Consumer Research Standards &Amp; Public Policy Formulation: the Case of Mickey Mouse &Amp; Old Joe

ABSTRACT - In December, 1991 a consumer research paper "Brand Logo Recognition by Children Aged 3 to 6 Years, Mickey Mouse and Old Joe the Camel," produced a crescendo of demands to make significant changes in public policy, including banning the "Old Joe" advertising campaign. This paper reviews the demands for public policy change and offers a summary of evaluations by four independent reviewers of the study addressing issues of reliability, validity, reporting of convergent results and support for conclusions in the data offered. Serious concern is raised as to its efficacy to underpin major changes in public policy.

Citation:

Claude R. Martin, Jr. (1994) ,"Consumer Research Standards &Amp; Public Policy Formulation: the Case of Mickey Mouse &Amp; Old Joe", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 380-386.

Advances in Consumer Research Volume 21, 1994      Pages 380-386

CONSUMER RESEARCH STANDARDS & PUBLIC POLICY FORMULATION: THE CASE OF MICKEY MOUSE & OLD JOE

Claude R. Martin, Jr., University of Michigan

ABSTRACT -

In December, 1991 a consumer research paper "Brand Logo Recognition by Children Aged 3 to 6 Years, Mickey Mouse and Old Joe the Camel," produced a crescendo of demands to make significant changes in public policy, including banning the "Old Joe" advertising campaign. This paper reviews the demands for public policy change and offers a summary of evaluations by four independent reviewers of the study addressing issues of reliability, validity, reporting of convergent results and support for conclusions in the data offered. Serious concern is raised as to its efficacy to underpin major changes in public policy.

INTRODUCTION

In December, 1991 a consumer research paper produced a crescendo of demands to make significant changes in advertising regulation, including the banning of one campaign specifically. The article by Fischer, et al. entitled "Brand Logo Recognition by Children Aged 3 to 6 Years, Mickey Mouse and Old Joe the Camel," appeared in the Journal of the American Medical Association [JAMA]. In the year following the publication of the articles there was extensive media coverage of calls for banning the ad campaign, including 72 individual wire service stories. While accompanied in the same issue of JAMA by two other research articles and three editorials, the Fischer, et al. research was the most cited in demands for a significant change in public policy regarding advertising.

This unusual response to consumer research provoked our interest in reviewing the study on four dimensions: reliability, validity, reporting of convergent results and support for conclusions in the data offered. The objective is to explore whether this research stands the test for changing public policy.

DEMANDS FOR PUBLIC POLICY CHANGES

Shortly after the JAMA publication the American Cancer Society, American Heart Association and American Lung Association formally petitioned the Federal Trade Commission to immediately ban the Old Joe campaign [Cimons 1992, Dagnoli 1991, Snider 1991]; Senator William Cohen called for the senate to be involved in an effort to ban Old Joe Camel [Scripps Howard News Service, 1992]; Representative Henry Waxman asked the House Subcommittee on Health and the Environment to recommend strict controls on all advertising and specifically to prohibit the RJR ad campaign [Waxman, 1991]; Advertising Age [January, 1992] called upon RJR to drop the Old Joe campaign; the attorneys general of 26 states began to lobby for more control over cigarette advertising calling for repeal of a section of the Cigarette Labeling and Advertising act [Zukin 1992 and Levine 1992]; the New York Metropolitan Transportation Authority, citing the research in JAMA, banned all tobacco ads on subways, buses and trains beginning in 1993 [Harrigan 1992]; The Surgeon General of the United States, the Secretary of the Department of Health and Human Services, and the American Medical Association demanded that R.J. Reynolds stop using Old Joe [Washington Times 1992; Brown 1991, Enrice 1992, Horovitz 1992, Fara Warner 1992, Vesey 1992, U.S. News & World Report 1992, Lipman 1992, Kong 1992, Roberts 1992, Standora 1992]; a resolution was passed by a committee of the Chicago City Council condemning R. J. Reynolds Tobacco Company for the Old Joe ads [Chicago Tribune 1992]; there was a parade by physicians and their families through the streets of Chicago to protest Old Joe and calling for a ban on the ad campaign [Reuters News Service 1992, Associated Press 1992]; and the Surgeon General Antonia Novello in collaboration with the American Medical Association launched a nationwide contest among elementary and junior high school students calling for letters, essays, poems or cartoons focused on a theme of "Say No, Old Joe" [Detroit Free Press 1993].

It is difficult to quantify the news coverage and subsequent demands for public policy reformulation, including banning Old Joe. We reviewed 172 newspaper articles and 71 wire service stories all of which reported on the Fischer, et al. research. To give a flavor for the reporting the following sampling of headlines is offered:

Puffing Camel Rivals Mickey in Kid's Eyes

The Charlotte Observer [12/11/91]

Joe Camel Is Also Pied Piper, Research Finds

The Wall Street Journal [12/11/91]

Most Kids Can Identify Camel Character

The Associated Press [12/11/91]

Study: Camel Cartoon Sends Kids Smoke Signals

The Boston Herald [12/11/91]

To Some, Cartoon Camel Isn't Funny

Seattle Post-Intelligencer [12/11/91]

Ban: Health Groups Say Study Proves All Tobacco Ads Must Be Crushed Out

The Detroit News [12/12/91]

We also reviewed 82 print media editorials and 23 syndicated columnists. Of these editorial efforts there were 32 specific editorial calls for the banning of Old Joe and/or a more stringent regulation of tobacco advertising including the Boston Globe [December 12, 1991], Atlanta Journal and Constitution [December 14, 1991], Fort Worth Star-Telegram [December 17, 1991], Syracuse Herald-Journal [March 11, 1992], The News & Observer (Raleigh, N.C.) [March 12, 1992], Denver Post [March 15, 1992], Philadelphia Inquirer [March 15, 1992], San Francisco Examiner [March 17, 1992], and the New York Observer [March 23, 1992].

THE EVALUATION PROCESS

A census of research associations and journals was undertaken to ascertain their research guidelines. These included the fields of statistics, advertising, marketing, economics, sociology and psychology. Input was generated from the American Statistical Association, American Association for Public Opinion Research, American Economic Review, Journal of Marketing Research, Journal of Marketing, American Marketing Association, and American Psychological Association. Also examined were the "instructions for authors" of the Journal of the American Medical Association. This produced a compendium of standards upon which to judge social science/business research:

TABLE

While there are no formal research guidelines for Association for Consumer Research, we perceive that there would be general acceptance of these by most ACR members.

The Fischer, et al. article was submitted to four researchers in varying social science fields (marketing, advertising research, consumer behavior and corporate strategy) from differing major universities for their independent reviews. They were not provided with the research standards as discussed above, but rather were asked to independently evaluate each of the three papers on the basis of the usual norms for their disciplines. All four are experienced in analytical techniques and are reviewers for journals in their selected fields. The net result were evaluative comments, both general and specific, concerning the Fischer, et al. research which were then correlated to these standards.

The elements on which the reviewers' focused were: reliability, validity, full disclosure, and support for conclusions in the data offered.

FISCHER, ET AL., DESCRIPTION

Fischer, et al. investigated "brand logo recognition by children aged 3 to 6 years." Children were instructed to match logos with one of 12 products pictured on a game board. Twenty-two logos were tested, including those representing children's products, adult products, and those for two popular cigarette brands (Camel and Marlboro). The setting for the research were preschools in Augusta and Atlanta, Georgia. A convenience sample of 229 children attending 10 preschools was used. Fischer, et al., claim the children demonstrated high rates of logo recognition and when analyzed by product category, the level of cigarette logos was intermediate between children's and adult products. The recognition of the Disney Channel logo and Old Joe were described as "highest in their respective product categories" [Fischer, et al., p. 3145]. The respective categories are "children's brands" (the Disney logo) and "cigarette brands" ( Old Joe). The data offered by Fischer, et al.shows hat the Disney Channel, McDonald's, Burger King, Dominos Pizza, Coca Cola, Pepsi, Nike, Chevrolet and Ford all had higher recognition scores than Old Joe. This is not discussed in the research findings by the authors.

INDEPENDENT EVALUATIONS

Marketing Management Reviewer

Overall Evaluation. This paper leads with an advocacy position and then seeks to prove that position, rather than scientifically examining the relation between advertising and smoking behavior. The result is a plethora of citations from anti-smoking advocates and a failure to consider research that has addressed the advertising-smoking relationship and found countervailing evidence. The authors go far beyond their basic methodology, inferring advertising recognition to a simple matching exercise involving children. They ignore the data that shows improvement in cognitive skills to correctly match logos to products among older children in the study.

Specific Comments. The authors cite a series of studies [Aitken, et al. 1987, 1988, 1990; Armstrong, et al. 1990; Chapman and Fitzgerald 1982; Davis 1987; Goldstein, et al. 1987; Levit, et al. 1981; Seldon and Dordoodian 1989; Warner 1987], and then state "Collectively these studies provide compelling evidence that cigarette advertisements are seen by adolescents ........cigarette advertising is causally linked to smoking behavior" [Fischer, et al. p. 3145]. In a recent review of these studies the conclusion is reached that "such studies have not examined the causal link between advertising message recall and smoking behavior" [Mazis, et al. 1992, p. 24]. The advocacy position of the Fischer research team is clear from their conclusion that there is "compelling evidence."

Are-reading of the Leckenby and Plummer article [1983] cited by the Fischer team shows advertising recognition is a complex, multi-variable measurement that is used, among others, to evaluate copy and as a surrogate measure for advertising effectiveness. That is not what they proposed. The methodology simply allows them to have subjects match certain symbols (logos) with product pictures. This is not advertising recognition, but simply a matching exercise. The Old Joe logo could be very difficult to match to any of the other product categories and the matching may simply reflect the cognitive ability of the participants to match through a process of elimination. Controls for such a possibility are necessary.

The data concerning non-matches needs to reported. In how many instances did participants mismatch the other logos to cigarettes and how many times did they mismatch other cigarette logos to other non-cigarette products? In this vein, how many times were multiple cards incorrectly assigned to the same product? This information is not reported.

In their conclusion, Fischer, et al. [1991] state that "R J Reynolds Tobacco Company is as effective as the Disney Channel in reaching six year old children." The authors go far beyond their data and what they studied. The data simply show that six year old children were better than younger children at the task of matching logos to products generally, and that they matched Old Joe to cigarettes as a product at a rate not significantly different than they matched the Disney Channel to Mickey Mouse. The authors fail to show whether there are other matches that are also not significantly different, thus reinforcing the conclusion that six year olds have developed a better cognitive skill for matching.

Advertising Reviewer

Overall Evaluation. This article is very demand ridden. In other words, it is biased to get the results the authors' desire. It was set up not to fail. A classic example follows:

The task is demanding in that the Camel ads always show the camel with cigarettes. In the study the cigarette is not present. You are really just asking the subjects "what is missing from this picture?" So if the subjects correctly identify the cigarette, they are identifying what is missing from the picture, a task right out of kindergarten curricula. It is an obvious violation of experimental design protocol to have the stimulus material and the dependent measure be parts of the same image, separated only for the purpose of this study. This makes this study nothing more than a classical conditioning or S-R (stimulus-response) replication. I'm not entirely sure, but I think some higher order primates such as chimpanzees would have done as well. I'm only partially facetious on this.

Specific Comments. Essentially the authors take 3 to 6 year old children, expose them to 22 different brand "logos" and ask them to match them to pictures of 12 different product categories. The basic problems are:

a. The choice set is relatively small. One only has twelve choices. To be more externally valid, one would need more alternative choices, as would be true in a real world setting or information environment. It is also true that give so few choices, it is likely that the children employed some heuristic, choice rules and elimination procedures which may have systematically biased the results. But, given the lack of controls, etc. we simply don't know.

b. To be valid the study should have included not only more alternatives, but more varied ones as well. This would allow us to know whether alternative explanations are likely. For example, what if the authors would have included a picture of a zoo? If subjects would have then matched Old Joe with the zoo, what would these results have meant? There are so many more credible alternative explanations for the data than the authors tested or even acknowledge. I'd suggest they could have used other cartoon characters such as Green Giant, Pillsbury Dough Boy, Trix Rabbit, Hamburger Helper, etc. A child may recognize the Green Giant at a fairly early age; may even like him, but that doesn't mean he will eat his green beans, broccoli or asparagus.

c. It is possible that subjects are merely responding to context. What is the study would have included elements that also appear in camel ads such as pool tables, guitars, automobiles, etc.? My guess is that you would have high matching rates there as well. Such a finding would support the idea of the results being due to nothing more than "recognizing things often found with Old Joe," which means very little beyond some vague context effect.

The authors also play a little loose with their third hand summaries of other's research. They state that "by age six years, half of all children regularly go shopping by themselves" [Fischer, et al. p. 3147]. The attribution for this is to a book by J.U. McNeal [1987, p. 33]. This is not an empirical finding. The authors cite and refer to it as if it were research by McNeal supportive of their conclusion. The fact is that it is not; they simply misrepresent it. What is its face validity? The last time you were in K-Mart, Walmart or Target did you notice a lot of kids six and under shopping by themselves?

Consumer Behavior Reviewer

This reviewer did not follow the general format of the others, but rather concentrated on three "major conclusions" of the Fischer, et al. paper.

Conclusion #1. "Children's knowledge of cigarette brands logos is likely the result of their exposure to environmental tobacco advertising..." [Fischer, et al. p. 3148].

a. "Knowledge" is measured by a widely used measure of memory - the recognition test. This measure is believed to capture the most basic form of memory [Krugman, 1982] that need not lead to future behavior. For example, children recognizing the no-smoking symbol would not logically be lead to later smoking cigarettes!

B. On the issue of reliability there is a concern that the recognition task used by Fischer, et al.[1991] to measure knowledge of logo/trade characters is subject to extensive guessing [Mizerski 1982]. Several aspects of the method used in this study would prompt levels of guessing or bias far beyond the levels (8.3%) assumed by the Fischer, et al. [1991, p.3146]. The following are a few major factors:

1. "Each subject's parent signed a parental consent form and completed a short questionnaire about...the use of cigarettes in the home" [p.3146]. It appears that the only product asked about was cigarette use. The very next day, the children were tested! There is a strong likelihood that the topic of cigarettes was discussed with at least some of the children, which would affect the recognition task the following day. This problem is often referred as a "demand artifact" and can impose a serious bias.

2. A "don't know" option was not offered. This has been shown to bias/inflate recognition scores [Mizerski 1982].

C. Remembering a trade character does not necessarily translate to liking or preference. Indeed, the negative aspects of smoking would be expected to have been exposed to these children, and in a more extensive and organized fashion (e.g., in school) as the child ages. Matching a negative symbol (picture of a cigarette - perhaps similar to a no smoking sign) to Old Joe may actually reflect the impact of no smoking education. This was found in two studies [Aitken, et al. 1986, 1987] that are cited in the Fischer, et al. article.

D. The use of the McNemar Test, to test for "a significant change in correct responses between two logos (the Disney Channel and Old Joe) [Fischer, et al., p. 3146] does not appear to be the appropriate statistic. Siegel [1956, p. 63] notes this statistic is used to measure changes in response to the same stimulus. The McNemar Test would not be appropriate for use to test for differences between two stimuli (logo vs. trade characters) that are quite different.

Conclusion #2. "It has been shown that children prefer brands that they see advertised. This effect has been shown to even influence their preference of products that they are too young to use, such as lipstick and diet soft drinks " [Fischer, et al. p. 3147].

This study simply does not provide primary data for this point. They cite previously published studies and imply that high recognition scores transmit into preference for brands-and more importantly-product category use.

a. Recognition, even if carefully measured, is not a surrogate for preference. Preference is best measured by seeing the choice of a user [See: Zajonc and Markus 1982]. There is some research that uses affect as a measure of preference, but few, if any, would suggest a very rudimentary measure of memory (recognition) is able to assess preference.

b. There is a citation [Goldberg, et al. 1978] that is used as support for the statement, "it has been shown that children prefer brands that they see advertised" [Fischer, et al. p. 3147]. Goldberg, et al. [1978] used a television advertisement in their research, this study did not, it used a logo. Furthermore, an alternative of a tennis ball, which was not advertised, was presented to the children as preferred by their mothers. The tennis ball tended to be the most preferred option by children despite exposure to a a televised toy ad. The Goldberg, et al. [1978] research amply demonstrated that even forced exposure to a TV ad for a toy is overcome by parental influence. My conclusion is that the citation cannot be used to support the Fischer, et al. [1991] conclusion.

c. The remaining part of the second conclusion, "this effect has been shown to even influence their preference of products that they are too young to use, such as lipstick and diet soft drinks," relies on a citation from Gorn and Florsheim [1985]. That study failed to find any effect of a diet soft drink TV commercial on preference - although Fischer, et al. [1991] state it did. Furthermore, the Gorn and Florsheim [1985] study used forced exposure to a television commercial in its methodology, not the different "logo" study offered by Fischer, et al. [1991].

Conclusion #3. "The R J Reynolds Tobacco Company is as effective as the Disney Channel reaching 6 year old children" [Fischer, et al. p. 3148].

a. The definition of "effective" and "reaching" is not clear. Simple recognition of the Disney Channel cannot imply "effectiveness." Recognition does not imply preference or other choice/behavioral activity-it is only a very basic measure of memory. "Reaching" is equally ambiguous and subject to the same criticism.

b. The sources of recognition of cigarette logos are not necessarily a derivative of advertising. Other possibilities include: (1) experimental bias already discussed under the first conclusion; (2) school based anti-smoking material, already cited by Fischer, et al. [1991]; and (3) various anti-smoking messages and no smoking logos.

Strategic Management & Public Policy Reviewer

Overall Evaluation. A fundamental problem with this study is that there is no basis for its underlying premise that early logo recognition leads to premature smoking.

Specific Comments. This reviewer offered the following more specific comments regarding the Fischer, et al. [1991] study:

1. Recognition of "children's brands" may be understated because of the logos used. Nike, for example, is not particularly a children's product, and the generic name "Kelloggs" is not necessarily a good proxy for particular Kellogg's cereals that are promoted to children. Moreover, all but one of the children's logos (Disney Channel) include a word as part of the logo, which may tend to reduce correct matching rates among children who cannot yet read.

2. Recognition rates of "adult brands" may be low because of the products used. Children had to match the logo with a product. The adult products included two computers and two television networks, which had to be matched with a television. It seems likely that children would confuse the "product" pictures of a TV and a computer, since the dominant feature of a picture of a computer is the TV monitor. For these reasons , recognition rates of cigarettes compared to other adult products are probably overstated.

3. The key comparison of Old Joe and Mickey Mouse lacks any control. Given the fact that there is very little difference among six year olds in recognition scores for cigarette brands, it is extremely likely that they would also do quite well on the best known adult logo, Chevrolet. In fact, Chevrolet is more widely recognized among all children than is Old Joe. "Chevrolet is as well recognized as Mickey Mouse," however, doesn't have quite the same punch. It is suggested that by age 6, kids may be able to recognize most logos.

4. Another factor producing the lack of significant differences in recognition on Old Joe and Mickey Mouse is sample size. Among 6 year olds, Mickey Mouse is more widely recognized, but the difference is not significant in part because there are only 23 6-year-olds in the sample. Sample sizes are larger for the other age groups (e.g., 60 at age 5).

5. The lack of any relationship between children's recognition of cigarette logos and the use of cigarettes in the home is curious. Kids should be more likely to recognize the logo of the brand their parents smoke, at least as long as the logo is on the package. Thus, the Marlboro red roof, the "Camel" name, and the camel and pyramids should be more widely recognized by children from households in which those brands are used.

SUMMARY

The summary of the findings concerning reviews of the Fischer, et al. article [1991] is seen in Table #1.

Using the major disciplinary standards discussed earlier, there are six on which the majority of the reviewers identified errors in the reported research: building in of controls, validity and reliability, outcome related to variables of investigation, reporting and/or discussion of convergent results, support for claims and assertions, an advocacy position by the authors and identification of authors' speculation.

Summarizing some of the key evaluations, on the issue of validity there is no basis offered for the study's underlying premise that early logo recognition leads to smoking. There is no evidence offered by the authors that logo recognition scores are correlates to behavior initiation. The failure of the authors to consider the basic works in recall and recognition [Bagozzi and Silk 1983; Singh and Rothschild 1983; Mizerski 1982; Valentine and Blum 1961; Wells, et al, 1989; and Finn 1988] is disturbing.

The authors go far beyond their basic methodology inferring advertising recognition to a simple matching exercise by children. Leckenby and Plummer [1983], cited by the authors, conclude that advertising recognition is a complex, multi-variable measurement, not the more simplistic matching exercise offered by Fischer, et al. [1991].

Fischer, et al [1991] ignore data showing improvement in correctly match logos to products among older children in the study and likewise do not report on mismatches.

TABLE 1

FISCHER, ET AL., "BRAND LOGO RECOGNITION BY CHILDREN AGED 3 TO 6 YEARS," JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, VOL. 266 NO. 22, DECEMBER 11, 1991, PP. 3145-3148

The study found that brand logo recognition increased with age, and that by age 6, recognition of Old Joe is comparable to the Disney Channel logo. The authors state that recognition of the Disney Channel logo and Old Joe were "highest in their respective product categories" [Fischer, et al., p. 3145]. They fail to point out in conjunction with the statement that the respective categories were "children's brands" and "cigarette brands" and that the Disney Channel, McDonald's, Burger King, Dominos Pizza, Coca Cola, Pepsi, Nike, Chevrolet and Ford had higher recognition scores than Old Joe.

Among conclusions reached are that "very young children see, understand, and remember advertising" [Fischer, et al., p. 3145]; that "children's knowledge of cigarette brand logos is most likely the result of their exposure to environmental tobacco advertising" [Fischer, et al., p. 3148]; and that "R. J. Reynolds Tobacco Company is as effective as the Disney Channel in reaching 6-year old children" [Fischer, et al., p. 3148]. There is nothing in the data reported to substantiate these conclusions.

DISCUSSION

One could question the time and effort spent in appraising an already published article. "After all, isn't this overkill for a mere 7 pages?" In many cases it probably would be so, but here we are dealing with a paper that has the potential to have a major impact on public policy. It was the crescendo of calls for public policy changes regarding advertising, specifically for cigarettes, and demands that a particular advertising campaign be banned that prompted our interest. Indicative of the general acceptance of the research was an accompanying editorial in JAMA entitled "Tobacco Marketing: Profiteering From Children." In it, Representative Henry Waxman makes the statement that "the tobacco companies' success at targeting young people is apparent from the data reported in this issue of THE JOURNAL. Old Joe Camel has demonstrated appeal and recognition among youth" [Waxman, p. 3185].

A reviewer of the original draft of this paper commented, "but what impact has occurred as a result of this research?" As recently as August, 1993 there was a strong recommendation of the Federal Trade Commission staff to completely ban the Joe Camel campaign [Wall Street Journal, 8/11/93, pp. B-1-B5]. The Fischer, et al. research was specfically cited as the major underpining for the staff recommendation. The implications for the first amendment and for other "controversial" ad campaigns is all too evident. The following is indicative of the impact of the staff recommendations:

FTC watchers say the move may signal a vigorous new activist mindset at the FTC. 'It is one of the most controversial isues to come before the FTC in recent years because it is such a strong step and such a strong statement,' says Linda A Goldstein, a partner with Hall, Dickler, Lawler, Kent & Friedman, a New York law firm. 'It would be the first time the FTC has ever taken such a Draconian step.' [Wall Street Journal [8/11/93, p. B-1]

We submit that when four reviewers from differing disciplines independently evaluated the Fischer, et al. [1991] study they raised significant questions about the quality of it. Particularly disturbing is the advocacy nature of the research and serious questions concerning reliability and validity. Our concern is not with supporting or attacking cigarette smoking or even cigarette advertising. Rather, our concern is with the quality of the research that can impact public policy, as far-reaching as that suggested by the FTC staff, that then can be applied to other product categories. We suggest the need to better codify research guidelines and more rigorously apply them to consumer research.

REFERENCES

Advertising Age, "Old Joe Must Go," editorial page, January 13, 1992

Aitken, P.P., D.S. Leathar, F.S. O'Hagen, and S.I. Squair, "Children's Awareness to Cigarette Advertisements and Brand Imagery," British Journal of Addiction, 1987, 82: pp. 615-622

Aitken, P.P., D.S. Leathar, A.L. Scott and S.I. Squair, "Cigarette Brand Preference of Teenagers and Adults," Health Promotion, 1988, 2: pp. 219-226

Aitken, P.P. and D.R. Eadie, "Reinforcing Effects of Cigarette Advertising on Underage Smoking," British Journal of Addiction, 1990, 85: pp. 399-412

Associated Press, "Most Kids Can Identify Camel Character," New York: December 11, 1991

Associated Press, "AMA Protest Parade Targets Old Joe Ads," Chicago: June 21, 1992

Atlanta Journal and Constitution , "For Kid's Sake, Scrap Cartoon Camel," December 14, 1991

Bagozzi, Richard P. and Alvin J. Silk, "Recall, Recognition and the Measurement of Memory of Print Advertisements," Marketing Science (1983) vol. 2 no. 2, pp. 95-134

Boston Globe, "What's Good For R.J. Reynolds," December 12, 1991

Boston Herald, "Study: Camel Cartoon Sends Kids Smoke Signals," December 11, 1991

Brown, David, "Old Joe and Mickey Mouse Nose to Nose," The Washington Post, December 11, 1991, p. A-23

Chapman, S. and B. Fitzgerald, "Brand Preference and Advertising Recall in Adolescent Smokers: Some Implications for Health Promotion," American Journal of Public Health, 1982, 72: pp. 491-494

Charlotte Observer, "Puffing Camel Rivals Mickey in Kid's Eyes," December 11, 1991, pp. 1-A & 6-A

Chicago Tribune, "Panel OK's Resolution Condemning Old Joe Ads," May 6, 1992

Cimons, Marlene, "Tobacco Foes Seek Federal Injunction," The News & Observer, Raleign, NC, March 29, 1992, p. 2G

Dagnoli, Judann, "JAMA Lights New Fire Under Camel's Ads," Advertising Age, 1991, pp. 3 & 32

Davis, R.M., "Current Trends in Cigarette Advertising and Marketing," New England Journal of Medicine, 1987, 316: pp. 725-732

Denver Post, "Joe Camel, Child Molester," March 15, 1992

Detroit Free Press, "Say No to Joe Camel," March 17, 1993, p. 1-E

Detroit News, "Do Your Kids Know Old Joe," December 12, 1991, p. 5

Enrice, Dottie, "Feds Walk A Mile To Ban Joe Camel," New York Newsday, March 14, 1992, p. 8

Finn, Adam, "Print Ad Recognition by Children Aged 3 to 6 Years," Journal of Marketing Research (May, 1988) Vol. 25, pp. 168-177

Fischer, Paul M., Meyer P. Schwartz, John W. Richards Jr., Adam O. Goldstein, and Tina H. Rojas, "Brand Logo Recognition by Children Aged 3 to 6 Years," Journal of the American Medical Association, December 11, 1991, vol. 266, no. 22, pp. 3145-3148

Fort Worth Star-Telegram, "Joe Camel: Brilliant Campaign, Unintended Results," December 17, 1991

Goldberg, M.E., G.J. Gorn and W. Gibson, "TV Messages for Snack and Breakfast Foods: Do They Influence Children's Preference?" Journal of Consumer Research, 1978, 5: pp. 73-81

Goldstein, A.O., P.M. Fischer, J.W. Richards, and D. Creten, "Relationship Between High SchoolStudent Smoking and Recognition of Cigarette Advertisement, " Journal of Pediatrics, 1987, 110: pp. 488-491

Gorn, G.J. and R. Florsheim, "The Effect of Commercials for ASdult Products on Children," Journal of Consumer Research, 1985, 11: pp. 962-967

Harrigan, Susan, "Snuffed Out!" New York Newsday, June 27, 1992, p. 5

Horovitz, Bruce, "Most -Advertised Cigarettes Are Teens Choice, Study Says," Los Angeles Times, March 13, 1992, p. 16

Kong, Dolores, "Do Ads Lure Youngsters To Drink, Smoke?" The Boston Globe, April 27, 1992, pp. 41-44

Krugman, Herbert, "Memory Without Recall, Exposure Without Perception," Journal of Advertising Research, Vol. 1 [1982] pp./ 80-85

Leckenby, John and Joseph Plummer, "Advertising Stimulus Measurement and Assessment Research: A Review of Advertising Testing Methods, Current Issues and Research in Advertising, James Leigh & Claude Martin, eds., 1983, pp. 135-165

Levine, Andrea C. (Assistant Attorney General of State of New York), letter to Senator John C. Danforth (R-MO.), July 8, 1992

Levit, E.M., C. Coate, and M. Grossman, "The Effects of Government Regulation on Teenage S moking," Journal of Law Economics, 1981, 24: pp. 545-569

Lipman, Joanne, "Surgeon General Puts Joe Camel on the Endangered List," Wall Street Journal Classroom Edition, April 1992, p. 8

Mazis, M.B., D.J. Ringold, E.S. Perry and D.W. Denman, "Perceived Age and Attractiveness of Models in Cigarette Advertisements," Journal of Marketing, Vol. 56 No 1 [1992], pp. 22-38

McNeal, J.U., Children As Consumers (1987), Lexington, MA: Lexington Books, pp. 5, 12, 47, 179

Mizerski, Richard, "Viewer Miscomprehension Findings are Measurement Bound," Journal of Marketing, Vol. 46 No 4, (1982), pp. 32-34

New York Observer, "Commentary," March 23, 1992

Philadelphia Inquirer, "Pushing Butts to Kids," March 15, 1992

Raleigh News & Observer (Raleigh, N.C.), "Old Joe, 'Friend' of Kids," March 12, 1992

Reuters News Service, "US Doctors Stage Protest Against Old Joe," Chicago: June 21, 1992

Roberts, Selena, "Surgeon General Wants to Snuff Out Joe Camel," Orlando Sentinel, May 25, 1992, p. A-4

San Francisco Examiner, "Just Say No to Joe Camel, March 17, 1992

Scripps Howard News Service, "Senator Wants Ban on RJR's 'Old Joe'" March 20, 1992

Seattle Post-Intelligencer, "To Some, Cartoon Camel Isn't Funny," December 11, 1991, p. 9

Seldon, B.J. and K. Dordoodian, "A Simultaneous Model of Cigarette Advertising: Effects on Demand and Industry Response to Public Policy," Review of Economic Statistics, 1989, 71: pp. 673-677

Siegel, S., Nonparametric Statistics, New York: McGraw-Hill Book CVompany [1956]

Singh, S.N. and M.L. Rothschild, "Recognition as a Measure of Learning from Television Commercials," Journal of Marketing Research, Vol. 20 [August 1983] pp. 235-248

Snider, Mike, "Doctors Turn Up The Heat on Smoking," USA Today, December 11, 1991, pp. 1-2D

Standora, Leo, "Docs: Time For Old Joe To Become a No-Show," New York Post, March 10, 1992, p. 3

Syracuse Herald-Journal, "Luring Kids to Smoke: The Uncool Side of Joe Camel," March 11, 1992

U.S. News & World Report, "An Anti Smoking Campaign Heats Up," p. 14

Valentine, Appel and M.I. Blum, "Ad Recognition and Response Set," Journal of Advertising Research , (June, 1961) pp. 13-21

Vesey, Susannah, "Up In Smoke," Atlanta Constitution, ("Peach Buzz" column) March 20, 1992

Wall Street Journal, "Joe Camel Is Also Pied Piper, Research Finds," December 11, 1991, pp. B1 & B4

Wall Street Journal, "FTC Staff Recommends Ban of Joe Camel Campaign," August 11, 1993, pp. B1 and B5

Warner, Fara, "Novello Throws Down Gauntlet," AdWeek's Marketing Week, March 16, 1992, p. 4

Warner, K.E., "A Ban on Promotion of Tobacco Products," New Eng land Journal of Medicine, 1987, 316: pp. 745-747

Waxman, Henry, "Tobacco Marketing: Profiteering From Children," Journal of the American Medical Association, December 11, 1991, vol. 266, no. 22, pp. 3185-3186

Webster's Ninth New Collegiate Dictionary, Springfield, MA: Merriam-Webster Inc. (1990)

Wells, William, John Burnett, and Sandra Moriarty, Advertising: Principles and Practices (1989) New York: Prentice-Hall Inc.

Zajonc, R.B. and H. Markus, "Affective and Cognitive Factors in Preferences," Journal of Consumer Research, Vol. 9 [September, 1982], pp. 123-131

Zukin, Helen E., "Brief Exposures," Indoor Pollution Law Report, August, 1992

Washington Times, "Camels 'Old Joe' Creates Young Smokers," letter from Louis W. Sullivan and Antonia C. Novello, March 30, 1992, p. E2

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