A Longitudinal Study of the Use of Children in Magazine Advertising

Douglas R. Hausknecht, University of Akron
Janelle K. Hood, University of Akron
Tomasita Chandler, University of Akron
Barbara Heinzerling,, University of Akron,
ABSTRACT - A content analysis of advertisements over a 36 year period (1953-1988) was used to examine the changes in portrayal of children in advertising. Of 30,000 advertisements in selected issues, 2,000 contained children and were classified further for analysis. Relative use of children has declined somewhat during the study period. The use of preschool age children, compared to school age or infants increased at the end of the period. Early in the study period, female children were more frequently used, supplanted by males later. The use of minority children has increased, commencing in the early 1960’s. Advertising planners are encouraged to reflect population trends in advertising design.
[ to cite ]:
Douglas R. Hausknecht, Janelle K. Hood, Tomasita Chandler, and Barbara Heinzerling, (1996) ,"A Longitudinal Study of the Use of Children in Magazine Advertising", in AP - Asia Pacific Advances in Consumer Research Volume 2, eds. Russel Belk and Ronald Groves, Provo, UT : Association for Consumer Research, Pages: 1-6.

Asia Pacific Advances in Consumer Research Volume 2, 1996      Pages 1-6

A LONGITUDINAL STUDY OF THE USE OF CHILDREN IN MAGAZINE ADVERTISING

Douglas R. Hausknecht, University of Akron

Janelle K. Hood, University of Akron

Tomasita Chandler, University of Akron

Barbara Heinzerling, University of Akron

ABSTRACT -

A content analysis of advertisements over a 36 year period (1953-1988) was used to examine the changes in portrayal of children in advertising. Of 30,000 advertisements in selected issues, 2,000 contained children and were classified further for analysis. Relative use of children has declined somewhat during the study period. The use of preschool age children, compared to school age or infants increased at the end of the period. Early in the study period, female children were more frequently used, supplanted by males later. The use of minority children has increased, commencing in the early 1960’s. Advertising planners are encouraged to reflect population trends in advertising design.

INTRODUCTION

Recent articles in several marketing and retailing industry trade publications have suggested an increase in the use of children in print advertising and a trend toward portraying children who appear in print advertisements in a more adult fashion. For example,in two articles in AdWeek’s Marketing Week, the authors described advertising campaigns that "show smart youngsters who look, talk, and act like scaled-down versions of their baby boom parents" (Enrico 1987, p. 47) and that "star divinely dressed kids looking hair-raisingly adorable" (Jordan 1987, p. 54). The author of a third article in a trade publication called Stores, wrote,

"Kids look different in department store and specialty store advertising, too. The trend is away from the cute, impish kid and toward a more serious, older-looking, somewhat sophisticated and even elegant child." (Wallach 1986, p. 44). Jordan (1987) concluded that these "hot shot kids with dynamite duds and aristocratic attitudes invite unfortunate comparison, not with ourselves, but with our children" (p. 54).

These articles described children who appeared in advertisements for children’s clothing. However, Enrico (1987) pointed out that "Kids aren’t only taking on new roles and images in ads for products targeted to them, they’re also playing a bigger role in ads for adult products" (p. 47). One reason given for the increased use of children in advertisements aimed at adults is that children are increasingly influencing the buying decisions of their parents (Enrico 1987). It appears from a search of the literature, however, that there have been no attempts to systematically explore how children are actually portrayed in advertising. Searches of the Social Science Subject Index from 1976 through 1990, the Business Periodicals Index from 1980 through May 1991, and of the Index for Advances in Consumer Research for the years 1980 through 1988 revealed no studies on this topic. A computer search of the PsychInfo data base for the years 1967 through February 1991 was also conducted and revealed only one study (North 1985) similar to the present study. A content analysis of South African consumer magazines was conducted to determine the use advertisers made of children and the different roles in which they were portrayed in advertising illustrations.

A number of studies have used content analysis to determine how other populations have been portrayed in print advertisements. African Americans (Shuey, King, and Griffith 1953; Cox 1969; Kassarjian 1969; Humphrey and Schuman 1984), women (Courtney and Lockeretz 1971; Wagner and Banos 1972; Belkaoui and Belkaoui 1976; Venkatesan and Losco 1975), men (Wolheter and Lammers 1980), and the elderly (Hollenshead and Ingersoll 1975; Ursic, Ursic, and Ursic 1986) have all been the subjects of such studies. In general, the purpose of each of these studies was to identify stereotypes of the population under study and/or to determine the frequency or accuracy with which the particular population was portrayed.

Purpose

This study describes the portrayal of children in print advertisements in selected magazines by using content analysis to explore such characteristics as age, sex, and race/ethnic origin. The findings from this study will assist marketing and advertising practitioners in understanding the dynamics of the portrayal of children in print advertising during the study period. This can provide guidance for the future use of children in advertising as well as contribute understanding to discussions of potential industry or government regulation of the use of children in this manner. This preliminary analysis also provides a foundation for further investigation of issues related to the use of children in magazine advertising.

Prior Studies

In the conclusions of several content analyses of other populations, the researchers asserted that advertisers have a responsibility to provide accurate, responsible portrayals of populations represented by models used in advertisements. Unerstanding the portrayal of a population in mass media is important because people believe in and respond to those portrayals. Rosalind Williams (1981) has hypothesised that,

"consumers faced with...contradictions to reality tend to ask why their lives don’t resemble the images of advertising rather than why advertising images are not true to life." (in Belk and Pollay 1985, p. 889)

Although content analyses of other populations are of little value in answering the question of how children are portrayed in print advertisements over time, in two of the studies that examined how African Americans were portrayed in print advertisements the authors did take a cursory look at the portrayal of children (Kassarjian 1969; Humphrey and Schuman 1984).

Kassarjian hypothesised that the frequency of African American appearance in magazine advertisements would be curvilinear. It was also hypothesised that more African American females than males would appear in magazine advertisements and that more African American children than adults would appear, because females and children would be less threatening to white readers than African American males or adults.

Kassarjian’s first hypothesis regarding frequency of appearance of African Americans in advertisements was supported with African Americans appearing in 35% of advertisements in 1946, 25% of advertisements in 1956, and 40 percent of advertisements in 1965.

Kassarjian’s hypotheses that more African American females than males would appear in the advertisements and that more children would appear than adults were both rejected. Of the 821 African Americans that appeared in the 503 different advertisements, 92 percent were adult with 80 percent male.

A second study analysing the portrayal of African Americans in magazine advertisements from 1950-1982 was done by Humphrey and Schuman in 1984. Humphrey and Schuman found that although there was a long term rise in the proportion of advertisements that contained African Americans, African Americans were still under represented in magazine advertisements. With respect to the portrayal of children, Humphrey and Schuman found that fifteen percent of African American advertisements (defined as any advertisement that contained an African American, whether or not a white person was also present in the advertisement) were represented solely by children whereas only 5 percent of white advertisements used only children. When children were shown with adults, 38% of African American children were shown under the exclusive supervision of white adults and 31% of African American children were shown under adult African American supervision. No white children were shown under the exclusive supervision of African American adults-66% were overseen by adult whites and the remaining one-third were unsupervised.

RESEARCH QUESTIONS

The central research question of this study is "How did print advertisements in selected magazines from 1953 through 1988 portray children?" The second research question posed in this study is "How has the portrayal of children in print advertisements in selected magazines changed from 1953 through 1988?"

Given the dearth of prior research germane to this area, the hypotheses were stated as though there were no interesting changes or variations. Thus, deviation from these hypotheses would suggest that some change had occurred in the study period.

Hypthesis 1. There will be no pattern of difference in the relative frequencies of printed advertisements that contain children from 1953 through 1988.

Hypothesis 2. There will be no significant differences in the percentages of printed advertisements that portray children of different ages from 1953 through 1988.

Hypothesis 3. There will be no significant differences in the percentages of printed advertisements that portray male or female children from 1953 through 1988.

Hypothesis 4. There will be no significant differences in the percentages of printed advertisements that portray children of various race/ethnic backgrounds from 1953 through 1988.

METHODOLOGY

Content Analysis

Content analysis has been defined as "...any technique for making inferences by systematically and objectively identifying specific characteristics of messages" (Holsti 1969, p. 601). In recent years, the marketing literature has seen a resurgence of this technique for studying the printed record. The goal is to summarise a vast amount of verbal information in a form that is objective, systematic and quantitative (Kassarjian 1977).

As a research method, content analysis is especially well suited to identifying trends or patterns and indicating the consistency of these patterns over time. In a summary of the usefulness of content analysis of advertisements Leiss, Kline, and Jhally (1986) wrote,

"The great strength of content analysis for the study of advertisements is its ability to supply some objective answers to the question #What is going on in these ads?’...It allows us to treat qualitative data in quantitative terms, thus helping ground our analysis of images and words in something more than individual and impressionistic interpretation. When dealing with large amounts of data, patterns that otherwise are difficult or impossible to detect often show up in these procedures." (p. 174)

Content analysis as a method does have certain limitations. The same authors write: (a) Content analysis is inadequate for the measurement of meaning because "...meaning is dependent upon the place of any particular item within an entire system of language and image" (p. 174); (b) Content analysis cannot describe how audiences read or interpret the messages presented; (c) Content analysis can only describe the manifest content of a message (Leiss, Kline and Jhally 1986). Although these limitations must be kept in mind, it is also important to remember the strengths of content analysis as a method. According to Kassarjian (1977), in content analysis,

"the signs and symbols are the units of analysis rather than the intent of the communicator or the actions of the interpreter. Of interest is what was said, the properties of the stimuli, rather than what the communicator claims he said or the interpreter perceived to have been said. Content analysis is the study of the message itself and not the communicator or the audience." (p. 8).

Instrument

The instrument used in this study was developed by college professors from the areas of consumer education, child development, clothing and textiles, and marketing. A detailed coding scheme enabled the judge to score each qualifying advertisement on a number of criteria. These included the number, age, sex, and race of any children in the advertisement. Also recorded were suc details as product (service) category being advertised, the kind of activities in which the child(ren) was engaged, whether the child(ren) appeared to be acting in adult role(s), and the presence of any other models in the advertisement. The present paper examines only the frequencies with which children were portrayed in advertising.

Sampling Procedure

A three stage sampling procedure was used for the magazine selection. In the first stage six classifications of magazines were selected from the Standard Rate and Data Service (1987) publication of consumer magazines. The six used were Youth, Women’s, General Editorial, Men’s, Home Service and Home, and Women’s and Men’s Beauty and Grooming. Within these classifications the magazines with the largest circulation were identified for each five-year interval beginning in 1953 and continuing through 1988. In the second stage, the two magazines that rated number one or number two in circulation within each classification were identified. In the third stage, the magazine title that most consistently appeared in the number one or number two circulation positions for each of the five year periods was identified. Because no magazine title consistently appeared in either of the top two positions in three of the six classifications, it was decided to eliminate those classifications from this study. The three remaining classifications were Home Service and Home, Women’s, and Women’s and Men’s Beauty and Grooming. The three periodicals from the three classifications that were used in this study were Better Homes and Gardens, McCall’s, and Glamour, respectively. The goal of the project was to examine advertisements using children, not necessarily advertisements targeted to children. Over the study period, these magazines were thought to be representative of vehicles in which advertisers might use children as models.

TABLE 1

PERCENT OF ADVERTISEMENTS WITH AND WITHOUT CHILDREN BY PERIOD

Advertisements appearing in the April, August, and December issues of each of these three periodicals were coded for each year beginning in 1953 through 1988. If the April, August, or December issue of a periodical was not available then the prior month’s issue of that periodical was used. The sample included all advertisements that were at least one-half page in size and contained one or more children whose face(s) was(were) visible and who appeared to be 12 years of age or younger. If an advertisement had multiple parts with each part less than one-half page in size but totalling at least one-half page, it was included in the sample. If an advertisement appeared in more than one magazine or in a different issue of the same magazine it was counted as a separate advertisement. The total sample included 324 issues and 2,047 advertisements. Failure to code the year on two advertisements resulted in an effective sample size of 2,045. The number of children and the age, sex, and race, of each child in an advertisement was recorded. The researchers always worked in teams of at least two persons when coding the data to insure resolution of any disagreements. No other tests of inter-rater reliability were conducted.

Initial frequency distributions were conducted on all variables pertinent to this study. Based on these initial frequencies it was decided to collapse the 36 years for which data were collected to nine, four-year categories.

RESULTS

Hypothesis 1

Hypothesis 1 stated that there would be no pattern of differences in the relative frequencies of printed advertisements that contained children from 1953 through 1988. This hypothesis was supported, as no consistent trend was found.

There were a total of 29,370 advertisements in the April, August, and December issues of the three magazins used in this study over the 36-year period. There were 2,045 advertisements that were at least one-half page in size and contained one or more children whose face(s) was(were) visible and who appeared to be 12 years of age or younger. The smallest percentage of advertisements that contained one or more children during a four-year period was 5.54% for the 1969-1972 survey period while the highest percentage was 9.88% for the 1957-1960 survey period (Table 1).

Use of children in print advertisements was highest in the early part of the survey period, 8.21% from 1953 through 1956 and 9.88% from 1957 through 1960, and then dropped steadily over the next 12 years. Use of children in print advertisements began to rise again in the 1973 through 1976 survey period (6.49%) and continued through 1980 (6.65%) when the percentage of printed advertisements containing children again began to decline. The percentage of advertisements containing children in the final four-year survey period was 6.22%.

Advertisements containing children made up less than ten percent of the total sample of advertisements in each four-year survey period. Although the percentages of advertisements with children differed over the 36-year period no obvious trend could be identified. If children were being used more frequently in advertisements for products aimed at them and in advertisements for products aimed at adults as was suggested by Enrico (1987), a positive linear trend might have been evident.

Hypothesis 2

Hypothesis 2 stated that there would be no significant differences in the percentages of printed advertisements that portrayed children of different age groups from 1953 through 1988. This hypothesis was partially accepted.

TABLE 2

PERCENT OF ADVERTISEMENTS WITH CHILDREN BY AGE GROUP

TABLE 3

PERCENT OF ADVERTISEMENTS WITH CHILDREN BY SEX

The differences in the percentages of advertisements containing infants was not significant. However, the differences in the percentages of advertisements containing both preschool age children and school age children were significant (Table 2).

When comparing all three age groups, it was evident that school age children appeared in printed advertisements more frequently than did either infants or preschool age children through eight of the nine, four-year survey periods. However, in the last four-year survey period, 50.67% of the advertisements containing children included preschoolers, compared to 49.33% of the advertisements containing children which included school age children.

Hypothesis 3

This hypothesis stated that there would be no significant changes in the percentages of print advertisements that portrayed male or female children over the 36-year period. This hypothesis was rejected.

Male children appeared in approximately 58% of all advertisements that contained children while female children appeared in nearly 63% of all advertisements that contained children. Because both females and males may have appeared together in an advertisement, these percentages exceed 100 percent. In each of the four, four-year categories from 1953 through 1968 print advertisements were more likely to portray female children than male children. From 1969 through 1972 male and female children appeared in exactly the same number of advertisements. However, in the latter part of the survey period, the four, four-year periods from 1973 through 1988, print advertisements were more likely to contain male children than female children (Table 3).

In two content analyses that examined the roles portrayed by women and men, it was reported that more men than women were pictured in the advertisements analysed (Courtney and Lockeretz 1971; Wagner and Banos 1972). While the measure in the present study was somewhat different (the number of advertisements that contain male and/or female children versus the number of males or females in the advertisement), female children appeared in a greater percentage of advertisements than did male children. Print advertisements portrayed female chldren more frequently in the first 16 years of the survey period and male children more frequently in the last 16 years of the survey period.

TABLE 4

FREQUENCY OF ADVERTISEMENTS WITH CHILDREN BY RACE/ETHNICITY

Hypothesis 4

Hypothesis 4 stated that there would be no significant differences in the percentages of advertisements that portrayed children of various race/ethnic backgrounds from 1953 through 1988. Three categories of race/ethnic backgrounds, African American, Asian, and Hispanic, had too few advertisements in each survey period to support statistical analysis. Two race/ethnic categories, Other and Unknown, were not analysed for differences in statistical significance. Based on the analysis of the data for white children, this hypothesis was rejected.

White children appeared in nearly 95% of all advertisements that contained children and appeared in print advertisements in all nine of the four-year survey periods. No minority children were portrayed in these advertisements before the 1961-1964 survey period. Table 4 presents the frequencies of white children in print advertisements and the frequencies of minority children in print advertisements. The frequencies of African American, Asian, Hispanic, and Other children were combined to produce the frequencies for minority children.

DISCUSSION

In their study of the portrayal of African Americans Humphrey and Schuman (1984) found a long term rise in the proportion of advertisements that included African Americans. Similarly, the present study found a long term rise in the percentage of advertisements that included African American children. The Humphrey and Schuman study also concluded that African Americans were under represented in magazine advertisements when compared with the African American population of the United States. In their study they found 5.7% of the total number of people in advertisements were African Americans. At the time of their study the African American population of the United States was about 12% (Humphrey and Schuman 1984). Although the current study measures the number and percentage of advertisements containing children rather than the number of people in advertisements, it appears that African American children were similarly under represented.

The findings of this study showed that advertisements containing children made up less than ten percent of the total sample of advertisements in each four-year survey period. Children were most likely to be portrayed as school aged and white. Over the entire 36-year survey period, female children were more likely to appear in print advertisements than were male children. However, beginning in the 1973 through 1977 survey period and continuing through 1988, the percentages of advertisements with male children were higher than the percentages of advertisements with female children.

Limitations

There are several limitations of this study that require mention. One limitation is that because the sample of advertisements was drawn from only three magazines, it is impossible to generalise the findings of this study to other periodicals. Also, no attempt was made to control for advertisements coded in one periodical that also appeared in one or both of the other periodicals or in other issues of the same periodical. This could result in over-reporting of the frequency of advertisements that contain children. In the Kassarjian (1969) study of African Americans in print advertising reported earlier, approximately 33% of the advertisements analysed appeared more than once. In the Belkaoui and Belkaoui (1976) study of the portrayal of women in print advertisements, repeated advertisements accounted for only 2% of the sample. By contrast, however, it is important to remember that not controlling for repetition of a given advertisement can be considered to be an approprite mirror of market exposure. For example, when a new product is introduced, marketers may schedule more advertisement inserts in a magazine to increase consumer exposure to the product. A final limitation of this study is that inter-rater reliability was limited to simultaneous coding by two or more coders. No other test was used.

CONCLUSION

Concern over the influence of the various media, and particularly advertising, on children remains high. However, most adults seldom consider how the media, including advertising, influences them or their interactions with others. As stated by Pollay (1986),

"...we like to think of ourselves as personally immune to advertising’s inducements. This is clearly a delusion for some or perhaps many or even most of the public....’Though at first the changes were primarily in manners, dress, taste, and food habits, sooner or later they began to affect more basic patterns: the structure of authority in the family, the role of children and young adults as independent consumers in the society, the pattern of morals, and the different meanings of achievement in the society’" (pp. 23-24).

Does the way children are portrayed in the media affect the way we respond to them? The results of this study cannot provide an answer to this question. However, this study does describe how advertisements in these selected magazines portrayed children and provides a base for future research that explores the effect the portrayal of children has on children and adults.

Studies that explore how other media portray children are necessary to determine whether differences exist in the portrayal of children among the various media. Analysis of print advertisements in the same magazines used in this study prior to 1953 and since 1988 would provide a better picture of how advertisements portrayed children and how that portrayal has changed. Analysis of advertisements from additional classifications of consumer magazines would provide a broader picture of how children are portrayed. Further analyses of the data collected for this study will more closely examine the roles of children in advertisements, products advertised, interaction with other models, and other more complex issues.

Advertisers are seeking to appeal to more and more specific lifestyle segments. As this occurs, the demographics of the customer base change. Selecting models, children as well as adults, that reflect these broader demographic categories will be an increasingly important consideration.

REFERENCES

Belk, Russell W., and Richard W. Pollay (1985). "Images of Ourselves: The Good Life in Twentieth Century Advertising," Journal of Consumer Research, ll, 887-897.

Belkaoui, Ahmed, and Janice M. Belkaoui (1976). "A Comparative Analysis of the Roles Portrayed by Women in Print Advertisements: 1953, 1970, 1972." Journal of Marketing Research, 13(5), 168-172.

Courtney, Alice E., and Sarah Wernick Lockeretz (1971), "A Woman’s Place: An Analysis of the Roles Portrayed by Women in Magazine Advertisements." Journal of Marketing Research, 8(2), 92-95.

Enrico, Dottie (1987). "Images of Children in Advertising: Like Father, Like Son." AdWeek’s Marketing Week, (November 30),p. 47.

Hollenshead, Carol, and Berit Ingersoll (1982). "Middle-aged and Older Women in Print Advertiements. Educational Gerontology, 8, 25-41.

Holsti, Ole R. (1969). Content Analysis for the Social Sciences and Humanities. Reading, MA: Addison Wesley.

Humphrey, Ronald and Howard Schuman (1984). "The Portrayal of Blacks in Magazine Advertisements: 1950-1982." Public Opinion Quarterly, 48, 551-563.

Jordan, Killian (1987). "Kiddie Fashion Ads: Children of a Greater God?" AdWeek’s Marketing Week, (November 30),p. 54.

Kassarjian, Harold H. (1969). "The Negro and American Advertising, 1946-1965". Journal of Marketing Research, 6(2), 29-39.

Kassarjian, Harold H. (1977). "Content Analysis in Consumer Research". Journal of Consumer Research, 4(6), 8-18.

Leiss, William, Stephen Kline, and Sut Jhally (1986). "Two Approaches to the Study of Advertisements". Social Communication in Advertising: Persons, Products, & Images of Well Being (pp.149-174). New York: Methuen, Inc.

North, Ernest John (1985). The Use of Children as Models in Magazine Advertising (Master’s thesis, University of South Africa, 1983. Masters Abstracts, 23(1), 65.

Pollay, Richard W. (1986). "The Distorted Mirror: Reflections on the Unintended Consequences of Advertising." Journal of Marketing, 50(4), 18-35.

SAS Institute Inc. (1989). SAS software release 6.07. Cary, NC: Author.

Shuey, Audrey, Nancy King and Barbara Griffith (1953). "Stereotyping of Negroes and whites: An analysis of magazine pictures." Public Opinion Quarterly, 17, 281-287.

Standard Rate and Data Service. (1987). Consumer Magazine and Agri media Rates and Data. (69)12. Oak Brook, IL: Author.

Ursic, Anthony C., Michael L. Ursic, and Virginia L. Ursic. (1986). "A Longitudinal Study of the Use of Elderly in Magazine Advertising." Journal of Consumer Research, 13, 131-133.

Venkatesan, M., and Jean Losco (1975). "Women in Magazine Ads: 1959-1971." Journal of Advertising Research, 15(5), 49-54.

Wagner, Louis C., and Janis B. Banos (1973). A Woman’s Place: A Follow-up Analysis of the Roles Portrayed by Women in Magazine Advertisements." Journal of Marketing Research, 10(5), 213-214.

Wolheter, Maralinda and H. Bruce Lammers (1980). "An Analysis of Male Roles in Print Advertisements Over a 20-year Span: 1958-1978." In Jerry C. Olson (Ed.), Advances in Consumer Research: Vol. 7 (pp. 760-761). Ann Arbor, MI: Association for Consumer Research.

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