Printer Friendly

Preliminary assessment of changes in labels required by the Nutrition Labeling and Education Act of 1990.

Since food labels first became required on certain consumer packaged food products 18 years ago, researchers have been concerned with consumers' perceptions and use of nutrition information on food labels (Asam and Bucklin 1973; Brown, Kelley, and Lee 1991; Daly 1976; Cole and Gaeth 1990; Lenahan et al. 1973; Levy et al. 1985; Moorman 1990). Results concerning the effectiveness of disclosure of nutrition information has often been equivocal; for example, in summarizing results from six studies Jacoby, Chestnut, and Silberman concluded that "the vast majority of consumers neither use nor comprehend nutrition information in arriving at good purchase decisions" (1977, 126). Additional concerns arose in 1987 when the Food and Drug Administration (FDA) lifted its ban on the use of health claims on food labels which led to some exaggerated use of terms as "high fiber," "low fat," "low sodium," and "lite." Summarizing current nutrition labeling conditions, U.S. Health and Human Services Secretary Louis Sullivan referred to the supermarket as "a Tower of Babel |where~ consumers need to be linguists, scientists, and mind readers" (Consumer Reports 1990, 326).

Such concerns led to the passage of the Nutrition Labeling and Education Act of 1990. While this Act does not specify the structural format required for presenting nutrition information, it identifies several changes concerning nutrition information that should be available on a label. A primary objective of the Act is to permit consumers to make more informed decisions about food purchases. The government has estimated that improvements in the quality of consumers' diets could result in $100 billion in reduced health care costs over 20 years (Ingersoll 1991).

The primary purpose of this study is to provide a preliminary comparison of consumer attitudes and perceptions between nutrition labels consistent with nutritional requirements in the Nutrition Labeling and Education Act and a label consistent with regulatory requirements prior to the NLEA. Two alternative labels that comply with the 1990 amendment were tested against the type of label in use at the time of the study. These three types of nutrition labels were tested across conditions where the nutritional value provided was varied along with inclusion or exclusion of a "warning" that focused on potential risks associated with poor nutrition.


Overview of the Nutrition Labeling and Education Act of 1990

The Nutrition Labeling and Education Act of 1990 (NLEA) contains a number of important provisions for consumers and food marketers. These provisions include the specification of minimum nutrition information that must be available on all processed foods, delineation of authority of the FDA, federal preemption of state food labeling and health claims jurisdictions, and the establishment of operational definitions of frequently used terms such as "lite," "low fat," and "reduced calories" (Mueller 1991).

The law allows claims of the presence (absence) of any nutrient as relating to disease or health-related conditions only when "there is significant scientific agreement, among experts qualified by scientific training and experience to evaluate such claims, that the claim is supported by such evidence" (Best 1991; Nutrition Labeling and Education Act 1990, 6). Also, legitimacy of claims about specific nutrients will be contingent upon the presence or absence of other specified nutrients. For example, under most conditions a "no cholesterol" content claim will not be allowed if the food in question contains a level of saturated fat determined to increase the risk of a disease or health-related condition. Such regulations will require changes in the on-package claims for health benefits by many food manufacturers (Colford 1991).

Of particular interest in this study are the provisions for the nutrition information required on a label found in section 2 of the Act. It specifies that labels must include (1) serving sizes (expressed in common household measures), (2) number of servings per container, (3) total number of calories (derived from all sources), (4) calories derived from total fat, and (5) amounts of each of the following nutrients in each serving: total fat, saturated fat, cholesterol, sodium, total carbohydrates, complex carbohydrates, sugars, dietary fiber, and total protein.

Relative to prior nutrition information requirements, key changes necessitated by the Act include the provision of information on the calories derived from fat, amount of saturated fat and cholesterol, breakdown between complex carbohydrates and sugars, and amount of dietary fiber. Thus, the law requires several types of additional information on a label that may interest consumers (Consumer Reports 1990).

It has been estimated that the new labeling requirements for processed packaged foods will affect up to 17,000 firms and 70 percent of the total dollar expenditures for grocery products in the more than $300 billion industry (Ingersoll 1991; Williams 1992). The industry will spend an estimated $2.8 billion as a result of changes in labels and associated expenses. Only companies with annual gross sales of less than $500,000 or food sales under $50,000 are exempt from the regulations. Product packages that are too small to provide required nutrition information are exempt if the label provides no nutrition information.

The nutrition labeling section of the NLEA also includes some broad guidelines concerning consumer education. It states activities will be performed that "educate consumers about (1) the availability of nutrition information in the label or labeling of food, and (2) the importance of the information in maintaining healthy dietary practices" (NLEA 1990, 5).

Research Issues

While the NLEA suggests a number of directions for empirical research, this study addresses four research questions. The focus of this study concerns inclusion of additional nutrition information required by the Act. A principal question pertains to whether inclusion of this information on a label is sufficient to affect consumer perceptions of product nutrition and product purchase likelihood. Much of the additional information concerns "negative" nutrients that consumers want to avoid (e.g., calories derived from fat, amount of saturated fat and cholesterol). Inclusion of such negative nutrition information may lower nutrition attitudes and perceptions across levels of nutritional value. However, any consequences of nutrition label changes depend on an array of variables including overall concern with nutrition, perceived importance of added nutrient data, and comprehension of and motivation to use the information (Jacoby, Chestnut, and Silberman 1977; Moorman 1990). Thus, it remains unclear whether changing a label by requiring specific additional information significantly affects consumers' attitudes and perceptions.

The impact of nutritional value (i.e., amounts of fat, cholesterol, sodium) provided on the label is also assessed. Nutritional value is examined for two primary reasons. First, it permits an assessment of whether the impact of added nutrition information required by the NLEA varies across conditions where nutritional value is relatively favorable versus relatively unfavorable (i.e., in addition to a main effect of label type, do manipulations of type of label and nutritional value interact?). As suggested by prospect theory, reactions to a "loss" may differ from reactions to a "gain" (Tversky and Kahneman 1981). Therefore, because negative nutrition information (e.g., high levels of cholesterol and calories from fat) may be perceived as more important than positive information (Heimbach and Stokes 1982; Russo et al. 1986), as nutritional value becomes more unfavorable, any impact relating to inclusion of additional information may become stronger. Second, a manipulation of nutritional value permits a comparison between the amount of variance in dependent variables that can be explained by the type of nutrition label and the actual nutritional value.

Inclusion (exclusion) of a nutrition "warning" that reminds consumers of potential health hazards associated with consumption of large amounts of fats, cholesterol, and sodium is also examined. Government warnings have long been in effect for cigarettes and have recently been instituted for alcohol. While it is unclear how the education provisions of the NLEA will be fulfilled, a nutrition "warning" is consistent with the NLEA goal of communicating to consumers "the importance of that |label~ information in maintaining healthy dietary practices" (NLEA 1990, 5).

Educational opportunities offered by the label are also supported by research conducted by the Roper Organization. They found the label to be the most frequently used source of nutrition information. Fifty-two percent of consumers reported use of food packages as a source of nutrition information compared to 24 percent for newspapers and magazines (Mueller 1991). A "warning" on a nutrition label can serve a dual function. First, it seems capable of informing consumers with minimal knowledge of nutrition that some level of risk exists (Bettman, Payne, and Staelin 1986). Second, for more knowledgeable consumers it may increase accessibility and potential relevance of information of which consumers are aware but may not actively consider at the point of purchase without the availability of a specific cue, such as a warning (Tulving and Pearlstone 1966).

A final question concerns consumers' opinions about the importance of various information that could be included on a label (Heimbach and Stokes 1982; Stokes 1980). Given the provision in the NLEA to add (or potentially delete) information, it seems beneficial to understand overall perceptions of importance and how such perceptions vary across demographic groups. From a nutrition education standpoint, any discrepancies between these perceptions and what is suggested by nutrition research are highly relevant.


Experimental Manipulations and Stimulus

Subjects responded to initial questions pertaining to their attitude about nutrition labeling and then were shown a simulated side panel of a package of macaroni and cheese which contained nutrition information. Macaroni and cheese was chosen for two primary reasons. First, macaroni and cheese was believed to be a relatively widely consumed packaged food product; this belief was supported by 85 percent of the sample that reported eating macaroni and cheese in a prior six-month period. Second, the percentage of calories from fat from the actual product was very close but slightly below the daily maximum level suggested by government guidelines, thus making the product a reasonable "baseline" for manipulation of fat content and percentage of calories derived from fat.

The type of nutrition label, the nutritional values provided, and the presence or absence of a warning label were manipulated on the side panel in a 3 x 3 x 2 between subjects experimental design. Each is discussed in turn.

Type of nutrition label

Subjects were exposed to one of three different nutrition label types which are summarized in the top portion of Table 1. The first type of label was consistent with the presentation of nutrition information as required by law at the time of the study (i.e., before the requirements resulting from the NLEA went into effect). The information presented was identical to that on the side panel of a national brand of macaroni and cheese on the market at the time. This condition may be viewed as a control against which changes in nutrition label type may be assessed. The second label condition was consistent with that required by the NLEA of 1990. The third label, also consistent with the Act, was based on a recommendation by the Consumers Union (Consumer Reports 1990, 327).

Nutritional values

The amounts per serving of fat, sodium, cholesterol, and calories derived from fat were manipulated at "poor," "moderate," and "good" levels. The moderate condition corresponded to the actual level in one serving of the macaroni and cheese product used as the baseline for the experiment. Nutrient levels for poor, moderate, and good conditions are shown in the bottom of Table 1.

The percentage of calories from fat used in the moderate nutrition condition (28 percent) was slightly below the daily recommended maximum percentage of calories that should be obtained from fat (i.e., 30 percent; Dietary Guidelines for Americans 1990, 16). In the "good" condition, the amount of fat was one-half of the fat per serving in the moderate condition. The amount of fat in the "poor" condition was TABULAR DATA OMITTED almost twice that of the moderate condition. Sodium and cholesterol content were set near zero in the good condition. In the poor condition, they were set at levels approximately two and one-half times the moderate condition (i.e., level of actual product).(1)

The warning

One-half of the labels contained an explicit warning pertaining to consumption of fat, cholesterol, and sodium; the other half had no warning. Information in the warning concerning specific health risks was drawn from that in the government brochure Dietary Guidelines for Americans (1990). The warning read as follows:

WARNING: U.S. government nutritional guidelines indicate that consumption of high levels of fat, saturated fat, and cholesterol may be associated with an increased risk of heart disease. High levels of sodium are related to high blood pressure, a factor associated with increased heart disease risk.

Results from a test sample offered support for the warning in terms of believability, comprehensibility, moderate level of fear evoked, and attitude toward the warning.(2)

Study Sample and Measures

The study used a convenience sample of 374 nonstudent subjects 18 years of age or older. Only subjects who reported they were primary grocery shoppers for their households were included in the sample. Subjects meeting these criteria were given a packet that included a cover letter, the simulated label containing the experimental manipulations, and survey questions pertaining to the dependent variables, perceived importance of various nutrients, and demographics. Due to missing values, 343 subjects were used in tests of the manipulations. Cell sizes ranged from 15 to 24. Median age of the sample was 41 years, median education level was some college, 71 percent of respondents were female, and median household income was $34,000. These age, income, and education medians were similar to those of the metropolitan area.

Dependent variables included measures of attitude toward nutrition content, overall perceptions of product nutritiousness, and product purchase likelihood. For the attitude toward nutrition measure, three seven-point scales with endpoints of unfavorable-favorable, bad-good, and negative-positive were used in response to the following item: "Based on the specific nutrition information shown in the label, what is your attitude about the nutrition content of the product?" For this three-item scale coefficient alpha was .98.

Subjects' overall perceptions of the nutritiousness of the product were measured with the following item: "Overall, how would you rate the level of nutritiousness suggested by the information on the label?" Endpoints for this item were "not nutritious at all" (coded as '1') and "very nutritious" (coded as '7'). The purchase likelihood measure assessed subjects' perceptions about their likelihood of purchasing the product given information on the label ("Would you be more likely or less likely to purchase the product, given the information shown on the label?"). Endpoints were "more likely" and "less likely" (reverse coded).

Manipulation check

To check if manipulation of nutritional value affected respondent perceptions about amounts of cholesterol, fat, and sodium and percentage of calories from fat, subjects rated the level of each of the four components on a scale from low to high. A MANOVA was run in which the four rating scales assessing amounts of cholesterol, fat, sodium, and calories from fat were used as dependent variables with the nutrition manipulation as the independent variable. The overall MANOVA was significant (Wilks' |Lambda~ = 0.73, F = 15.1, p |is less than~ .01). Univariate tests were significant for each of the four dependent variables with F-values ranging from 6.9 to 62.8. Followup contrasts indicated significant differences between moderate and good nutritional values for each of the four dependent variables, but differences between the moderate and poor conditions were relatively small and not significant.


Manipulations of type of nutrition label, nutritional value, and presence (absence) of the warning were tested with a multivariate analysis of variance (MANOVA). Cell means and standard deviations for the dependent variables are reported in Table 2. MANOVA and univariate analysis of variance results are shown in Table 3.



The bottom portion of Table 3 indicates that all two-way and three-way interactions among the manipulated variables were nonsignificant. Significant multivariate main effects were found for type of nutrition label (Wilks' |Lambda~ = .95, F = 2.8, p |is less than~ .01, ||Eta~.sup.2~ = .03) and the nutritional value manipulations (Wilks' |Lambda~ = .95, F = 3.1, p |is less than~ .01, ||Eta~.sup.2~ = .03), but the overall main effect of the warning was not statistically significant (Wilks' |Lambda~ = 1.0, F = 0.0).(3)

Results Pertaining to Type of Label

The F-values associated with the followup univariate tests show type of nutrition label manipulation had a significant effect on the perception of product nutritiousness (F = 6.2, p |is less than~ .01, ||Eta~.sup.2~ = .04) and product purchase likelihood (F = 3.6, p |is less than~ .05, ||Eta~.sup.2~ = .04) and a marginal effect on attitude toward nutrition content (F = 2.7, p |is less than~ .10, ||Eta~.sup.2~ = .02). Contrasts between the type of nutrition label conditions indicate more favorable attitudes toward nutrition content, perceptions of nutritiousness, and product purchase likelihood for the nutrition label type in effect at the time of the study (i.e., control condition) than for the NLEA and the Consumer Reports conditions, both are consistent with the requirements set by the Act. Specifically, for perception of product nutritiousness the contrast between the pre-NLEA (control) condition and the pooled data for the other two conditions consistent with the new labeling requirements was significant (t = 2.9, p |is less than~ .01). Also, the contrast between the control and the NLEA condition was significant (t = 2.5, p |is less than~ .025), and the control condition was significantly higher than the type of label proposed by Consumer Reports (t = 2.5, p |is less than~ .025). The difference between the NLEA and Consumer Reports label types were small and nonsignificant.

Contrasts pertaining to the product purchase likelihood dependent variable produced generally similar results. Mean purchase likelihood for the control label type was significantly higher than the pooled mean for the NLEA and Consumer Reports label type (t = 2.3, p |is less than~ .025). The difference between the control and the Consumer Reports label was significant (t = 2.5, p |is less than~ .025), but the difference between the control and NLEA, while in the same direction, did not reach significance (t = 1.4).

For nutrition attitude, the difference between the control and the other pooled treatment conditions was marginally significant (t = 1.9, p |is less than~ .06). The control group mean was greater than the NLEA condition (t = 2.1, p |is less than~ .05). The t-value associated with the difference between the control and Consumer Reports condition was nonsignificant (t = 1.2). Consistent with findings for the other dependent variables, differences in means between the NLEA and Consumer Reports labels were not significant. In sum, these results indicate that the type of nutrition label recommended by the NLEA can significantly lower attitudes and perceptions of nutrition and product purchase likelihood and thus seem to have important implications.

Results Concerning Nutritional Value and Warning

The univariate main effect of nutritional value was significant (p |is less than~ .01) for the dependent variable of attitude toward nutrition content (||Eta~.sup.2~ = .03), nutrition perception (||Eta~.sup.2~ = .04), and product purchase likelihood (||Eta~.sup.2~ = .04). For each dependent variable, contrasts between each of the nutritional value conditions revealed significant differences (i.e., contrasts between both good and moderate levels and moderate and poor levels were significant at p |is less than~ .005). The univariate results in Table 3 and means in Table 2 show that the inclusion of a warning on the label had virtually no effect on the dependent variables.

Perceptions of Importance of Specific Nutrition Information

While the NLEA is specific in recommendations for inclusion of certain nutrition information, it leaves great latitude in determining if additional information should be included (NLEA 1990, 2). To assess what information was perceived as important and thus most desirable on labels, subjects rated the importance of inclusion of 17 separate types of nutrition information on labels. Table 4 shows the results. Amounts of total fat, calories, and cholesterol were perceived of greatest importance followed by amounts of saturated fat, calories from fat, and sodium. Percentages (e.g., percentage of calories from fat, percentage of saturated versus unsaturated fat) were viewed of TABULAR DATA OMITTED less importance than absolute amounts. These importance ratings were generally consistent with changes specified in the NLEA of 1990; for example, inclusion of information about cholesterol, amount of saturated fat, and calories from fat were viewed as very important. However, somewhat surprisingly, the amount of fiber (which is also required by the NLEA) was not viewed as critical by subjects.

Demographic variables of sex, age, and household income significantly affected perceptions of importance. MANOVA results indicated a significant effect of sex on the ratings (Wilks' |Lambda~ = .90, F = 2.27, p |is less than~ .01). Women viewed inclusion of all 17 items as more important than men. The univariate F-values in Table 4 indicate significant differences for 14 of the 17 ratings. Age also had a significant multivariate effect (Wilks' |Lambda~ = .86, F = 1.6, p |is less than~ .05); there was a monotonic relationship in which older subjects viewed various types of nutrition information as more important than younger subjects. Followup univariate tests revealed significant results for nine of the 17 items. Household income also affected ratings (Wilks' |Lambda~ = .80, F = 1.43, p |is less than~ .05). As shown in Table 4, followup univariate tests indicated a significant effect for ten of the items. While higher income households (above $40,000) had higher ratings than lower income households, perceptions of importance generally peaked between $40,000 and $60,000. Interestingly, education had little effect with only one of 17 importance ratings significantly impacted.


The primary purpose of this paper was to assess some potential consequences of changes in nutrition labels indicated by the NLEA of 1990. Specifically, the concern was whether changes in nutrition labels impact variables such as attitudes and perceptions of product nutritiousness and self-reported purchase likelihood. Results indicate additional information required can affect nutrition perceptions and product purchase likelihood. Nutrition labels that included the additional information (i.e., cholesterol, amount of saturated fat, calories from fat, and fiber) required by the NLEA led to lower nutrition perceptions and likelihood of purchase than nutrition labels in use at the time of the study. Thus, while this study stopped short of gauging actual purchase behavior, results suggest the new nutrition information requirements of the NLEA at least have potential to impact consumer perceptions of nutritiousness. In addition, results concerning the importance of inclusion of various nutrition information reinforce consumers' interest in (and presumably desire to use) newly required information such as cholesterol, saturated fat, and calories from fat.

These results have some potentially important implications for consumers, consumer policy advocates, and food product marketers. Results appear positive for consumers and policy advocates; the NLEA seems to require nutrition information that consumers generally desire, is useful in determining perceptions of nutritiousness, and could be used as one factor in product purchase decisions. Mandated changes thus appear useful to comparison shoppers concerned with nutrition. From the food marketers' perspective, results suggest potential negative effects for products with unfavorable nutritional values. These results suggest NLEA provisions concerning inclusion of additional nutrition information and health claims and terminology may encourage some brands to examine more nutritious alternative formulations.

The study also tested for differences between two alternative labels that were consistent with the NLEA guidelines. One label, drawn from a recommendation by Consumer Reports, went beyond NLEA specifications by including additional information (e.g., fat composition) as well as providing graphical summaries of some nutrition information. Results showed no significant differences between this enhanced label and the more straightforward alternative, suggesting benefits associated with pictorial representation of nutrition information may not exceed associated costs.

Because consumers may weigh negative information more heavily than positive information, there was interest in whether or not the label type or warning would interact with the nutritional values provided. These two-way interactions were not significant, but, as might be expected, nutritional value manipulation significantly impacted attitudes and perceptions of nutrition and product purchase likelihood.

While these main effects of nutritional value seem of little surprise, for these data the relative amounts of variance explained by label type and nutritional value manipulation appear interesting. While neither manipulation explained a large amount of variance in dependent variables, the multivariate ||Eta~.sup.2~ values indicate type of nutrition label explained almost as much variance as nutritional value. These findings reinforce the importance of the specific nutrition information required by law.

Perhaps the failure of nutritional value to have stronger effects relative to type of label is less of a surprise when taken in conjunction with findings concerning manipulation checks of nutritional value. While large differences across conditions in levels of fat, calories from fat, cholesterol, and sodium produced significant differences in perceptions of the amount of each of these in the product, differences between the "moderate" and "poor" conditions were small and nonsignificant. This result suggests the need for additional educational measures pertaining to what constitutes highly unfavorable amounts of nutrients. Such results also may argue for labels that classify fat, cholesterol, and other nutrition components as "low," "medium," or "high" based on government guidelines, in addition to providing absolute amounts (Hellmich 1991; McGuire 1980).

Relevant to this concern, this study compared a label prototype that included a "warning" about levels of fat, saturated fat, cholesterol, and sodium to one that did not. The warning had no effect on nutrition attitudes and perceptions or product purchase likelihood. These results may be due to the general nature of the warning tested. An attempt was made to create a warning that was general enough to be consistent with nutrition research findings and government guidelines (Dietary Guidelines for Americans 1990) and was believable and comprehensible. Failure to find main effects or interactions concerning the warning suggests that if warnings are to have some impact, they may need to be more specific in recommendations concerning nutrition.

Limitations and Future Research

Several factors relate to the generalizability of the results and thus specific conclusions and implications which should be drawn. The study did not assess purchase behavior; thus, whether any effects extend beyond attitudes, perceptions, and purchase likelihood are unclear. A mock nutrition label was presented to subjects as part of a survey packet and responses were generated in a nonstore environment. Effects of the wide variety of factors (e.g., brand name, price, packaging, time constraints) that influence perceptions and behavior in the in-store environment were not assessed. Research may extend these findings by assessing effects of manipulations on actual purchase behavior and other dependent variables across multiple product categories in more naturalistic field settings (McGuire 1980).

There are also limitations pertaining to the generalizability of effects of the specific manipulations. For example, two labels that are consistent with NLEA recommendations and a control condition consistent with regulations at the time of the study were tested. While no differences were found in effects between the NLEA-consistent labels, and as the new law offers great flexibility in terms of final implementation, there could be significant differences among other (untested) label types that also satisfy NLEA requirements. Similarly, the results of this study may not extend to different manipulations of nutritional values or alternative "warnings" that could be included on a label. Because of the concern with potential effects associated with different nutritional values and not with generalizing results to specific macaroni and cheese products, the "good" and "poor" manipulations of nutritional value did not correspond to products on the market and probably were not plausible from a nutritional composition perspective. Generalizability of results also is limited by the nonprobability sample.

Beyond research designed to extend the generalizability of these results, a variety of applied and theoretical research issues are suggested by the NLEA. For example, studies could examine effects associated with use of "larger type, bold type, or contrasting color" (NLEA 1990, 1), the manner in which information from various types of labels is processed by different demographic segments, and effectiveness of other educational methods such as in-store displays that focus on government nutrition guidelines. Results of this study indicate the inclusion of additional information on a nutrition label, as recommended by the NLEA, can affect nutrition perceptions and product purchase likelihood. Hopefully, these findings may serve as a springboard for research motivated by the NLEA.

1 For the "good" and "poor" conditions, the amount and percentage of calories from carbohydrates were altered so that fat content could be manipulated without impacting total calories per serving. The protein amount was held constant across conditions. Nutrient amounts in the "good" and "poor" conditions were designed with the purpose of being sufficiently large to produce some differences in overall perceptions of nutritiousness and did not correspond to actual products on the market.

2 A small sample (n = 29) of subjects was presented with the warning and then asked to respond to multi-item seven-point scales about believability, comprehensibility, level of fear, and attitude toward the warning. Results indicated that the label was perceived as believable and easy to comprehend (average item scores = 5.6 and 5.7, respectively). An intermediate level of fear evoked by a warning is generally recommended for effective persuasion (McGuire 1980). Results indicated that the level of fear evoked by the warning was moderate (mean item score = 3.5). These test results were viewed as supportive of the warning used.

3 Because the nutrition manipulation check revealed moderate and poor nutritional value conditions were not perceived as significantly different from each other, the data were also analyzed by collapsing these two conditions. The results were similar to those reported.


Asam, Edward H. and Louis P. Bucklin (1973), "Nutrition Labeling for Canned Goods: A Study of Consumer Response," Journal of Marketing, 37(April): 32-37.

Best, Daniel (1991), "New Labeling Law Thrusts at Regulatory Standards," Prepared Foods (January): 50-52.

Bettman, James R., John W. Payne, and Richard Staelin (1986), "Cognitive Considerations in Designing Effective Labels for Presenting Risk Information," Journal of Public Policy and Marketing, 5: 1-28.

Brown, Vernon, Craig Kelley, and Ming Tung Lee (1991), "The State of the Art in Labeling Research Revisited: Developments in Labeling Research 1978-1990," in 1991 American Marketing Association Educators' Proceedings, M. C. Gilly (ed.), Chicago: American Marketing Association: 717-725.

Cole, Catherine A. and Gary J. Gaeth (1990), "Cognitive and Age-related Differences in the Ability to Use Nutritional Information in a Complex Environment," Journal of Marketing Research, 37(May): 175-184.

Colford, Steven W. (1991), "FDA, FTC May Team Up To Tackle Label, Ad Claims," Advertising Age (May 27): 3, 35.

Consumer Reports (1990), "What's Wrong With This Label?" 55(May): 326-327.

Daly, Patricia A. (1976), "The Response of Consumers to Nutritional Labeling," The Journal of Consumer Affairs, 10(Winter): 170-178.

Dietary Guidelines for Americans (1990), U.S. Department of Agriculture; U.S. Department of Health and Human Services.

Heimbach, James T. and Raymond C. Stokes (1982), "Nutrition Labeling and Public Health: Survey of American Institute of Nutrition Members, Food Industry, and Consumers," American Journal of Clinical Nutrition, 36: 700-708.

Hellmich, Nanci (1991), "Adjectival Food Label Gets High Marks," USA Today (June 24): D1.

Ingersoll, Bruce (1991), "U.S. Is Ready to Limit What Food Processors Can Claim on Labels," The Wall Street Journal (November 6): A1.

Jacoby, Jacob, Richard W. Chestnut, and William Silberman (1977), "Consumer Use and Comprehension of Nutrition Information," Journal of Consumer Research, 4(September): 119-128.

Lenahan, R. J., J. A. Thomas, D. A. Taylor, D. L. Call, and D. I. Padberg (1973), "Consumer Reaction to Nutritional Labels on Food Products," The Journal of Consumer Affairs, 7: 1-12.

Levy, Alan S. Odonna Matthews, Marilyn Stephenson, Janet E. Tenney, and Raymond E. Shuker (1985), "The Impact of a Nutrition Information Program on Food Purchases," Journal of Public Policy and Marketing, 4: 1-13.

McGuire, William J. (1980), "The Communication-Persuasion Model and Health Risk Labeling," in Product Labeling and Health Risks: Banbury Report 6, L. Morris, M. Mazis, and I. Barofsky (eds.), Cold Springs Harbor, NY: Cold Springs Harbor Laboratory: 99-122.

Moorman, Christine (1990), "The Effects of Stimulus and Consumer Characteristics on the Utilization of Nutrition Information," Journal of Consumer Research, 17(December): 362-374.

Mueller, William (1991), "Who Reads the Label?" American Demographics (January): 36-41.

Nutrition Labeling and Education Act of 1990, Public Law 101-535, 21 USC 301 (November 8).

Russo, J. Edward, Richard Staelin, Catherine A. Nolan, Gary J. Russell, and Barbara L. Metcalf (1986), "Nutrition Information in the Supermarket," Journal of Consumer Research, 13(June): 48-70.

Stokes, Raymond C. (1980), "Consumer Attitudes Toward Product Labeling," in Product Labeling and Health Risks: Banbury Report 6, L. Morris, M. Mazis, and I. Barofsky (eds.), Cold Springs Harbor, NY: Cold Springs Harbor Laboratory: 79-88.

Tulving, Endel and Zena Pearlstone (1966), "Availability Versus Accessibility of Information in Memory for Words," Journal of Verbal Learning and Verbal Behavior, 5: 381-391.

Tversky, Amos and Daniel Kahneman (1981), "The Framing of Decisions and the Psychology of Choice," Science, 211(January): 453-458.

Williams, Richard, Industry Activities Section, Food and Drug Administration (1992), personal telephone communication (February 5).

Scot Burton is Associate Professor and Wal-Mart Chairholder, Marketing, University of Arkansas, Fayetteville; and Abhijit Biswas is Assistant Professor, Marketing, Louisiana State University, Baton Rouge.

The authors express their thanks to Craig Andrews and two JCA reviewers for their helpful comments on previous versions of this article and to the College of Business Administration at Louisiana State University for its support.
COPYRIGHT 1993 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Burton, Scot; Biswas, Abhijit
Publication:Journal of Consumer Affairs
Date:Jun 22, 1993
Previous Article:Perceived time pressure and recommended dietary practices: the moderating effect of knowledge of nutrition.
Next Article:Media advocacy: a case study of Philip Sokolof's cholesterol awareness campaigns.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters