The Effects of Tv Messages For High and Low Nutritional Foods on Children's Snack and Breakfast Food Choices

ABSTRACT - This study investigated the capacity for food messages developed for television to influence the nature of children's snack food and cereal selections. In a controlled experiment, children were exposed to varying numbers of commercials for highly sugared snack foods and cereals, or Public Service Announcements (PSA's) for less sugared, nutritious fruits and vegetables. In addition, a 24 minute edition of the 'Fat Albert' program starring Bill Cosby called 'Junk Food' was utilized. After exposure to one or another of these stimuli, children were presented with a number of snack food and cereal choices. Exposure to the Fat Albert program (even when it was interrupted by the commercials for snack foods) was most effective in reducing the number of highly sugared foods selected and increasing the number of fruits and vegetables selected. Exposure to the PSA's or the snack food commercials was also capable of creating differences in the foods selected by the children. These differences were noted even though children in all the conditions were aware of which foods were "healthy" or "unhealthy".

Citation:

Marvin E. Goldberg, Gerald J. Gorn, and Wendy Gibson (1978) ,"The Effects of Tv Messages For High and Low Nutritional Foods on Children's Snack and Breakfast Food Choices", in NA - Advances in Consumer Research Volume 05, eds. Kent Hunt, Ann Abor, MI : Association for Consumer Research, Pages: 540-545.

Advances in Consumer Research Volume 5, 1978      Pages 540-545

THE EFFECTS OF TV MESSAGES FOR HIGH AND LOW NUTRITIONAL FOODS ON CHILDREN'S SNACK AND BREAKFAST FOOD CHOICES

Marvin E. Goldberg, McGill University

Gerald J. Gorn, McGill University

Wendy Gibson, (student) Stanford University

ABSTRACT -

This study investigated the capacity for food messages developed for television to influence the nature of children's snack food and cereal selections. In a controlled experiment, children were exposed to varying numbers of commercials for highly sugared snack foods and cereals, or Public Service Announcements (PSA's) for less sugared, nutritious fruits and vegetables. In addition, a 24 minute edition of the 'Fat Albert' program starring Bill Cosby called 'Junk Food' was utilized. After exposure to one or another of these stimuli, children were presented with a number of snack food and cereal choices. Exposure to the Fat Albert program (even when it was interrupted by the commercials for snack foods) was most effective in reducing the number of highly sugared foods selected and increasing the number of fruits and vegetables selected. Exposure to the PSA's or the snack food commercials was also capable of creating differences in the foods selected by the children. These differences were noted even though children in all the conditions were aware of which foods were "healthy" or "unhealthy".

In April 1977, the consumer activist group, Action for Children's Television (ACT) petitioned the FTC to prohibit the advertising of candy to children on television. At just about the same, the office of Research and Analysis of the New York State Assembly made virtually an identical recommendation to the New York State Legislature related to the advertising of all heavily sugared products to children (Mauro and Feins, 1977).

The focus of the discontent with present advertising practices revolves around the postulate that given their relatively underdeveloped cognitive abilities, young children are likely to be unduly influenced by the more than 20,000 TV commercials that they are exposed to each year (Adler et al, 1977); the majority (55%) of which are for either breakfast cereals or snack foods (Gussow, 1972). ACR (1977) has cited evidence from associations such as the American Dental Association and the Society for Nutrition Education to document the fact that frequent sugar consumption in the form of the typical pre-sweetened breakfast cereals and snack foods usually advertised on TV contributes especially to tooth decay in children, as well as other potential physical problems such as obesity. While there seems to be relatively little controversy regarding the potentially negative effects of an overconsumption of sugar and sugared products, what appears to remain an issue of greater contention is the establishment of a causal link between exposure to children's TV advertising for sugared snacks and breakfast foods and subsequent attitudes and behavior regarding these food products.

Correlational evidence (Clancy-Hepburn, Hickey and Nevill, 1974) is suggestive of such a link, as is observational date (Atkin, 1975; Galst and White, 1976), and experimental evidence (Goldberg and Gorn, 1977; Poulos, 1975). These notwithstanding, a recent summary of the research in this field concludes that the nature and extent of these effects is not clear and that much remains yet to be assessed (Adler et al, 1977).

While the majority of both policy-makers and researchers have tended to focus on the degree to which TV commercials affect children adversely, consideration has more recently also been given to the manner in which TV can be used to encourage positive or pro-social behavior, including sound dietary habits in children. Thus a variety of groups have been involved in developing Public Service Announcements and program material that encourage children to eat nutritious foods such as fruits and vegetables and to moderate their intake of highly sweetened snacks and cereals they see advertised on TV (see Society for Nutrition Education, 1975). The White House Conference on Food, Nutrition and Health (1970) specifically recommended that the President establish a permanent task force to formulate a mass-media nutrition education program. More recently the New York State Office of Research and Analysis recommended in their report (March, 1977) that stations be requested to air PSA's emphasizing proper nutrition for children, to be paid for by a tax levied on all broadcasters as a function of their advertising revenues. These efforts suggest that at least these policy makers and other interested parties believe there is a link between the material children are exposed to on TV (messages featuring either low or high nutritious foods) and their subsequent attitudes and behavior.

Research has documented that: very young children in particular are most likely to believe television commercials (Ward and Wackman, 1973); are most likely not to understand disclaimers (Atkin, 1975), and are least likely to understand the sales intent of the commercials (Rubin, 1972; Robertson & Rossiter, 1974). Thus television may have an especially strong impact on the very young, with the potential to maximally influence their food habits which for these children may not be as yet fully formed.

One of the most elaborate nutrition education efforts via the mass media is a series of programs called "Mulligan Stew". It includes a series of six half hour films integrated with written teaching materials that attempts to get children to learn the various nutrients associated with the four basic food categories. An experiment evaluating the effectiveness of this program revealed that it was successful in communicating the nutritional information bus was, on the whole, unable to get the children to change their consumption behavior (Jenkins et al, 1975).

The present study sought to test the potential of TV as a medium for influencing children's snack and breakfast food preferences. The 30 second message - either Public Service Announcement for highly nutritious foods or commercials for less nutritious foods was one vehicle used. The second was a 24 minute program entitled "Junk Food", one of a number of issues treated by a series called "Fat Albert" starring Bill Cosby who notes:

What this series attempts to do is harness the natural interest (enjoyment of cartoons) and act as a conveyor of learning to instill an awareness of life and develop codes of behavior. Employing animation as the primary medium to teach its (Fat Albert's) goals differs from Sesame Street and the Electric Company in that its emphasis is upon the affective development of young viewers (1976, p. 60-61).

It was anticipated that exposure to these materials, insofar as they made the various foods attractive and salient, would serve as cues to the children and encourage the selection of either more or less nutritious foods, as the case might be. It was also considered likely that an entire program devoted to nutrition would be more effective than the 30 second PSA's. The increased time would not only direct attention to nutritional issues for a longer period of time, but potentially ought to allow for greater creativity in developing a meaningful theme.

In a further effort to vary the degree of exposure to messages for more or less nutritious snack and breakfast foods, conditions were developed in which the number of PSA's or commercials either approximated the number permitted per half hour by the NAB Code or was double that number.

METHOD

Subjects

The twin objectives of utilizing children who were as young as possible, yet at the same time could respond to simple paper and pencil measures resulted in the selection of a sample of first grade children.

Subjects were drawn from three Northern California-Bay area schools, about one mile from one another. The communities surrounding the schools were part of the same upper-middle class suburban area and were essentially indistinguishable from one another. In all, there were seven first grade classes in the three schools (three in one school and two in each of the remaining two schools). Total enrollment in those classes was approximately 175 of whom 122 or 70% participated in the study, (having returned signed permission statements from their parents). Subjects were randomly allocated to one of eight conditions (described below). At each school, assistants would draw a randomly selected subset of children from different classes and take them to a room where the experiment was conducted. Typically seven or eight would then watch one of the following profession- ally videotaped and edited programs.

Independent Variables

1. "Yogi's Gang", a 24 minute animated cartoon program provided the context for a 30 second commercial for each of the nine food products listed below. These were typically inserted, two at a time at five appropriate points in the program, for the total of 4.5 minutes of commercial time; an amount roughly corresponding to what children typically see in a half hour on childrens TV (N.A.B. Code, 1975). Commercials were for: Mounds Candy Bar, Lollipop Lifesavers, Crackerjacks, Hershey Candy Bar, Blow Pops, Milky Way Candy Bar, Kool-Aid, Sugar Crisps, and Fruity Pebbles.

2. "Yogi's Gang" plus the commercials for the nine food products, each repeated once for a total of 9 minutes of commercial time. These were inserted three at a time at six points during the program.

3. "Yogi's Gang" plus eight Public Service Announcements (for a total time of 4.5 minutes) stressing the values of eating fruits, vegetables, milk; and suggesting moderation in the use of candy bars, sugared cereal and other sugared snacks. (The majority of these were selected from among those produced under the direction of the Bay area Committee on Children's Television; two produced by ABC were taped off-air). One or two of these Public Service Announcements were inserted at the same five points in the program as in condition 1.

4. "Yogi's Gang" plus eight Public Service Announcements each repeated once (for a total time of 9 minutes). These were inserted two or three at a time at the same 6 breaks as in condition 2.

5. A 24 minute "Fat Albert" program entitled "Junk Food". This is an animated childrens cartoon aired weekly on CBS, starring Bill Cosby. The program depicts "Fat Albert" and his friend "Slim" who eat only Junk Food (mostly non descript candy).This leads to a toothache for Slim followed by an anxiety producing visit to the dentist. This convinces Fat Albert, but not Slim, to turn to good healthy foods. Finally, Slim having skipped meals in favor of junk food is so weakened that he is responsible for losing the "big" football game. Most of the program is animated, with Bill Cosby appearing occasionally to underline a point made in the course of the animated cartoon. The theme of the program generally is: "This is Bill Cosby coming at you with music and fun and if you're not careful you may learn something before it's done."

6. The Fat Albert "Junk Food" program with the eight Public Service Announcements described above - each repeated once for a total of 9 minutes. In essence this condition represented the maximum message with regard to eating highly nutritious foods. The PSA's were typically inserted three at a time at six points in the program; (because of limitations of time and subject population we were unable to run this condition with the eight Public Service Announcements shown once only; i.e. for a total of 4.5 minutes).

7. The Fat Albert "Junk Food" program with the nine 30 second food commercials described above for a total of 4.5 commercial minutes. These were typically inserted two at a time at five points in the program. (Because this program was originally aired on CBS during Saturday morning, this condition comes close to replicating reality; i.e. what children viewing the program off-air would actually have viewed: the "Junk Food" program with its message to moderate candy and sugared product intake, ironically along with commercials for several examples of these sugared products.

8. A control group not exposed to any program at all. (Although the focus of this study was on the direct comparisons permitted by the groups described above, it was decided to incorporate this group to establish "baseline" data, as well as to assess the comparability of children in each of the three schools used in this study).

Dependent Variables

Following the program the children were involved in a four-five minute discussion of their favorite TV programs so as to break any momentary set that the program material might have established. At that point the children turned to the questionnaire that was used to obtain the dependent measures.

Subjects were first asked to indicate if they were a boy or girl and were then shown the first of a series of 3' x 2' boards each divided by lines into six of the six rectangles so that a child saw six actual snack foods mounted on a board. Following consultation with nutritionists, three of the six foods were selected because they represented "healthy" highly nutritious snacks and the remaining foods were selected because they represented snacks that would be "unhealthy" and low in nutrition if over-eaten at the expense of other more nutritious foods. The first board had the following three snacks considered as highly nutritious: banana, peanuts, raisins; and three sugared snacks considered as less nutritious: Mounds Candy Bar, Jelly Beans and Lollipop Lifesavers. The first page of the questionnaire had a pictorial representation of the food sketched on the page in the same location as on the board and identified in writing as well. The experimenter reviewed the items on the board in front of the children and subsequently on the first page of the questionnaire and made sure they recognized the correspondence between the board and the questionnaire page. No problems were encountered in this regard throughout the experiment.

The experimenter then structured the situation for the children's selection of the snack foods by indicating the following:

Now I want you all to pretend something. Let's pretend that your Mommy and Daddy were going away on a holiday, and they asked me to babysit for you while they were gone. Now I don't know the kind of foods you would want while they were gone. So suppose I said here are six snacks - you can eat three of them. Now you can tell me which three you would want by putting a big X through the three snacks on your page you would want most.

After they had indicated their three choices the experimenter removed the first board and revealed a second board with a second set of six snack foods; once again with three foods considered highly nutritious (apple, carrots, orange) and three considered less nutritious (Crackerjacks, chocolate "stars" and Ju-Jubes). The experimenter introduced the second board by suggesting:

Now lets suppose it was the second day I was babysitting for you while your Mommy and Daddy were away on a holiday. And again, because I didn't know the kind of snacks you like, I said here are six snacks and you can pick three of them to eat today.

After the children had once again made their three choices, the experimenter went on to the choices for the "third day of babysitting" (pear, cheese and crackers, fruit cocktail-snack cup; Hershey Candy Bar, Malted Milk Balls and Blow Pops) and then the "fourth day of babysitting" (strawberries, peaches-snack cup, raisins; Milky Way Candy Bar, Hershey Chocolate Kisses, Gum Drops).

It is important to note that in each set of six foods one or two of the sugared candy products were among the ones advertised in the commercial conditions, while one or two were more akin to the nondescript sugar products used in the Fat Albert program. Among the 3 more nutritious snacks presented in each set, one or two were among those featured in the Public Service Announcements while one or two were not.

Following the four boards of snack foods the experimenter went on to introduce two boards consisting of breakfast foods:

O.K. now what about breakfast. I wouldn't know what you would want for breakfast so you would have to tell me. Let's see, you could have: ....

Once again the six choices presented on the board consisted of three foods selected with the help of nutritionists: milk, orange, eggs and toast; and three considered less nutritious: Koolaid, Super Sugar Crisps and Fruity Pebbles (sugared cereal). These foods were pictorially represented in the same fashion in their questionnaire and here too, the children were asked to "put an X through the 3 foods you would want for breakfast". Once completed, the children were provided with a second set of breakfast choices consisting of the same more nutritious alternatives (milk, orange, eggs and toast); and Cap'n Crunch, Frankenberries and Koolaid as the less nutritious alternatives. (Only the cereals were changed in an effort to provide the same reasonably realistic breakfast choices in each set). While the four cereals are all roughly equally popular, the first two (Sugar Crisps, Fruity Pebbles) were among the nine advertised products in the commercial conditions while the latter two (Cap'n Crunch, Frankenberry) were not.

The second set of questions focused on the children's level of awareness of the healthy or unhealthy nature of each of the foods they just had seen on each of the six boards (four sets of six snack foods plus two sets of six breakfast foods). After all the children's choices had been made, the experimenter returned to the first board of snack foods and asked the children to turn to the next page of their questionnair which once again included a pictorial representation of the foods corresponding to the first board. In this instance, however, each food had a "smiling face" with the word "good" and a "frowning face" with the word "bad" immediately below it. The experimenter indicated:

Now I want to ask you a different kind of question. Let's think about ....... Do you think ..... are good for you and healthy or bad for you and not healthy? If you think ..... are good for you and healthy put a big X through the happy face that says good for you. If you think ..... are bad for you and not healthy then put a big X through the sad face that says bad for you.

Now what about .... ? Are ..... good for you and healthy or bad for you and unhealthy? Put an X through the happy face that says good for you and healthy or an X through the sad face that says bad for you and unhealthy.

The experimenter continued in this way for each of the foods on each of the remaining boards.

At the conclusion of the entire set of questions, where time permitted, the experimenter engaged in a group discussion with the children, asking questions such as: why they had made choices they did; what they thought could be done given the "we all sometimes eat things that are not healthy for us."

RESULTS

The children were able to select a maximum of either three more nutritious or three less nutritious foods on each of six boards, making a maximum of 18 possible in either direction. The control group selected a mean number of 10.2 less nutritious foods (and 7.8 more nutritious foods). As indicated in Table 1 these figures are very similar for the children in each of the three schools suggesting comparability on this key dimension.

TABLE 1

MEAN NUMBER OF LESS NUTRITIOUS FOODS SELECTED BY CHILDREN AT EACH SCHOOL

Initial examination of the mean scores revealed no significant difference between the groups exposed to either 4.5 minutes of PSA's (x=8.67) or 9 minutes of PSA's (x=8.73) in the context of the Yogi's Gang program. The difference between those exposed to either 4.5 minutes of commercials (x=12.12) or 9 minutes (x=13.14) was almost as small and equally insignificant. Given the similarity of data in the two groups exposed to Public Service Announcements and the similarity of data in the two groups exposed to commercials, it appeared more parsimonious to combine each pair in further analysis. [In fact, the similarity of the results with repetitive exposure parallels several earlier studies suggesting that repetition does not further change short term attitudes much beyond the impact of the initial exposure (Goldberg and Gorn, 1974, 1977)].

An analysis of variance conducted with the four original conditions and the two merged conditions described above, revealed highly significant overall differences (F=9.62, 5, 116 df, p < .001). Table 2 presents the means for each of the six groups in ascending order.

TABLE 2

MEAN NUMBER OF LESS NUTRITIOUS FOODS SELECTED

Post Hoc Newman Keuls analyses suggested the following significant differences between particular groups. Children exposed to Fat Albert selected significantly fewer less nutritious foods (x=2.87) than those exposed to Yogi's Gang and PSA's (x=8.70), who in turn selected significantly fewer than those who were exposed to Yogi's Gang and commercials (x=12.58). The control group fell between the latter two (x=10.20) but was not significantly different from either.

As might be expected, children who viewed Fat Albert with commercial insertions selected a somewhat greater among of the less nutritious foods (x=6.06) than those who viewed the program uninterrupted (x=2.87) but this difference was not significant. Similarly, but perhaps more surprisingly, adding 9 minutes of PSA's to the Fat Albert program also tended to increase the number of less nutritious foods selected (5.60) but, again, not to a statistically significant extent.

In sum, while Fat Albert by itself was significantly more effective than Yogi's Gang plus PSA's in reducing the number of less nutritious foods selected, inserting commercials or Public Service Announcements into Fat Albert tended to reduce the effectiveness of the program to the level of Yogi's Gang plus PSA's. Nevertheless, children in each of these latter conditions still selected significantly fewer less nutritious foods than the controls or those in the Yogi's Gang plus commercial condition.

An analysis of variance performed on just the 4 sets of snack food choices considered together was equally significant(F=10.69, 5, 116 df, p<.001). Post Hoc analysis revealed essentially the same pattern of results as were obtained for the total set of foods.

While the overall analysis of variance was also significant for the 2 sets of breakfast foods considered separately (F=4.32, 5, 116 df, p<.001), Post Hoc Newman Keuls revealed only one significant difference: the group seeing Fat Albert selected significantly fewer less nutritious breakfast foods (x=1.53) relative to either the control group (x=3.45) or the groups exposed to Yogi's Gang plus commercials (x=3.87).

The second set of questions asked children to indicate whether each of the 36 foods they were presented with was "good for you and healthy" or "bad for you and not healthy". Very few errors were made in response to this question. In fact, the mean number of errors for all 122 subjects was 2.62. The overall analysis of variance revealed no significant differences between the groups (F=1.12, 5, 116 df, p >.05).

DISCUSSION

The results of this study suggest that the type of snack and breakfast foods children select may be a function of the type of food messages they are exposed to on TV. Those who viewed material stressing the attractiveness and value of eating more nutritious foods (Fat Albert, PSA's) selected significantly more of these types of foods than those who viewed commercials stressing the attractiveness and value of eating less nutritious foods.

The Fat Albert program "Junk Food" was more effective than the PSA's in reducing the number of less nutritious foods selected. A number of explanations may be provided for this finding. For one, the unique constellation of creative factors associated with the Fat Albert program in general, and with Bill Cosby it's star in particular, is probably responsible in large measure for the program's effectiveness.

By contrast, at least some of the PSA's seemed less imaginative, slower moving etc. Of course, a program format with 24 minutes allows for the development of themes with antecedent events leading to subsequent consequences, while the 30 second PSA's make this approach difficult if not impossible. The question remains still, how effective could the Fat Albert approach be in a thirty second PSA?

The evidence in this study suggest that the degree of effectiveness of the various TV messages was not a matter of mere minutes of exposure. Doubling either the number of minutes of PSA's or the number of minutes of commercials had no increased effect on the children who were exposed to the messages. Further, adding 9 minutes of PSA's to the Fat Albert program did not enhance it's effectiveness. In fact, the number of less nutritious foods selected increased in this condition, even if not significantly so. Observation of the children in this condition suggested that the redundancy of the messages contained in the nine minutes of food PSA's added to the 24 minute "Junk Food" program resulted in the children paying little attention to the TV by the end of the sequence. This type of viewer satiation has been observed among adults who are overexposed to TV commercials (Greenberg and Suttoni, 1974) and among children as well (Goldberg and Gorn, 1974, 1977).

While Fat Albert appeared most effective in reducing the number of less nutritious foods selected, it should be emphasized that the Public Service Announcements were also able to significantly reduce this number (and concomitantly to increase the number of more nutritious foods selected). Of course real-life insertion of these PSA's would likely occur together with other commercials many of which might well be for candy, chocolate and other sugared foods. The impact of such "mixed" TV exposure remains to be tested. However, it is interesting to note that the real-life condition of Fat Albert with sugared snack and breakfast commercials inserted throughout the program was still effective in significantly reducing the number of less nutritious foods the children selected.

It is evident that at least this sample of children knew which foods were more nutritious and which were less so. Nevertheless, this appears to have made little difference with regard to the foods they actually selected (something often observed with adults as well). Changes in food choices appear to have developed as a result of what Cosby (1976) calls "affective education" rather than cognitive education. The latter approach is often typified by schools which:

.... often concern themselves solely with the development of cognitive skills, leaving the development of the affective domain to the parents. In doing so what they overlook is the total development of the child (Cosby, 1976, p. 61).

Fat Albert is a human hero who children can empathize with as he struggles with value conflicts and peer group problems. The program themes explore the intangibles of values and ethics, influencing behavior and feelings. It is attitude formation (Cosby 1976, p. 33).

Using Fat Albert or the PSA's to increase the attractiveness and salience of the more nutritious foods seems to have served as a cue that led children to develop more positive attitudes towards these foods and to consider them more actively and positively than they otherwise might have. Similarly the primary purpose for exposing children to snack food commercials in the course of the experiment was to make these foods also more salient. The appropriate directionality of the differences between the control and commercial groups suggests that this goal was accomplished. It may be noted that because the "control" group would typically have been exposed to the same commercials during their normal viewing hours they did not constitute a "control" in the strict sense of the term. Thus, while the groups exposed to food commercials selected a larger number of less nutritious foods than did the control group, it is not surprising to note that this difference was not significant.

The possibility of demand characteristics inducing change in an advocated direction is an ever present problem in experimental design, in this study, the fact that the Fat Albert program was more effective than PSA's would seem to argue against an explanation of results based on demand characteristics. The experimenters' intentions in each condition should have been equally clear and had both groups responded on the basis of demand characteristics they both should have changed approximately equally, which was not the case. The experimenter also tried to eliminate any momentary set the children may have had after the program by interposing a 5 minute irrelevant discussion between the program and the questions the children were asked. While this diversion could have been longer, it should be noted that children's response sets have generally been found to change particularly rapidly (Mischel, 1972). Indeed their excitement in expressing their attitudes about their favorite TV programs (the focus of the intervening discussion) strongly suggested that they had moved away from any initial set they may have developed during the TV program.

With regard to demand characteristics, it is also important to mote that a number of studies have observed that young children do not understand the motives and intent behind TV commercials (e.g. Robertson & Rossiter, 1974). As a result, they are not likely to be moved to hypothesize what is expected of them in an experiment, as a function of the messages to which they are exposed. Utilizing young children in an experiment, in many regards ensures a relatively "naive" population.

This study considered sugared snack foods and sugared breakfast cereals as belonging together toward one end of the low-high nutritions spectrum. Indeed, at least anecdotal evidence suggests that many children treat sugared breakfast cereals as snacks rather than breakfast foods thus blurring the distinction between the two. Nevertheless, there may be good reason to treat the two types of foods separately. For example, the role of cereal in a balanced breakfast is a complex issue and one that perhaps could be addressed more comprehensively than was the case in this study. In fact, the Fat Albert program contained only one very brief reference to breakfast cereals and only one of the PSA's dealt with breakfast cereals. This suggests that an increased focus on breakfast foods in this study might have resulted in greater differences with regard to breakfast food choices. However, the results provide some evidence counter to this hypothesis. The one PSA that did deal with breakfast cereal was much more explicit with regard to the problems associated with sugared cereals than the Fat Albert program was, yet only the children who viewed Fat Albert selected fewer less nutritious breakfast foods relative to the control group or those seeing Yogi's Gang plus commercials. The relative absence of significance with regard to the breakfast food choices in this study may be more readily understood by examining the dependent measures. The inclusion of only two sets of breakfast foods provided a relatively reduced range of potential responses. Moreover, the repetition of all the alternatives except the cereals further constrained the range of possible responses. In sum, future studies that treat breakfast foods and sugared snacks separately could encourage greater specification and elaboration with regard to both the independent and dependent variables.

The general discussion that followed the administration of the questionnaire suggested that children may well consider the issues raised during the experiment with much more specificity than the global "good-bad" "healthy-non healthy" measure that was obtained. For example, when asked what people should do if they were constantly tempted to "eat things that were not good for them" responses included the following: "they should try and avoid the aisles in the supermarket with all of these foods"; "they should brush their teeth after each snack"; "they should try to get good things like fruit inside the chocolate". Thus when pressed further, children appear to be able to treat the nutritional issue in a more multidimensional manner. In fact, while the global awareness of whether foods are or are not nutritious may not be predictive of actual choice behavior, the more specific cognitions children have, or can be taught to have, may be more useful as predictors in future research.

FOOTNOTES

The cooperation of many individuals at the Institute for Communication Research, Stanford University, in particular Don Roberts, is gratefully acknowledged. The assistance of Doreen Croft, Carol Goldberg and Laurel Ward was also appreciated. Thanks are due to Sally Williams of the Committee on Children's Television, San Francisco; Lou Scheimer and Norman Prescott of Filmation, Reseda CA and Lee Hambleton of CFCF-TV, Montreal for much of the TV materials provided.

While Kool-Aid may seem to be an unusual breakfast alternative, in fact, it has been found (e.g. Jenkins et al, 1975) that a small but significant proportion of children do drink Kool-Aid or Pop for breakfast.

REFERENCES

Action for Childrens Television, Newton, Massachusetts, Petition to Promulgate a Rule Prohibiting the Advertising of Candy to Children on Television, Petition to Federal Trade Commission, April 1977.

Adler Richard et al, Research on the Effects of Television Advertising on Children: Review and Recommendation. Report prepared for National Science Foundation Research Applications Directorate, 1977.

Atkin, Charles, "Effects of Television Advertising on Children -- Parent-Child Communication in Supermarket Breakfast Selection," Report #7, Michigan State University, June, 1975a.

Atkin, Charles, "Effects of Television Advertising on Children -- First Year Experimental Evidence," Report #1, Michigan State University, June, 1975b.

Clancy-Hepburn, Katherine, Anthony A. Hickey and Gayle Neville, Childrens Behavior Responses to TV Food Advertisements, Journal of Nutrition Education, 6 (3), 1974, pp. 93-96.

Cosby, William, Jr., An Integration of the Visual Media Via Fat Albert and The Cosby Kids into the Elementary School Curriculum as a Teaching Aid and Vehicle to Achieve Increased Learning, Unpublished Ph.D. Thesis, University of Massachusetts, Amherst Mass., 1976.

Galst, Joann and Mary White, "The Unhealthy Persuader: The Reinforcing Value of Television and Children's Purchase-Influencing Attempts at the Supermarket," Child Development, 1976, 47.

Goldberg, Marvin E. and Gerald J. Gorn, "Children's Reactions to Television Advertising: An Experimental Approach," Journal of Consumer Research, Vol. 1, Sept. 1974, pp. 69-75.

Goldberg, Marvin E. and Gerald J. Gorn, "Childrens TV Commercials: Do Child Viewers Become Satiated Too?" Working Paper, Faculty of Management, McGill University, Montreal.

Gorn, Gerald J. and Marvin E. Goldberg, "The Impact of Television Advertising on Children from Low Income Families," Journal of Consumer Research, In Press, September, 1977.

Greenberg, Allan and Charles Suttoni, "Television Commercial Wearout," Journal of Advertising Research, Vol. 13, No. 5, October, 1973, p.47-54.

Gussow, Joan, "Counternutritional Messages of TV Ads Aimed at Children," Journal of Nutrition, 4 (1972), pp. 48-52.

Jenkins, Suzanne, Mary Stumo and Jane Voichick, "Evaluation of the Nutrition Film Series Mulligan Stew'", Journal of Nutrition Education, 7 (1), 1975 p. 17-19.

Mauro, Frank and Feins, R. P. "Kids, Food and Television: The Compelling Case for State Action", Office of Research and Analysis Program and Committee Staff. New York State Assembly, Albany N.Y., March 1977.

Mischel, Walter, Ebbe B. Ebbesen and Antonette P. Zeiss, "Cognitive and Attentional Mechanisms in Delay of Gratification," Journal of Personality and Social Psychology, 21 (2), 1972, pp. 204-218.

National Association of Broadcasters, The Television Code, 18th edition, June 1975.

Poulos, Rita Wicks, "Unintentional Negative Effects of Food Commercials on Children: A Case Study," Media Action Research Center, 1975.

Robertson, Thomas S. and John R. Rossiter, "Children and Commercial Persuasion: An Attribution Theory Analysis," Journal of Consumer Research, 1 (June 1974), pp. 13-20.

Rubin, Ronald S., "An Exploratory Investigation of of Children's Responses to Commercial Content of Television Advertising in Relation to their Stages of Cognitive Development," unpublished doctoral dissertation, University of Massachusetts, Amherst, Mass., 1972.

Society for Nutrition Education (ed.), Audiovisuals for Nutrition Education, Nutrition Education Resource Series, No. 9, Society for Nutrition Education, Berkeley, California.

Ward, Scott and Daniel Wackman, "Family and Media Influences on Adolescent Consumer Learning," American Behavioral Scientist, 14 (3) January 1971.

White House Conference on Food, Nutrition and Health Final Report, Washington, D.C., Government Printing Office, 1971.

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