizing and measuring deception from a methodological perspective. First, all of these authors recognized that experimental research designs (vis-a-vis cross-sectional studies) are the most appropriate methodological framework for investigating deception. In other words, these studies illustrate a need for empirical research in consumer deception to be based upon "cause and effect" research designs--where degrees of alleged deception can be directly related to specific stimuli (i.e., either a product label or an advertisement).

The second methodological insight from these early studies in deceptive advertising concerns the selection of appropriate dependent variables to measure deception. In all of these studies, such broad or "global" variables as consumer attitudes toward the "product brand" or "brand manufacturer" were used. In other words, traditional aggregate consumer response variables were used in these studies. While this measurement approach reflected the "state of the art" at that time, other academicians recognized a need to conceptualize innovative variables for examining deception which are more precise and reliable in nature.

Defining Deception in Conceptual Terms

Conceptual contributions to defining consumer deception encompass two major themes. First, many authors (Aaker, 1974; Armstrong and McLennan, 1973; Cohen, 1969 and 1972; Dillon, 1973; Permut, 1974; Ford, Kuehl and Reksten, 1975; Rosch, 1975) developed typologies for classifying and measuring characteristics of alleged deceptive practices. In essence, these authors have illustrated that the term "deception" is multidimensional in nature and that it is possible to define a variety of different forms of deception (i.e., the "blatant lie," the "partial truth," "claim-fact discrepancy," the "false certification," etc.). These discussions suggest that empirical research on deception must recognize the specific type or form of deception represented in the alleged misleading stimuli vis-a-vis adopting an aggregate meaning for the term "deception."

Second, several other authors (Jones, 1971; Wilkie and Gardner, 1974; Jacoby and Small, 1975; Brandt and Preston, 1976) developed conceptual frameworks for examining deception within the policymaking processes of public sector institutions like the FTC and FDA. In essence, these authors developed approaches for empirically defining consumer deception in an operational context that is consistent with traditional legal doctrine based approaches.

Brand Beliefs: Concepts and Research

The preceding two dimensions of academic work on defining and measuring consumer deception were valuable in stimulating a growing professional dialogue in the subject area. However, it was not until the contributions of Wilkie and Gardner in 1973 that the "belief" component of the generalized Fishbein attitude model was recognized as an appropriate and precise variable to measure the effects of alleged deceptive practices in the marketplace. In proposing the use of brand attribute "beliefs" vis-a-vis overall attitudinal variables to study deception, these two authors suggest that consumer beliefs about specific product attribute claims in specific marketing stimuli should be used to measure deception. Related to this study, a product label may be viewed as stimulating an "incorrect" or "misleading" consumer perception about selected product attributes (e.g., the ingredients of the product) but may not mislead the consumer on other product attributes (e.g., the nutritional value of the product).

The conceptual efforts of Wilkie and Gardner have spawned four research efforts to date (Dyer and Kuehl, 1974; Mazis and Adkinson, 1976; Kuehl and Dyer, 1976; and Kuehl and Dyer, 1977). In each of these studies, classic experimental research designs have been used to measure consumer "beliefs" about specific claims made about the product in a firm's promotional efforts. Further, the relatively consistent results from these studies support the views of Wilkie and Gardner that consumer beliefs--measured within an experimental research design--should be used to assess the impact of alleged deceptive marketing stimuli.

Summary Conclusions

The preceding discussion has highlighted the major conceptual and empirical contributions which consumer research has made to the study of deception in advertising. In relating this work to the present study, the following summary conclusions are appropriate:

1. An experimental research design should be used to study deception--so that consumer response measures can be directly related to specific claims made in product labeling through a "cause and effect" methodology.

2. Consumer "beliefs"--examined on a specific product attribute basis--should be used to measure deception in labeling.

3. The specific type of deception reflected in alleged misleading statements about the product must be identified. In this study, it appears a priori that labeling used for "turkey-ham" meat products is a "claim-fact" discrepancy. In other words, consumer perceptions of the "claim" that "turkey-ham" is made from a combination of meat and poultry products are not supported by the technical "fact" (i.e., this product is made only from turkey meat).

4. The research should be consistent with viable public policymaking alternatives in the agency with jurisdictional responsibility for the alleged deception. In this study, for instance, deception in labeling should be related to potential, (a) pre-market, (b) post-market, and (c) consumer education policy actions at USDA.

PUBLIC POLICY FRAMEWORK

The framework for the public policy objective of the study encompasses two classes of issues. From a general perspective, it should be noted that product labeling--as a point-of-sale marketing stimulus--is receiving increased attention among public policymakers at the present time. For instance, FTC recently has instituted hearings on "information disclosure" proposals affecting non-prescription drug product labeling regulations. Similarly, the FDA had undertaken a systematic review of food product labeling policies related to provisions of the Fair Packaging and Labeling Act. As a result, research which examines product labeling as a potential source of consumer deception contributes knowledge in support of this increased policymaking activity.

From the perspective of USDA policymaking, the study extends the work of Miller, Topel, and Rust (1976). Their empirical work, which focused on USDA labeling regulations affecting beef grading, supported the view that there is a need to investigate the meat labeling practices and that USDA should reconsider the ". . . role it plays as a source of information to food marketers and consumers alike" (Miller, Topel, and Rust, 1976; p. 31).

In relating this viewpoint to the present study, the USDA's promulgation of two regulations affecting the labeling of "ham" and "turkey-ham" meat products provide the specific focus for the experiment. First, USDA policymakers have adopted a dictionary-based definition for the technical term "ham" (i.e., the thigh cut of meat from the hind leg of any animal). Second, USDA has approved the terminology "turkey-ham" for the labeling of a meat product which includes only turkey meat (i.e., no "ham" meat).

The justification for this label wording approval, according to USDA, is based on two major policy arguments. First, since the term "ham," in USDA parlance, refers to a cut of meat (i.e., the thigh portion of the hind leg of any animal) rather than a type of meat (i.e., smoked and cured pork meat), it is technically correct to associate "turkey" and "ham" wording on a product label. Second, while turkeys have only two legs, USDA has decided that the wings of this bird constitute two additional locomotive extremities on the body. As a result, meat taken from the thigh portion of the turkey leg can be labeled with the term "ham."

Although this USDA logic appears unduly naive from a consumer research view, the major issues in USDA policymaking are clarified by the following questions:

1. Do consumers, as a result of their exposure to USDA approved product labeling, exhibit beliefs about specific "turkey-ham" product attributes which are not supported by factual evidence on this product?

2. Do consumers, as a result of their exposure to USDA approved product labeling, exhibit beliefs about specific "turkey-ham" product attributes which are supported by factual evidence on this product?

3. What are consumer perceptions or understanding of the USDA definition of the term "ham"?

METHODOLOGY

The Experimental Design

The basic experimental design used in the study is an "after-only control group" approach. This design was chosen for three major reasons. First, this design contains two basic factors needed for classic experimental research (i.e., (1) "control" and "experimental" groups, and (2) random assignment of respondents to the two groups). Second, this design is particularly well-suited for multi-site field interviewing since the administrative and supervisory procedures needed to support this design are not complex (i.e., as a result, potential errors in the implementation of the experiment are reduced). Third, this design is especially appropriate to examine the major policy issues in the study. In other words, "ham" product labeling can viewed as a "control" variable in the experiment since it is the traditional meat product for which "turkey-ham" is a substitute. Furthermore, if consumers are deceived by the "turkey-ham" label wording (i.e., the experimental variable) it is likely that the effect of such deception most likely will be a decreasing market share for "ham" meat products.

FIGURE 1

THE "AFTER-ONLY WITH CONTROL GROUP" EXPERIMENT DESIGN

Study Procedures

Four interviewing sites (i.e., New York, Chicago, Kansas City, and Los Angeles) were used to implement the study design. At each site, the following procedures were used:

1. Professional interviewers were recruited and trained in the methods and procedures to be used in the study. All of the interviewers were employed by local, independent interviewing services in each city.

2. A nonprobability sample of 100 respondents was selected for the study at each site (N=400). None of the respondents were paid to participate in the study.

3. Each respondent at each site was assigned a number from "1" to "100" and the Phase 1 data (general food knowledge) was collected from all respondents.

4. In Phase 2, odd numbered respondents (i.e., the "ham" product group) examined the ham package and label. The even numbered respondents (i.e., the "turkey-ham" group) followed the same procedure by examining a "turkey-ham" package and label. After examining their respective package and label, a seven-point, seven item attribute scale was used to obtain belief data. The attributes examined in the study included:

a. Nutrition Beliefs

b. Packaging Safety Beliefs

c. Product Content Beliefs: "Pork" Meat Only

d. Product Content Beliefs: More Than One Meat

e. Product Content Beliefs: Some "Turkey" Meat

f. Product Content Beliefs: "Turkey and Ham" Meat

g. Product Content Beliefs: Turkey Meat Only

5. Actual product packages and labels were used in the experiment: ITT Gwaltney Ham and Rich's Turkey-Ham.

6. Phase 3 (sociodemographic data) was collected from all respondents).

Data Analysis

Traditional data analyses were conducted on the study data in the following ways:

1. Phase 1 data were: (a) tabulated to provide general insights on respondent food knowledge, and (b) statistically analyzed to ensure that the "turkey-ham" and "ham" groups were equivalent in terms of general food knowledge.

2. Phase 2 data were analyzed by T-Tests which statistically examined whether or not consumer beliefs about "turkey-ham" and "ham" are significantly different as a result of exposure to the two product labels.

3. Phase 3 data were: (a) tabulated to provide a general profile on overall study respondent sociodemographic characteristics, and (b) statistically analyzed to ensure that the "turkey- ham" and "ham" groups were equivalent in terms of sociodemographic characteristics.

RESULTS

Table 1 presents the product attribute beliefs of respondents which were generated through the experimental procedures used in the study. This discussion is followed by findings illustrating the general food knowledge and sociodemographic characteristics of the respondents. The discussion is organized in terms of the major public policy issues presented previously.

TABLE 1

"HAM" GROUP AND "TURKEY-HAM" GROUP BELIEFS

Attribute: More Than One Type of Meat

After their exposure to the ham label, respondents in the "ham" group tended to believe (X = 4.02) that the product they examined did not contain more than one type of meat product. On the other hand, respondents who were exposed to the "turkey-ham" product label tended to believe that this product did contain more than one type of meat product (X = 4.75).

In general, the factually "correct" answer on the seven-point scale for respondents in both the "ham" and "turkey-ham'' groups is "1" since, in fact, both of the representative products contain only one type of meat (i.e., "ham" and "turkey" meat respectively). Stated differently, respondents in each group who perceive correctly that the product package and label they have examined has only one meat ingredient should have indicated a low number on the seven-point scale (i.e., a higher number on the seven-point scale would suggest that a consumer is likely to believe that the product has more than one meat product as an ingredient).

In the present data, the statistically significant lower belief scores for the "ham" group respondents--(compared to "turkey-ham" group respondents)--suggest that ham group respondents tended to correctly perceive that the ham product is made only of one meat product. Conversely, the statistically higher mean belief values for turkey-ham group respondents (compared to ham group respondents) suggest that these consumers believe the turkey-ham product is made of more than one meat product. Since ham group respondent beliefs (evolving from their exposure to the ham product label) agree with "fact," while turkey-ham group respondent perceptions do not agree with "fact," it appears that consumers are deceived and mislead by the "turkey-ham" wording in the label in both package size categories.

In summary, it appears consumers are deceived in their beliefs about the "more than one type of meat" attribute of the turkey-ham product--based on the exposure to the label on the product. Consumers are not deceived in their beliefs about the "more than one type of meat" attribute of the ham product--based on their exposure to the label on the product.

Attribute: Both "Turkey" and "Ham" Meat. As expected, the "ham" group respondents do not believe that the ham product contains both turkey and ham meat since the mean belief value is extremely low (X = 1.70). On the other hand, the "turkey-ham" respondent group mean value (X = 4.31)--a significantly higher belief value than the ham group--suggests that respondents believe the turkey-ham product to contain both turkey and ham meat.

This finding suggests that ham group respondents, based on their exposure to a ham product label, correctly perceive that ham does not contain both turkey and ham meat. Conversely, the significantly higher mean belief score from the turkey-ham group respondents suggest that there respondents believe that turkey-ham contains both ham and turkey meat--a perception which does not correspond to the actual ingredient mix of the product. In other words, the "correct" answer for respondents being exposed to both product labels should be "1" on the seven-point scale. This pattern is much more characteristic of "ham" product group respondents than "turkey-ham" group respondents.

In summary, it appears that consumers are deceived and mislead by the "turkey-ham" wording on the label in both package size categories. Consumers are deceived in their beliefs about the "turkey and ham" attribute of the turkey-ham product. Consumers are not deceived in the beliefs about the "turkey and ham" attribute of the ham label product.

Attribute: "Turkey" Meat Only. The "ham" group respondents produced an extremely low mean belief score (X = 1.16) when asked if ". . . the food product in this package is made only from turkey meat." However, the low mean belief value produced by "turkey-ham" group respondents (X = 3.35)--which is significantly higher than the ham group belief values--suggests that consumers are deceived by their product label for this attribute.

The "correct" answer for the "ham" respondent group is, in fact, a "1" on the seven-point scale since the ham product does not contain any turkey meat. Conversely, the "correct" answer for the "turkey-ham" respondent groups is, in fact, a "7" on the seven-point scale since the turkey-ham product contains only turkey meat. In essence, the very low group beliefs for the ham group show consumers have a correct perception of reality. On the other hand, the higher belief mean for the turkey-ham group should have been much closer to the upper limit (i.e., a "7" on the scale). In other words, "turkey-ham" group respondents do not believe that turkey-ham is made only from turkey meat, when in fact, this is the only ingredient of the product.

In summary, it appears that consumers are deceived and mislead by the turkey-ham wording on the label in both package size categories. Consumers are deceived in their beliefs about the "turkey only" attribute of the turkey-ham product--based on their exposure to the label on the product. Consumers are not deceived in their beliefs about the "turkey only" attribute of the ham product--based on their exposure to the label of the product.

Consumer Attribute Beliefs Supported by Factual Evidence

Attribute: Nutrition. Based on their exposure to the package label, "ham" group respondents (X = 4.72) did not have a statistically significant higher belief than the "turkey-ham" group respondents (X = 4.66) that ". . . the food product in this package is relatively nutritious for people to eat."

Since the mean scores representing consumer beliefs (i.e., evolving from the label) about the nutritional value of the products based on exposure to the labeling are correct in terms of technical evidence, it is reasonable to conclude that consumers in both groups were not mislead or deceived by the product labels. In other words, both products are, in fact, relatively nutritious to eat and most consumer correctly perceive this fact.

In summary, it appears that consumers are not deceived in their beliefs about the "nutritional" value attribute of the two products--based on their exposure to ham and turkey-ham labels.

Attribute: Federal Government Safety Standards in Packaging. There is no statistically significant difference between beliefs held by ham group (X = 6.16) and turkey-ham group (X = 6.25) respondents on the attribute that ". . . the package for this food product meets all federal government safety standards for food packaging" based on their exposure to the package label.

Consumers in both groups hold equally strong and equivalent beliefs that the packaging technology--which represents federal government standards--results in "safe" (in terms of purity or contamination) products. Further, all belief data is "high" on the seven-point scale which is "correct" in a normative sense. In summary, consumers are not deceived in their beliefs about the "packaging safety" attribute of two products--based on their exposure to ham and turkey-ham labels.

Attribute: "Pork" Meat Only. The mean value (X = 4.00) of consumers exposed to the "ham" label was significantly higher than the mean value (X = 1.74) of the turkey-ham group respondents.

Since the "ham" products are made only from pork meat and respondent belief scores show recognition of this fact (i.e., a relatively "high" belief score) and since turkey-ham products are not made only from pork meat and respondent beliefs are consistent with this fact (i.e., with relatively low belief scores) it is reasonable to conclude that consumers in both groups were not mislead or deceived by the product labels. In essence, it appears consumers are not deceived in their beliefs about the "pork meat only" attribute of the two products--based on their exposure to ham and turkey-ham labels.

Attribute: Some "Turkey" Meat. The belief scores of the "turkey-ham" group respondents (X = 5.24) are significantly higher-than the belief scores for the "ham" group respondents (X = 1.63) for the attribute that ". . . the food product in this package contains some turkey meat."

The consumer belief data, in terms of the "turkey-ham" and "ham" group scores on the seven-point scale, reflects an acceptable pattern of results in terms of marketplace reality. In other words, respondents who were exposed to the turkey-ham product label tend to believe there is actually turkey meat in the product (i.e., producing a mean belief value in excess of "5" on the seven-point scale). Similarly, respondents who were exposed to the ham product label tend to believe that there is no turkey meat in the product (i.e., producing mean value near "1" on the seven-point scale). These data suggest that consumers are not deceived in their beliefs about the "turkey'' meat attribute of the two products--based on their exposure to ham and turkey-ham labels.

General Food Knowledge Data. Less than one percent (.7 percent) of the sample respondents felt that the term "ham" meant a "cut of any animal," while a majority (51.7 percent) felt the term "ham" referred to a "cut of pork meat" and another (28.5 percent) felt this term referred to a "cut of ham meat." Further, only 3.2 percent felt ham referred to the "smoked and cured thigh of any animal,'' while 11.2 percent said it was a "smoked and cured thigh of swine." These data suggest that the term "ham" refers to pork or ham meat and very few believe this term applies to a "cut from any animal" or the thigh portion of any animal.

For general comparative purposes, it should be noted that 78.6 percent of the respondents correctly associated the term "hot dog" or "frankfurter" with beef and/or pork products. Very few (.5 percent) of the respondents felt these products are made of poultry products. Similarly, 91.9 percent of the respondents correctly perceived that a "chicken hot dog" is made: (a) only of chicken, or (b) some combination of chicken and beef or pork.

Sociodemographic Data of Respondents. The following statements summarize the results from the sociodemographic questions included in the study. It should be noted that post-experimental analysis of these data indicated that there was no statistically significant difference between "ham" group and "turkey-ham" group respondents in terms of any of those characteristics.

1. About one-third of the sample (31.5 percent were between 25 to 34 years old, while 17.0 percent and 16.0 percent were between 35 to 44 years old and 45 to 54 years respectively. Persons in the under 25 year-old age group (15.5 percent), 55 to 64 year-old group (12.5 percent), and 65 and older (8.0 percent) accounted for smaller sub- segments in the sample.

2. One-half (50.0 percent) of the respondents came from families with annual incomes between $10,000 and $19,999. On the other hand, about one-quarter (23.5 percent) of the families in the sample had incomes in excess of $20,000 per year, while 20.8 percent of the sample families had incomes less than $10,000 per year.

3. Nearly all of the sample respondents (90.2 per- cent) were the head of their household (31.0 percent) or their spouse (59.2 percent). Nearly two-thirds (64.0 percent) of the respondents were female.

4. Almost one-half (49.0 percent) of the respondents were from families with 2 to 4 members, while one-quarter (25.0 percent) of the respondents were from two person families. Few of the respondents were from one person (9.7 percent) or 5 or more person families (16.0 percent).

5. Nearly one-half (49.0 percent) of the respondents were high school graduates and 21.2 per- cent had some college experience. About one- fifth (18.2 percent) held a college or advanced/ professional degree.

6. Over one-half (52.0 percent) of the sample were employed--on a full-time (33.5 percent) basis or as part-time (18.8 percent) employees.

7. About three-quarters (73.0 percent) of the respondents were white and 11.0 percent were non- white. Less than one-tenth (6.3 percent) were Spanish-speaking.

SUMMARY AND IMPLICATIONS

Data from this study suggest three summary findings. First, consumers incorrectly believe--after examining an actual product package and label--that "turkey-ham" products: (1) contain more than one type of meat, (2) contain both "turkey" and "ham" meat, and (3) do not contain only "turkey" meat. This causal relationship between exposure to the "turkey-ham" wording on the product label and these incorrect consumer beliefs suggest that this label is a source of "claim-fact" discrepancy deception. On the other hand, consumer beliefs about "turkey-ham" nutritional, packaging, pork meat content, and turkey meat content attributes appear to correctly reflect available scientific evidence. Second, consumers who were exposed to "ham" product label wording did not exhibit false beliefs about any of the seven product attributes measured in the study. Third, study data indicate that very few consumers correctly perceive the technical definition of "ham" used by the USDA. The implications of these findings are presented in terms of the two major study objectives.

Conceptual and Methodological Implications

Three major implications related to research in consumer deception are suggested by this study. First, the study results indicate--on a preliminary basis--that selected conceptual and methodological perspectives from the existing research tradition in advertising deception are applicable to the study of potentially deceptive label wording. In other words, falsely held consumer beliefs associated with specific product attributes represented in product label wording were identified by the study methodology.

Second, future research should extend the findings from this study in two ways. First, additional research on potentially deceptive label wording in other consumer product categories should be conducted to provide documentation of the present study results. Second, other forms of marketing stimuli which may represent sources of marketplace deception--sales catalog product and pricing descriptions, point of sale materials, consumer brochures, etc.--should be investigated. In this way, knowledge on the sources and impact of consumer deception will be broadened in a meaningful way.

Third, the present study illustrates that previous laboratory-based approaches to the study of deception can be applied to "real world" products and consumers in a field experiment context. This factor illustrates that increasing degrees of external validity in research on deception can be achieved by consumer researchers. As such, the credibility of consumer research input public policy decision-making on issues involving deception is strengthened. In stating these consumer research implications the authors recognize also that future research must overcome the major limitations of the present study (i.e., non-probability sampling processes were used at four subjectively selected interviewing locations; testing effects related to exposure to products in a single category; "one time" analysis vis-a-vis longitudinal analysis, etc.).

Public Policymaking Implications. Findings from this study suggest that USDA should review its policy position of allowing the "turkey-ham" wording on the label of products in this category since consumer beliefs about the ingredient or content attributes of these products are incorrect. Within the framework of existing policy alternatives at USDA, a post-market regulatory strategy--whereby this wording is removed from the label--would seem to be an appropriate remedy. Conversely, USDA labeling policies for "ham" products do not stimulate incorrect attribute beliefs, and as a result, it appears that no regulatory action affecting the wording on these labels is needed.

These two post-market policy implications are supported by two additional long run considerations. First, since many new combination or multi-ingredient meat products are introduced in the marketplace each year, USDA should consider developing a continuing and on-going program of consumer research in product labeling using: (1) experimental methodologies, and (2) belief variables. Such a research program would gather empirical data on product labeling (in all USDA jurisdictional product categories) in support of pre-market regulatory policies at USDA. The objective of such a program would be to reduce the likelihood of consumer deception emanating from product label wording by preventing such labeling from reaching the marketplace.

Second, since consumers appear to relate the term "ham" to pork and/or ham meat products, the USDA should review the appropriateness of its dictionary-based definition within the context existing consumer knowledge and terminology. The alternative policy approach at USDA--which does not seem to be cost-effective based on findings in this study--would involve a large-scale consumer education program designed to promote a consistent relationship between the public policy definition and consumer understanding of the term "ham." In any event-thc preceding discussion of post-market, pre-market, and consumer education implications emanating from this study highlight three potential roles consumer research can fulfill in public policymaking processes in agencies like the USDA.

REFERENCES

D. A. Aaker, "Deceptive Advertising," in D. A. Aaker and G. S. Day, eds., Consumerism: Search for the Consumer Interest, 2nd ed. (New York: The Free Press, 1974), 137-156.

G. M. Armstrong and J. P. McLennan, "The FTC and the Investigation and Regulation of Deception in Advertising," in Thomas V. Greer, ed., Increasing Marketing Productivity/Conceptual and Methodological Foundations of Marketing (Chicago: American Marketing Association, 1973), 430-434.

G. M. Armstrong and F. A. Russ, "Detecting Deception in Advertising," MSU Business Topics, 23, No. 2 (Spring, 1975), 21-31.

M. T. Brandt and I. L. Preston, "The Federal Trade Commission's Use of Evidence to Determine Deception," Journal of Marketing, 41(January, 1977), 54-62.

D. Cohen, "The Federal Trade Commission and the Regulation of Advertising in the Consumer Interests" Journal of Marketing, 33(January, 1969), 40-44.

D. Cohen, "Surrogate Indicators and Deception in Advertising,'' Journal of Marketing, 36(July, 1972), 10-15.

T. Dillon, "What is Deceptive Advertising?" Journal of Advertising Research, 13(October, 1973), 9-12.

R. F. Dyer and P. G. Kuehl, "The 'Corrective Advertising' Remedy of the FTC: An Experimental Approach," Journal of Marketing, 38(January, 1974), 48-54.

R. F. Dyer, "The Effectiveness of the FTC's 'Corrective Advertising' Policy: New Empirical Insights," in Sharon K. Zeigler, ed., Proceedings of the 1974 National Conference of the American Academy of Advertising (Boston: American Academy of Advertising, 1974), 166-177.

G. T. Ford, P. G. Kuehl and O. Reksten, "Classifying and Measuring Deceptive Advertising: An Experimental Approach,'' in Proceedings: 1975 Fall Educators' Conference of the American Marketing Association, Rochester, New York. (pages unassigned)

D. M. Gardner, "Deception in Advertising: A Conceptual Approach," Journal of Marketing, 39(January, 1975), 40-46. (Based on "Can Deceptive Advertising be Measured?'' Working Series in Marketing Research, No. 18. The Pennsylvania State University, College of Business Administration, June 1973. 22 pages.)

J. E. Haefner, "The Legal Versus the Behavioral Meaning of Deception," in M. Venkatesan, ed., Proceedings of the Association for Consumer Research, Fall, 1972, 356-360.

J. E. Haefner, M. T. Birchmore and S. E. Permut, "Applications of Belief Measures to Specific and Overall Advertising Claims: A Program for Research," in S. K. Zeigler, ed., Proceedings of the 1974 National Conference for University Professors of Advertising, American Academy of Advertising, 311-317.

J. E. Haefner and S. E. Permut, "An Approach to the Evaluation of Deception in Television Advertising," Journal of Advertising, 3, No. 4, 40-45.

H. Keith Hunt, "Source Effects, Message Effects, and General Effects in Counteradvertising," in M. Venkatesan, ed., Proceedings, 3rd Annual Conference, Association for Consumer Research, 1972, 370-381.

H. Keith Hunt, "Effects of Corrective Advertising," Journal of Advertising Research, 13(October, 1973), 15-24.

J. Jacoby and C. Small, "The FDA Approach to Defining Misleading Advertising," Journal of Marketing, 39(October, 1975), 65-68.

M. G. Jones, "The FTC's Need for Social Science Research," Proceedings, 2nd Annual Conference, Association for Consumer Research, 1971, 1-9.

H. H. Kassarjian, C. J. Carlson and P. E. Rosin, "A Corrective Advertising Study," in Mary Jane Schlinger, ed., Advances in Consumer Research, Vol. 2, 1975, 631-642.

P. G. Kuehl and R. F. Dyer, "Brand Belief Measures in Deceptive-Corrective Advertising: An Experimental Assessment," in Kenneth L. Bernhardt, ed., Proceedings of the 1976 American Marketing Association Fall Conference, 1976, 373-379.

P. G. Kuehl, "Applications of the 'Normative Belief' Technique for Measuring the Effectiveness of Deceptive and Corrective Advertisements," in William D. Perreault, Jr., Advances in Consumer Research, Vol. 4, 1977, 204-212.

Michael B. Mazis and Janice E. Adkinson, "An Experimental Evaluation of a Proposed Corrective Advertising Remedy," Journal of Marketing Research, XIII (May, 1976), 178-183.

J. A. Miller, D. A. Topel and R. E. Rust, "USDA Beef Grading: A Failure in Consumer Information?" Journal of Marketing, 40(January, 1976), 25-31.

J. T. Rosch, "Marketing Research and the Legal Requirements of Advertising," Journal of Marketing, 39(July, 1975), 69-79.

"United States of America Before Federal Trade Commission in the Matter of Warner-Lambert Company," Docket No. 8891, Final Order, December 9, 1975 (103 pages).

W. L. Wilkie, "Research on Counter and Corrective Advertising.'' Paper presented at the American Marketing Association Conference on Advertising and the Public Interest, Washington, D.C., May 9-11, 1973, 16 pages.

W. L. Wilkie and D. M. Gardner, "The Role of Marketing Research in Public Policy Decision-Making," Journal of Marketing, 38(January, 1974), 38-47.

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