Printer Friendly

Effects of consumption frequency on believability and attitudes toward alcohol warning labels.

INTRODUCTION

Recent testimony before a U.S. subcommittee presented alarming evidence regarding the link between alcohol consumption and abuse and birth defects, driving impairment, hypertension, liver disease, cancer, dangerous effects with other drugs, and addiction ("Alcohol Warning Labels" 1988). Enoch Gordis, Director of the National Institute on Alcohol Abuse and Alcoholism, indicated that fetal alcohol syndrome is the leading cause of mental retardation, 51 percent of traffic fatalities are alcohol related, and alcohol abuse can be a contributing factor in the development of oral cancer, liver disease, and high blood pressure (Gordis 1988). In response to these and other potential hazards of alcohol consumption and abuse, the U.S. Congress enacted the Alcohol Beverage Labeling Act of 1988, requiring that two specific health warnings appear on the labels of all imported or domestic beverages bottled on or after November 18, 1989, for sale and distribution in the United States (Alcohol Beverage Labeling Act of 1988). These two federally mandated warnings are:

GOVERNMENT WARNING: (1) According to the Surgeon General, women should not drink alcoholic beverages during pregnancy because of the risk of birth defects. (2) Consumption of alcoholic beverages impairs your ability to drive a car or operate machinery, and may cause health problems.

There have been arguments from state lawmakers and citizen groups, however, that these two warnings should be strengthened. They suggest three additional alcohol warning labels alerting individuals to (a) alcohol's effects on hypertension, liver disease, and cancer; (b) its risks in combination with other drugs; and (c) its addictive qualities as a drug (Ferguson 1990; Lipman 1990; The New York Times 1989; Prial 1988). All five alcohol warning labels (i.e., the two federally mandated and three suggested) were originally proposed in the Alcohol Beverage Labeling Act. However, only the birth defects and driving impairment warnings were required by the federal government ("Alcohol Warning Labels" 1988).

While extensive research has examined other product warning labels (e.g., cigarettes, toxic products, appliances, prescription drugs; cf., Beltramini 1988; Bettman, Payne, and Staelin 1986), research on the effectiveness of alcohol warning labels is virtually nonexistent. Two recent exceptions are studies of the effect of prior alcohol attitudes and beliefs on warning perceptions (Andrews, Netemeyer, and Durvasula 1990) and the manipulation of warning severity and transmission mode (Ducoffe 1990). However, research is needed to determine what other important moderating factors serve to influence acceptance of alcohol warning label information. For example, based on studies suggesting a differential impact of alcohol awareness campaigns depending on stages of alcohol use/addiction (Bozinoff, Roth, and May 1989; DePaulo, Rubin, and Milner 1987), alcohol consumption frequency is expected to serve as an important moderating variable in the acceptance of alcohol warning label information. In Bozinoff, Roth, and May (1989), alcohol awareness campaign recall, ad evaluation, and perceived ad effectiveness were all found to be higher for nonusers than heavy users of alcohol. This suggests that the impact of alcohol awareness information can be differentially affected by a person's degree of alcohol usage. Will similar results occur for the acceptance of alcohol warning label information under differing alcohol consumption rates? In particular, will the occasional/nonuser of alcohol exhibit greater belief in and more favorable attitudes toward alcohol warning labels than the regular/heavy user? Will the regular/heavy user of alcohol hold more confident (and negative) attitudes toward alcohol warning labels than the occasional/nonuser? This study seeks to provide answers to these questions.

Another important issue pertains to the degree to which certain affected segments of the population will believe and react favorably to the warning labels. One of the regulatory objectives of alcohol warning labels is to "increase consumer knowledge and have an impact on consumer behavior" regarding potential health risks of alcohol consumption and abuse ("Alcohol Warning Labels" 1988, 6). Thus, it would be instructive to examine the alcohol consumption-label acceptance relationships for those population segments who may benefit the most from the labels. For instance, the alcohol industry has been recently criticized for its promotion of alcoholic beverages to university students (cf., Lipman 1989; Seessel 1988). While young adults (e.g., university students) represent an attractive target market for the alcohol industry (Tenowitz 1988), considerable debate and controversy has ensued regarding the persistence of alcohol-related problems on campus (Lightner 1984; Magner 1988; Redfield 1984). Consequently, will university students (especially those with more frequent alcohol consumption) believe and react favorably to the alcohol warning labels?

In order to address these questions, this study examines the effects of alcohol consumption frequency on perceived believability of, attitudes, and attitude confidence toward the five proposed alcohol warning labels. As a potentially vulnerable segment, university students serve as the sample. This paper proceeds as follows. First, the theoretical background concerning processing of warning label information and role of consumption experience are examined. Then, research hypotheses, study overview, and results are presented. Finally, alternative explanations for findings and implications for public policy are discussed.

THEORETICAL BACKGROUND

Effectiveness of Warning Label Information

As indicated by Bettman, Payne, and Staelin (1986), recall of warning label risk depends on the frequency of information communicated, whether the information is dramatic, sensational, or heavily reported in the media, and the vividness and personal relevance of the information. Label processors may ignore or discount information that is perceived as providing few rewards or involving little risk or danger. Individuals may elect to use heuristics (i.e., simple decision rules) to evaluate warning labels when the decision becomes complex, or if the labels result in information overload. Therefore, gauging the appropriate level of impact of a warning label is a difficult challenge for public policy-makers. On one hand, it is hoped that the risk information has considerable impact on consumers. On the other hand, the level of risk should not be so strong as to provide a punitive indictment of the product (cf., Bettman, Payne, and Staelin 1986).

The effectiveness of warning label information is further complicated by whether or not the information is accepted (i.e., believed). Acceptance serves as an important step in information processing and persuasion (cf., McGuire 1969, 1980). As Fishbein and Ajzen indicate:

. . . with persuasive communications, the major problem is to ensure that the receiver accepts (i.e., believes) the communication [e.g., the warning] which attempts to link the object [e.g., consuming alcohol] and the attribute [e.g., driving impairment]. (1975, 389)

Additionally, Petty and Cacioppo (1986) suggested that message content designed to be relatively strong may prove to be unpersuasive because the content is considered unbelievable, strains credulity, or is perceived as implausible in the message context. However, credible sources (e.g., "According to the Surgeon General") and/or message personalization (e.g., "consuming alcohol impairs your ability to operate machinery") have been found to enhance message believability and persuasion (cf., Ajzen and Fishbein 1980).

In perhaps the first study to examine the perceived believability of warning label information, Beltramini (1988) found that cigarette warning labels indicating specific risk outcomes (e.g., lung cancer, heart disease, emphysema, fetal injury, premature birth) were significantly more believable than those labels suggesting remedial action (e.g., quitting smoking) or harmful contents (e.g., carbon monoxide). However, general believability and attitudes toward warning labels may also be influenced by the interaction between the label information and the consumer's experience or "evidence" found to confirm the warnings.

Alcohol Warning Label Information and Confirming Evidence

According to Deighton's (1984) hypothesis regarding the interaction of advertising and confirming evidence, exposure to alcohol warning label information is likely to arouse curiosity and expectations, and subsequently lead to an individual's disposition to confirm the warnings based on their experience. That is, warning labels do not operate on their own, but in relationship to the individual's own experience with the product. As suggested by Hoch and Ha (1986; also Ha and Hoch 1989), consumers form "tentative hypotheses" about products (e.g., alcohol) based on advertising (e.g., alcohol warning labels). These tentative hypotheses are tested through an individual's experiences and acquired evidence. If the evidence is ambiguous, advertising and other persuasive communications can influence product judgments. This is particularly true when people use top-down processing (i.e., they are asked to form overall impressions). However, if the evidence is unambiguous or if a bottom-up processing strategy is used (i.e., exhaustive examination of message details), message information (e.g., alcohol warning label) is suggested not to influence product judgments (Hoch and Ha 1986). In the case of alcohol warning labels, perceptions that the evidence is clear or unambiguous may increase for those with greater alcohol experience and usage. These perceptions of alcohol warning labels are also likely to be negative for frequent users of alcohol.

Role of Direct Consumption Experience

Consistent with Smith and Swingyard's (1982) Integrated Information Response Model, product beliefs and attitude confidence are found to be higher for subjects exposed to product sampling, as opposed to product advertising (Marks and Kamins 1988; Smith and Swinyard 1983). The rationale for the greater attitude confidence under product trial is that people rarely derogate themselves as a source. That is, an individual's consumption experience and usage are trusted to a greater extent than external sources of information (e.g., the government in the case of alcohol warning labels). This presents a problem for government officials trying to enhance the effectiveness of alcohol warning labels for frequent alcohol users who may disbelieve and have unfavorable (and confident) attitudes toward the labels. Moreover, in the case of cigarette warning labels, Bhalla and Lastovicka (1984) suggested that people may engage in schematic or scripted behavior and systematically ignore part of it, not because the information is irrelevant, but because it is already known or avoided. [1] This schematic or scripted behavior occurs because risk information (e.g., fetal alcohol syndrome, drinking and driving) may have been seen many times and details of its structure are already known. This information is likely to be at odds with the frequent user's attitudes and consumption behavior, thereby increasing the likelihood of defensive or schematic reactions. As Bozinoff, Roth, and May (1989) discovered, the effectiveness of alcohol awareness ad campaigns is significantly greater for nonusers than for heavy users of alcohol.

Label Attitude Confidence

Similar to arguments and theory presented by Smith and Swinyard (1982, 1983) on product trial, research conducted by Fazio and Zanna (1978, 1981) suggests that the confidence with which an attitude is held increases with an individual's direct experience. The reason is that the more behavioral experiences (e.g., consuming alcohol) an individual has with respect to a related attitude object (e.g., warnings regarding consuming alcohol), the greater will be the opportunity to assess and infer attitudes from the behavior. Attitudes toward the object also are expected to be more clearly focused and consistent with each additional behavioral experince associated with the object (Fazio and Zanna 1981).

In the present study, it is expected that those with more frequent alcohol consumption experience also will hold more confident attitudes toward warnings regarding alcohol consumption. However, these confidently held attitudes are expected to be negative, reflecting a defensiveness on the part of the frequent alcohol user when confronted with the dangers of alcohol consumption and abuse.

RESEARCH HYPOTHESES AND STUDY OVERVIEW

Based on research regarding the perceived believability and attitudes toward cigarette warning labels (Beltramini 1988; Bhalla and Lastovicka 1984), the role of trial experience and evidence in attitude confidence (Deighton 1984; Fazio and Zanna 1978, 1981; Ha and Hoch 1989; Hoch and Ha 1986; Marks and Kamins 1988; Smith and Swinyard 1983), and the impact of alcohol usage on the effectiveness of alcohol awareness campaigns (Bozinoff, Roth, and May 1989; DePaulo, Rubin, and Milner 1987), the following is expected to occur.

H1: Frequent users of alcohol will exhibit (a) less perceived believability, (b) more unfavorable attitudes, and (c) greater attitude confidence regarding alcohol warning label information than will occasional/nonusers of alcohol.

To aid in the subsequent discussion of these hypothses, they will be referred to as the "label believability hypothesis" (H1a), "label attitude hypothesis" (H1b), and "label attitude confidence hypothesis" (H1c), respectively.

Relative differences among five different alcohol warning labels are also examined for their influence on the dependent measures (i.e., label believability, attitude toward the label, and label attitude confidence). While multivariate analysis of variance (MANOVA) is used to determine the influence of consumption frequency (frequent versus occasional/nonuser) and warning label type (i.e., the five different alcohol warning labels) on the dependent measures, a priori contrasts of cell means provide the crucial tests of the hypotheses for

[TABULAR DATA OMITTED]

consumption frequency. Consumption frequency effects for the dependent measures are also examined for each warning label. Table 1 presents the five different alcohol warning labels that were randomly assigned to subjects in the study. The first two are the federally mandated labels, while the last three are those suggested by various state lawmakers and citizen groups.

Sample

Because the primary objective of alcohol warning labels is to inform individuals (especially vulnerable ones) of the health risks associated with alcohol consumption, university students represent an appropriate sample which might benefit from the alcohol warning labels (cf., Lightner 1984; Magner 1988). While students do not necessarily reflect the general population (thereby limiting the generalizability of results), they do represent a relevant sample because of their relatively high alcohol consumption. It is estimated that college students spend $2 to $3 billion annually on purchases of beer alone (Lipman 1989). Magner (1988) reported that 20 percent of college student drinkers can be classified as heavy/abusive drinkers (six or more drinks at one sitting more than once per week), a figure that has remained constant since 1985. (Note: Ninety-four percent of the sample were student drinkers. Therefore, approximately 19 percent of the total sample can be labeled as heavy/abusive drinkers, as defined by Magner (1988).) In contrast, nine percent of the general population are classified as heavy/abusive (i.e., problem) drinkers ("Hearings on Health Warnings in Alcohol Advertisements" 1990). Driving impairment, addiction, alcohol-related disease, birth defects, and drug combination warnings represent legitimate concerns for the university student population due to their relatively high level of consumption and abuse. Regarding the birth defect warning, the average age of the sample falls directly into the range of childbearing years for women. In fact, the National Council on Alcoholism has indicated that one in six women of childbearing age drink enough to pose a risk to an unborn child (Egan 1989).

The sample for the study consists of 273 undergraduate marketing students that provide a complete set of responses. Males represent 47 percent of the sample and females 53 percent. The mean age is 21.4 years (SD = 0.91) with a range from 19 to 27. Forty-seven percent of the sample are frequent users of alcohol (drank more than once a week), while 53 percent are occasional users (drank only once a week or less) or nonusers. All data were collected prior to November 18, 1989, the date on which the Alcohol Labeling Act of 1988 took effect.

Measures

Frequency of alcohol consumption is measured on a three-point scale with response options of "frequently" (more than once a week), "occasionally" (once a week or less), and "never" (nonuser). As relatively few students are classified as nonusers (six percent), the nonuser and occasional user categories are combined.

The perceived believability of each label is measured with ten, seven-point items based on Beltramin's (1988) examination of the believability of cigarette warning labels. The ten items are believable/unbelievable, trustworthy/untrustworthy, convincing/unconvincing, credible/not credible, reasonable/unreasonable, honest/dishonest, unquestionable/questionable, conclusive/inconclusive, authentic/not authentic, and likely/unlikely (coefficient [alpha] - .91). Respondents' attitude toward the label is measured by the summation of three, seven-point items. These items are favorable/unfavorable, good/bad, and positive/negative (coefficient [alpha] = .97). Finally, consistent with measures used in attitude confidence research (cf., Fazio and Zanna 1978), each person's confidence level in their attitude held toward the label is measured on a seven-point scale ranging from "not confident in my response" to "confident in my response."

[TABULAR DATA OMITTED]

RESULTS

A multivariate analysis of variance (MANOVA) is used to examine the overall influence of consumption frequency (frequent versus occasional/nonuser) and warning label type on label believability, attitude toward the label, and attitude confidence. Because label believability, attitude toward the label, and attitude confidence measures are all theoretically correlated, MANOVA is the appropriate technique for analysis (cf., Winer 1971).

Warning Label Type

The multivariate F for warning label type [2] indicates a significant overall effect of the different warning labels on the dependent variables (Wilks' A = .81; F = 4.64; p < .001). As Table 2 indicates, the univariate reults from the MANOVA reveal that both label believability (F = 11.27, df 4, 263; p < .001) and attitude toward the label (F = 6.12, df 4, 263; p < .001) are significantly different across the five different warning labels. Label attitude confidence does not differ among the five warning label types. Nondirectional, Student-Newman-Keuls (SNK) pairwise comparisons reveal that the birth defects and driving impairment warning labels are perceived as significantly more believable than the other three labels. Regarding attitude toward the label, the birth defects label is viewed as significantly more favorable than all other labels. However, it should be mentioned that while relative differences are detected among the five different warning labels, all are above the scale midpoint for label believability (Md = 40) and attitude toward the label (Md = 12).

Consumption Frequency

The multivariate F for consumption frequency reveals a marginally significant overall effect of alcohol usage on the dependent variables (Wilks' A = .97; F = 2.59; p < .055). The univariate results indicate marginally significant effects for label believability (F = 3.68, df 1, 263; p < .056), and label attitude (F = 2.70, df 1, 263; p < .101), and a significant effect for label attitude confidence (F = 5.78, df 1, 263; p < .017) between frequent and occasional/nonusers of alcohol (Table 2). A priori contrasts between frequent and occasional/nonusers for each dependent variable provide the tests of directional hypotheses. Hypotheses tests are calculated for average responses across the five warning labels (Table 2, "all labels").

Across labels, the results in Table 2 indicate that label believability is greater for occasional/nonusers than for frequent users, providing support for the label believability hypothesis (H1a). Regarding specific labels, the birth defects and driving impairment labels are found to be significantly more believable for occasional/nonusers as opposed to frequent users. Consistent with predictions of the label attitude hypothesis (H1b), attitude toward the label is found to be significantly more favorable for occasional/nonusers versus frequent users. In particular, attitude toward the driving impairment label is found to be significantly more favorable for the occasional/nonuser than for the frequent user of alcohol. Contrary to expectations in the label attitude confidence hypothesis (H1c), occasional/nonusers are found to be significantly more confident in their label attitudes than frequent users of alcohol. For specific labels, occasional/nonusers exhibit significantly more confident attitudes toward the hypertension and drug combination labels than frequent users of alcohol.

DISCUSSION

To begin, the findings should be viewed with caution because the student sample does not necessarily reflect the general population. Therefore, one should be careful in generalizing the results of this study to the population at large. The sample is nonetheless relevant because of the relatively high levels of alcohol consumption and abuse among college students (Manger 1988). In this respect, the study produces several results of interest.

Findings for the differential impact of the five alcohol warning labels on label believability indicate that the birth defects and driving impairment warnings are perceived as significantly more believable than the other warnings. In addition, the birth defects warning is found to be significantly more favorable than all other warnings, including driving impairment. The results for the driving impairment warning are interesting in that while students believed the driving impairment risks to a greater extent than many other risks, they did not like being told these risks. However, it should be mentioned tat while relative differences are found across the five different alcohol warning labels, each label's believability and attitude scores are above the scale midpoint for these measures. This latter finding, in conjunction with research from other samples, should be of importance to those advocating the use of all five alcohol warning labels in a rotating manner on packaging and advertising (cf., Ferguson 1990; Gordis 1988; Lipman 1990; The New York Times 1989). In fact, recent nationwide surveys have favored placing alcohol warnings on both containers and in advertisements (Freedman 1989).

Regarding the effects of alcohol consumption frequency, this research indicates that frequent alcohol users perceive the alcohol warning labels to be significantly less believable and are less favorable toward the labels than occasional/nonusers of alcohol, as expected. Contrary to expectations, occasional/nonusers of alcohol are more confident in their attitudes toward the label than frequent users. There are two possible explanations for this unexpected finding. First, from a practical viewpoint, it is possible that occasional/nonusers may possess considerable experience with alcohol, but in a different sense than usage frequency. That is, important vicarious experiences with alcohol (e.g., friends' alcohol-related driving accidents, family medical and psychological problems due to alcohol, media emphasis on alcohol problems) may lead to relatively confident and favorable attitudes toward alcohol warning labels.

A second likelihood is based on the tenets of cognitive dissonance theory (Festinger 1957). According to Festinger, when people read a counter-aatitudinal message, dissonance can occur due to the inconsistency between one's beliefs (e.g., as a frequent user of alcohol) and the fact that they are reading the counter-attitudinal message (e.g., an alcohol warning label). As a result of this dissonance, people are believed to engage in rationalized thinking to help reduce the inconsistency between one's beliefs and the discrepant message. This justification/rationalization process may have attenuated the strength of the frequent alcohol user's attitude toward the warning labels.

While frequent users' attitude confidence toward the warning labels is not as strong as occasional/nonusers', their relatively less believable and less favorable attitudes toward the labels may suggest a defensiveness on the part of frequent alcohol users. If so, this implies that alcohol warning label information may be partially ignored or discounted by those who presumably need the warnings the most. This problem parallels that of cigarette warning labels and may be one reason why they have had little market impact (cf., Pollay 1989).

The results also point to the importance of research examining alcohol warning label effects for those in different stages of alcohol use and/or abuse (Bozinoff, Roth, and May 1989; DePaulo, Rubin, and Milner 1987). For example, will processing of warning labels by dependent alcohol users create greater anxiety and lead to greater abuse? Because addictive behavior is commonly characterized as irrational behavior (Falk, Dews, and Schuster 1983), will alcohol warning labels have any effect on the heavy/addictive user? While research into stages of alcohol usage and addiction will help answer these questions, the impact of alcohol warning labels on the different stages of information processing (e.g., exposure, attention, comprehension, yielding/acceptance, and retention) should also be considered. As indicated by Shimp and Dyer, ". . . investigation of the advertising-drug use link would benefit greatly by focusing on intermediate processing stages (e.g., comprehension; agreement; McGuire 1976). . ." (1979, 42).

Future Research

While the research investigated the influence of alcohol consumption frequency and warning label type on measures of the acceptance of and agreement with alcohol warning labels, the effects of other moderating variables and repreentative populations should also be explored. For example, the relative amount of consumption (as opposed to consumption frequency) may exhibit a similar pattern of results as found in this study. Those in different stages of alcohol abuse or those with related alcohol experiences (e.g., alcohol-related problems with friends or family) may reveal additional insight into the persuasive impact of alcohol warning label information. In addition, the impact of alcohol warning labels on other vulnerable population groups (e.g., pregnant women, older addicts, etc.) should be investigated. Another issue is whether public service advertisements could be effectively combined with the labels to further explain and elaborate on the warning information. Such information could enhance the understanding and believability of the health warnings in the labels. Many research opportunities also exist for examining the effects of different formats (e.g. label size, design, contrast) and manipulations of alcohol warning label information (e.g., source credibility, message personalization, degree of risk). However, consumer researchers and governmental agencies should make a concerted effort to investigate the impact of these alcohol warning label modifications on stages of information processing and persuasion.

(1) A "schema" represents an organized structure of knowledge about an object based on experience (Tesser 1978). A "script" is a type of schema and is defined as "a predetermined, stereotyped sequence of actions that define a well-known situation" (Schank and Abelson 1977, 41). When a message processor is confronted with incoming information, prior "scripts," written when information was once new, may be stereotypically acted upon (cf., Bhalla and Lastovicka 1984).

(2) The warning label type results are further explored in Andrews, Netemeyer, and Durvasula (1990). In their study, prior attitudes and beliefs toward drinking (as covariates) are examined for their differential impact on the warning labels. It is necessary to present the warning label main effect results in this study in the examination of possible interaction effects with consumption frequency.

REFERENCES

Ajzen, Icek and Martin Fishbein (1988), Understanding Attitudes and Predicting Social Behavior, Englewood Cliffs, NJ: Prentice-Hall, Inc.

Alcohol Beverage Labeling Act of 1988, Public Law 100-690, 27 USC 201-211 (November 18).

"Alcohol Warning Labels" (1988), Hearings before the Subcommittee on the Consumer, Senate Commerce, Science and Transportation Committee (August 10), Washington DC: U.S. Government Printing Office.

Andrews, J. Craig, Richard G. Netemeyer, and Srinivas Durvasula (1990), "Believability and Attitudes Toward Alcohol Warning Label information: The Role of Persuasive Communitions Theory," Journal of Public Policy & Marketing, 9: 1-15.

Beltramini, Richard F. (1988), "Perceived Believability of Warning Label Information Presented in Cigarette Advertising," Journal of Advertising, 17(1): 26-32.

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

Bhalla, Gaurav and John L. Lastovicka (1984), "The Impact of Changing Cigarette Warning Message Content and Format," in Advances in Consumer Research, Volume XI, Thomas Kinnear (ed.), Provo, UT: Association of Consumer Research: 305-310.

Bozinoff, Lorne, Victor Roth, and Colin May (1989), "Stages of Involvement with Drugs and Alcohol: Analysis of Effects of Drugs and Alcohol Abuse Advertising," in Advances in Consumer Research, Volume XVI, Thomas K. Srull (ed.), Provo UT: Association for Consumer Research: 215-220.

Deighton, John (1984), "The Interaction of Advertising and Evidence," Journal of Consumer Research, 11(December): 763-770.

DePaulo, Peter J., Mary Rubin, and Brenton Milner (1987), "Stages of Involvement with Alcohol and Heroin: Analysis of the Effects of Marketing on Addiction," in Advances in Consumer Research, Volume XIV, Melanie Wallendorf and Paul Anderson (eds.), Provo, UT: Association for Consumer Research: 521-525.

Ducoffe, Sandra J. (1990), "The Impact of Product Usage Warnings in Alcoholic Beverage Advertisements," Journal of Public Policy & Marketing, 9: 16-29.

Egan, Timothy (1989), "A Worried Liquor Industry Readies for Birth-Defect Suit," New York Times (April 21): A1.

Falk, John P., Peter B. Dews, and Charles R. Schuster (1983), "Commonalities in the Environmental Control of Behavior," in Commonalities in Substance Abuse and Habitual Behavior, Peter K. Levison, Dean R. Gerstein, and Deborah R. Maloff (eds.), Lexington, MA: Lexington Books.

Fazio, Russell H. and Mark P. Zanna (1987), "On the Predictive Validity of Attitudes: The Role of Direct Experience and Confidence," Journal of Personality, 46: 228-243.

Fazio, Russell H. and Mark P. Zanna (1981), "Direct Experience and Attitude-Behavior Consistency," in Advances in Experimental Social Psychology, Vo. 14, L. Berkowitz (ed.), New York: Academic Press: 161-202.

Ferguson, Tim W. (199), "Prop. 65 Populism Is on the Prowl, but Don't Cry Wolf," The Wall Street Journal (March 6): A19.

Festinger, Leon (1957), A Theory of Cognitive Dissonance, Stanford, CA: Stanford University Press.

Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research, Reading, MA: Addison-Wesley.

Freedman, Alix M. (1989), "Rebelling Against Alcohol, Tobacco Ads: Consumers Back Curbing Pitches, Not Products," The Wall Street Journal (November 14): B1, B11.

Gordis, Enoch (1988), Director, National Institute on Alcohol Abuse and Alcoholism, testimony on "Alcohol Warning Labels" before the Subcommittee on the Consumer, Senate Commerce, Science, and Transportation Committee, August 10, Washington, DC: U.S. Government Printing Office: 24-26.

Ha, Young-Won and Stephen J. Hoch (1989), "Ambiguity, Processing Strategy, and Advertising-Evidence Interactions," Journal of Consumer Research, 16(December): 354-360.

"Hearings on Health Warnings in Alcohol Advertisements" (1990), Subcommittee on Transportation and Hazardous Materials, U.S. House of Representatives Committee on Energy and Commerce, July 18, Washington, DC.

Hoch, Stephen J. and Young-Won Ha (1986), "Consumer Learning: Advertising and the Ambiguity of Product Experience," Journal of Consumer Research, 13(September): 221-233.

Lightner, Candy (1984), "Does Marketing of Alcoholic Beverages on Campuses Encourage Abuse? Yes--Causes Drunk Driving, Academic Failure," Los Angeles Times (May 6): V3.

Lipman, Joanne (1989), "Alcohol Firms Retain Campus Presence," The Wall Street Journal (October 6); B3.

Lipman, Joanne (1990), "Alcohol Marketers Are Fuming over Bill to Put Warnings in Ads," The Wall Street Journal (April 5): B8.

Magner, Denise K. (1988), "Alcohol-Related Problems Have Not Decreased on Most College Campuses, Survey Indicates," The Chronicle of Higher Education (November 9): A35, A37.

Marks, Lawrence J. and Michael A. Kamins (1988), "The Use of Product Sampling and Advertising: Effects of Sequence of Exposure and Degree of Advertising Claim Exaggeration on Consumers' Belief Strength, Belief Confidence, and Attitudes," Journal of Marketing Research, 25(August): 266-281.

McGuire, William J. (1969), "The Nature of Attitudes and Attitude Change," in The Handbook of Social Psychology, Volume 3, 2nd edition, G. Lindzey and Elliot Aronson (eds.), Reading, MA: Addison-Wesley: 136-314.

McGuire, William J. (1976), "Some Internal Psychological Factors Influencing Consumer Choice," Journal of Consumer Research, 2(March): 302-319.

McGuire, William J. (1980), "The Communication-Persuasion Model and Health Risk Labeling," in Product Labelling and Health Risks: Banbury Report 6, Louis A. Morris, Michael B. Mazis, and Ivan Barofsky (eds.), Cold Spring Harbor, NY: Cold Spring Harbor Laboratory: 99-122.

The New York Times (1989), "Alcohol Warning: Impact is Debated" (November 15): C1, C6.

Petty, Richard E. and John T. Cacioppo (1986), Communication and Persuasion: Central and Peripheral Routes to Attitude Change, New York: Springer-Verlag.

Pollay, Richard W. (1989), "Filters, Flavors ... Flim-Flam, Too! on 'Health Information' and Policy Implications in Cigarette Advertising,' Journal of Public Policy & Marketing, 8: 30-39.

Prial, Frank J. (1988), "California Considers Cancer Warning on Alcohol," The New York Times (May 25): C12.

Redfield, David (1984), "Does Marketing of Alcoholic Beverages on Campuses Encourage Abuse? No--Ban Might Actually Increase Drinking," Los Angeles Times (May 6): V3.

Schank, Roger A. and Robert P. Abelson (1977), Scripts, Plans, Goals, and Understanding, Hillsdale, NJ: Lawrence Erlbaum.

Seesel, Thomas V. (1988), "Two Ways to Reduce Underage Drinking," The New York Times (January 1): 16N, 30L.

Shimp, Terence A. and Robert F. Dyer (1979), "The Pain-Pill-Pleasure Model and Illicit Drug Consumption," Journal of Consumer Research, 6(June): 36-46.

Smith, Robert E. and William R. Swinyard (1982), "Information Response Models: An Integrated Approach," Journal of Marketing, 46(Winter): 81-93.

Smith, Robert E. and William R. Swinyard (1983), "Attitude-Behavior Consistency: The Impact of Product Trial Versus Advertising," Journal of Marketing Research, 20(August): 257-267.

Tenowitz, Ira (1988), "Miller Looks for Lite Touch to Tap Youth," Advertising Age (August 1): 3, 50.

Tesser, Abraham (1987), "Self-Generated Attitude Change," in Advances in Experimental Social Psychology, Volume 11, L. Berkowitz (ed.), New York: Academic Press: 289-338.

Winer, B. J. (1971), Statistical Principles in Experimental Design, New York: McGraw-Hill.

J. Craig Andrews is Assistant Professor of Marketing, Marquette University, Milwaukee, WI; Richard G. Netemeyer is Assistant Professor of Marketing, Louisiana State University, Baton Rouge, LA; and Srinivas Durvasula is Assistant Professor of Marketing, Marquette University, Milwaukee, WI.
COPYRIGHT 1991 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Andrews, J. Craig; Netemeyer, Richard G.; Durvasula, Srinivas
Publication:Journal of Consumer Affairs
Date:Dec 22, 1991
Words:5341
Previous Article:Dynamics of complaint management in the service organization.
Next Article:Consumer information and education effects on knowledge and choice of fire resistant upholstery.
Topics:


Related Articles
Alcohol warning labels receive complaints from Congress.
Two studies show that health warnings are seen, but the effects are unknown.
What Is Moderate Drinking?
BATF Opens Up for Comments Health-Related Alcohol Label Statements.
The Center for the Advancement of Health claims that alcohol abuse costs $250 billion.
The Price of Alcohol, Wife Abuse, and Husband Abuse.
ATF rejects label change.
What am I drinking? The effects of serving facts information on alcohol beverage containers.

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