Gender role stereotyping in advertisements on two British radio stations

Adrian Furnham

There has been considerable recent interest in the possible contributions of the mass media to the origins and maintenance of gender roles (Courtney & Whipple, 1974; Culley & Bennett, 1976; Dominick & Rausch, 1972; Furnham, Abramsky & Gunter, 1997; Furnham & Skae, 1996; Kolbe & Langefeld, 1993; O'Donnell & O'Donnell, 1978). Durkin (1985) and others have called for the continuous monitoring of television to chart consistency and change of imagery over time.

McArthur and Resko (1975) found that overall men appeared more often than women in television advertisements and that men and women differed in terms of credibility (men being authorities and women users), role (women portrayed in terms of their relationship to others and men in a role independent of others), location (men shown in occupational settings and women in the home), persuasive arguments (men gave more `scientific' arguments than women), rewards (women were shown obtaining approval of family and males, while obtained men social and career advancement) and product type (men were authorities on products used primarily by women). Manstead and McCulloch (1981) assessed the situation in Great Britain using 170 television commercials and a coding scheme modelled closely on that used by McArthur and Resko (1975) so legitimate comparisons could be made. The overall results were unambiguous and comparable to those of the American study, but the portrayal of men and women on television showed British advertisements at the time to be more gender role stereotyped.

More recent studies have been done, specifically on television advertisements, in Australia (Mazzella, Durkin, Cerini & Buralli, 1992), Kenya (Mwangi, 1996), as well as America (Gilly, 1985; Kolbe & Langefeld, 1993), Canada (Rak & McCullen, 1987), Italy (Furnham & Voli, 1989) and Great Britain (Furnham & Bitar, 1993; Furnham & Skae, 1996). A recent review has compared twelve studies using similar coding categories done on five continents over a twenty year period (Furnham & Mak, 1998). Earlier, Bretl & Cantor (1988) did a content analysis of 15 US television advertisement studies which compared male and female portrayals. Replications over time have shown surprisingly few differences. Ferrante, Hayes and Kingsley (1988) replicated the television based study of Dominick and Rausch (1972), in which they coded six features: the product advertised, gender of the voice-over announcer, gender of the on-camera product representative, setting, age, and occupation of the characters. Although there were some changes (women were portrayed in a wider range of occupations and appeared more frequently outside the home), the authors predicted more changes than actually occurred. Similarly, Lovdal (1989) replicated and extended the study of O'Donnell and O'Donnell (1978) to determine whether the practices of television commercials have change since the late 1970's. There were almost no differences between the 1978 and 1988 data and the authors concluded that there was: "a consistent biased representation of women in the labour force ... Clearly, changes need to be made on a variety of fronts. Increased representations of women in media equality in pay, and employment are only a few" (pp. 722-723).

Other mass communication media have been somewhat neglected by studies of gender role stereotyping. Studies using educational books (Lobban, 1975), picture books (Weitzman, Eiffer, Hokada, & Ross, 1972), and comic strips (Potkay & Potkay, 1984) have shown that men and women are also portrayed in stereotypic fashion suggesting that the media are by and large consistent in their gender role stereotyping. Even less work appears to have been done on gender roles on radio. Surveys have shown that about 36% of a representative sample of the British population listen to three hours of radio or more per day (IBA, 1985), clearly making this a substantial component of many people's exposure to the media. These numbers are fairly constant though listening figures attached to particular radio stations have risen and fallen. Not all of this sample would listen to commercial radio and thus the distribution of those being exposed to radio advertisements is probably very large. Thus, while heavy commercial listeners may hear well over 100 advertisements per day, some may almost never hear them.

A study by Furnham and Schofield (1986) compared the extent of gender role stereotyping in commercials on British radio with that of the content of commercials on television content by Manstead and McCulloch (1981) using the original coding categories. They found that in radio advertisements men were more often portrayed as authorities on products and women as users of products; men were more likely to be portrayed as narrators or celebrities than women; and women were more likely to be portrayed in the home than man. Compared with similar studies of television advertisements there were fewer significant differences in the portrayal of male and female characters. Manstead and McCulloch (1981) found significant differences in nearly all of their eight content-analytic categories, whereas Furnham and Schofield (1986) found significant differences in less than half of the categories and concluded that, compared with advertisements on British television, British radio advertisements were gender role stereotyped on fewer dimensions.

Hurtz & Durkin (1997) have recently pointed out that this has been the only radio-based gender role content analysis. They replicated the study using 100 Western Australian radio advertisements. They found that males were more often central characters; more often in authority roles. Females were most often portrayed in dependent roles and in their home, while they were portrayed as customers or girlfriends in the workplace. Only one of their analyses yielded a different pattern from the Furnham & Schofield (1986) data: in the British data, females were more likely to portray "characters" than males, whilst the reverse was true for the Australian data.

Nearly all of the studies in this area have used a coding frame adapted from the study of McArthur & Resko (1975). The dilemma for the researcher is this: should one update and improve this coding frame to make it more able to capture modern subtle aspects of gender (such as use of humor or cultural allusions), or use the old categories so as to ensure comparability over time. Further, it may be advisable to have media-specific coding frames such that radio is different from television. In this study, the previous used coding schema is retained for comparability though its limitations are acknowledged.

The present study was intended to replicate and update Furnham & Schofield (1986) and to provide a topical replication of Hurtz & Durkin (1997). The aim was two-fold: firstly, to determine whether British gender role stereotyping in radio advertisements has changed since Furnham and Schofield's study twelve years ago; and secondly to see if the amount of stereotyping varies between advertisements on different commercial radio stations. It was expected that the amount of stereotyping has decreased: that is, that fewer than a third of the tests of significance on the eight dimensions would be significant. To replicate the findings, advertisements from two very different radio stations were studied. One was Capital Radio (a London based commercial station), which Furnham and Schofield (1986) used in their study, so that legitimate comparisons between their findings and those arising from the present investigation could be made. Capital is a very popular music station, and is aimed primarily at people under 30 years old. The second radio station was LBC (London Broadcasting Company) which is a station that carries news and current affairs, holds serious discussions on social, economic and political issues, phone-ins and is more "conservative." No data was available on the advertisement profiles of each station, such as the products advertised. However, it did seem that products aimed at younger, less affluent listeners (tapes, CD's, fast food, cosmetics) predominated on Capital, while products for older, more affluent listeners (financial services, computers) were more common on LBC. It was predicted that more sex-role stereotyped advertisements would be associated with the generally older and more conservative audience of LBC than with the audience of Capital.

METHOD

Sample of Advertisements

Radio advertisements from both LBC (London Broadcasting Company) and Capital Radio were recorded during the morning (0700-0900 hrs), afternoon (1400-1700 hrs), and evening (1900-2100 hrs) on each day (Monday to Sunday) over a week in February 1995. The sample obtained from each radio station consisted of 100 advertisements (ranging from 10 to 60 seconds), after all repeats of advertisements and those difficult to code had been excluded. There did not appear to be any relationship between length of advertisement and the sex of the central character. The aim was to achieve the same number of advertisements for each radio station. It was, however, not clear if the number of, or total amount of time, dedicated to advertisements was the same on both channels, though it seemed they were very similar as it took roughly the same amount of time to obtain 108 advertisements from each.

Coding Procedure

Each of the 100 recorded radio advertisements broadcast on the two stations were coded by the two investigators. Both were white English speaking females. They both coded 100% of the advertisements and the Cohen's kappas ranged from .91 to .94. This high level of agreement was similar to both previous radio content analyses. The coding system employed was modelled closely on that used by McArthur and Resko (1975), Furnham & Schofield (1986), and Hurtz & Durkin (1997). The characteristics of each central figure depicted in the advertisements were coded according to the sex of the central figure and the eight variables of credibility, role, location, arguments, reward type, product type, accent, and narrator. Adults playing a central role (namely main speaking parts) in an advertisement were classed as central figures. No more than two central figures were coded for any one advertisement, and, if more than two adults were depicted, those featured most prominently were selected for further coding. Although it did occur, in 10-20% of the time, it was comparatively rare to find an advertisement with no male or no female characters. Certainly as always, more females than males appeared overall in the advertisements, although males were more frequently in the central role.

Coding

Credibility. Central figures were, categorized as "user" when they were depicted primarily as users of the advertised product, while those who were depicted primarily a sources of information concerning the product were categorized as "authority." Central figures depicted as neither use nor authorities were categorized as "other."

Role. Central figures were classified according to one of the following apparent roles: "dependent," meaning primarily financially dependent (spouse, home-maker, girlfriend), "narrator/celebrity," "professional," or "other" (including "worker").

Location. Central figures were categorized according to the location in which they were depicted, either: "home," "occupational setting," or "other."

Arguments. Central figures were categorized according to the type of arguments they presented in favor of the advertised product. Arguments were classified as "scientific" if they contained or purported to contain factual evidence concerning the product or as "nonscientific" if they simply consisted of opinions or testimonials, and as "none" if the central figure offered no argument.

Type of reward. Four categories of reward were coded: "self-enhancement" where the purported benefit of the product was an improvement in health or appearance, "practical" where the purported benefit was a saving of time or effort, or where the main emphasis was on the relative in expensiveness of the product, "social or career advancement" where it was suggested that ownership of the product would assist progress in some social or occupational hierarchy, "other" where the rewards could not be coded in any of the above (including `family approval' and `fun/enjoyment'").

Type of Product. Four categories were coded: "Body/Home/Food" where the product or service involved bodily health, hygiene, cleansing, the home or housework, food, and drink, "Auto/Technical/Occupational" which included automobiles and accessories, and technical and occupational products; and "other" if none of the above categories was applicable.

Accent. The accent used by each central figure was coded into two categories "standard" when an English BBC type accent was spoken, or "other" if it was any other accent (including "Londoner" and "regional").

Narrator. Central figures were categorized according to whether they portrayed a character, ("character") or narrator/presenter ("neutral").

RESULTS

The total number of male and female central figures for both radio stations were calculated, and the totals in each category of the variables. From the advertisements on the London Broadcasting Company (LBC), 128 central figures were recorded, 80% were male and 20% female. From the Capital advertisements 126 central figures were recorded, 78% male and 22% female. Table I shows the main findings concerning the relationship between sex of central figure and the categories of each of the dependent variables. It shows a number of differences in frequency between male and female central figures in certain categories, especially in the "Reward" and "Product" variables. A chi-squared test was performed on the data. Three tests were performed on the frequencies for each variable, testing the significance firstly between male and female central figures on LBC secondly between male and female central figures on Capital and thirdly between the female central figures in advertisements on LBC and Capital. Combined percentages were also calculated from the frequencies so that the more generally able figures could be compared more easily to studies from other regions.

Table I. Gender of Central Figures in the Major Categories of Variables

                                                      LBC

                                                Men       Women
Variable                Category             (n = 102)   (n = 26)

Credibility   User                              17          17
              Authority                         79          19
              Other                              6           0

Role          Dependent                          0           2
              Narrator or celebrity             76          17
              Professional                      13           2
              Other                             13           5

Location      Home                               0           2
              Occupational setting              18           4
              Other                             84          20

Argument      Scientific                        60          13
              Non-scientific                    20           4
              None                              22           9

Reward        Self-enhancement                   2           5
              Practical                         38           6
              Social or career advancement      35           8
              Other                             27           7

Product       Body, home or food                10           9
              Auto or technical or              54           9
                occupational
              Entertainments or leisure         21           5
              Other                             17           3

Accent        Standard                          84          21
              Other                             18           5

Narrator      Character                         19           4
              Neutral                           83          22

                                     Capital

                                 Men       Women
          Category             (n = 98)   (n = 28)

User                               7          2
Authority                         82         22
Other                              9          4

Dependent                          0          0
Narrator or celebrity             85         22
Professional                       4          0
Other                              9          6

Home                               0          0
Occupational setting               1          1
Other                             97         27

Scientific                        66         13
Non-scientific                    25         13
None                               7          2

Self-enhancement                   4          5
Practical                         31          5
Social or career advancement       8          5
Other                             55         13

Body, home or food                16         11
Auto or technical or              30          3
  occupational
Entertainments or leisure         42          8
Other                             10          6

Standard                         108         27
Other                             80          1

Character                         15          5
Neutral                           83         23

Credibility

Both males and females were more likely to be shown as "authorities" than "users." For the advertisements on Capital there was no difference between the portrayal of male (84% authorities, 7% users) and females (79% authorities, 7% users) portrayal ([chi square] = 0.61, ns). There was a small but nonsignificant difference in the LBC advertisements, as 17% of the males were portrayed as users and 79% as authorities, whereas 27% of the females were portrayed as users and 73% as authorities ([chi square] = 1.10, ns). Likewise no significant difference was found between the portrayals of female or male central figures in the advertisements on the two stations for this variable.

Role

Results indicated that in the advertisements on LBC, females were far more likely to be portrayed in dependent roles than males, with 8% of the women and 0% of the men in such roles. Males were also significantly more likely to have a role of narrator/celebrity than females were (75% vs 65%) ([chi square] = 4.08, p [is less than] .05). There were no significant differences in the such portrayals on Capital however. No males or females were depicted as dependent, and 88% of the males and 79% of the females were portrayed as narrator/celebrities ([chi square] = 0.00, ns). Although a significant association between gender of central figure and role was found in advertisements on LBC and not on Capital, no significant difference overall was found between the two radio stations portrayal of female or male central figures for this variable.

Location

Analysis of 2 x 3 contingency tables were carried out but no significant difference in location was found between males and females in advertisements on LBC (X = 1.09, ns) or on Capital ([chi square] = 0.68, ns), or in the portrayal of female central figures between the advertisements on the two radio station ([chi square] = 4.6, ns). Nearly all the locations were coded as "other."

Argument

Analysis indicated there was no significant association relating the sex of central figure to type of argument for advertisements on LBC ([chi square] = 1.10, ns) or for Capital ([chi square] = 4.52, ns). However, on LBC, males were more likely to use scientific arguments than females (73% vs 59%) and on Capital males were more likely to use scientific arguments (67%) than nonscientific (26%), whereas women were equally likely to use either (46% and 46%). No significant difference in the type of argument used by female or male central figures was found between the two radio stations.

Reward

A significant association between the sex of the central figure and type of reward was found for advertisements on both radio stations. On LBC, women were more likely than men to be shown suggesting self-enhancement as a reward (19% vs 2%) and men more likely to be shown suggesting practicality (38% vs 23%), ([chi square] = 1293, p [is less than] .005). On Capital, 18% of the women suggested self-enhancement versus 4% of the men, and 32% of the men suggested practicality against 18% of the women ([chi square] = 9.81, 12 [is less than] .05). There were no significant differences between stations' portrayal of males and females on the two different radio stations.

Product

On LBC, men were less likely than women to appear in advertisement for products concerning body, home, food (10% vs 35%) but more likely to appear in advertisements for auto, technological or occupational products (53% vs 35%). These differences were significant ([chi square] = 10.13, p [is less than] 0.03). On Capital, men were less likely than women to appear in body, home, food advertisements (16% vs 39%) but more likely to appear in auto, technical, occupational advertisements (31% vs 11%) and advertisements for entertainment and leisure (43% vs 29%). These differences were significant ([chi square] = 11.9, p [is less than] .01). No significant difference was found between males and females portrayed on the two stations for this variable.

Accent

There was no difference between accents used by male and female central figures in advertisements on LBC. 82% of the males and 80% of the females had a standard (British) accent ([chi square] = 0.03, ns). The same was found on Capital radio--98% of the men and 96% of the women had standard accents and 2% of the males and 4% of the females had other accents ([chi square] = 0.69, ns). The difference in the accent used by female or male central figures between the two radio stations was not significant.

Narrator

No difference in the portrayal of different sex central figures was found in advertisements on LBC, 81% of the men and 85% of the women were neutral ([chi square] = 0.15, ns), likewise on Capital, 85% of the men and 83% of the women were portrayed as neutral ([chi square] = 0.00, ns). There was no significant difference between males or females on the two radio stations portrayal in this variable.

Finally, the data from the two radio stations was combined and a simple [chi square] run on the gender role differences. In all, only three were significant: role: ([chi square] = 7.02, p [is less than] .05), where males were more likely to be the narrator or celebrity; reward: ([chi square] = 18.97, p [is less than] .001), where it was more likely to be practical for males and other for females; product ([chi square] = 18.14, p [is less than] .001), where it was more likely to be body, home or food for females, and auto, technical or occupational for males.

These results do not support the hypothesis that sex-role stereotyping has decreased in radio advertising since Furnham and Schofield's 1986 study, nor that any differences exist between the two radio stations. However, the dimensions upon which the gender roles differ are somewhat different to those from the two published studies in the area (Furnham & Schofield, 1986; Hurtz & Durkin, 1997).

DISCUSSION

The results show that on two specific criteria, men and women were portrayed in significantly different ways on British radio advertisements. Overall (i.e., on both radio stations) men were more often portrayed suggesting practical and social career advancement as a reward for product purchase, and women as suggesting self-enhancement as a reward for the product. Men were more likely to be portrayed in advertisements for automobile, technical and occupational products and women more likely to appear in advertisements for body, home and food products. In addition, on LBC, women were more likely than men to be shown in dependent roles, but this was not true on Capital. There was one difference in portrayal between the two stations; women were more likely to use scientific arguments in advertisements on LBC than on Capital.

Compared to earlier comparable studies, there were fewer significant differences in the portrayal of male and female characters. Furnham and Schofield (1986) found sex differences in less than 40% of the coding categories: credibility of the central figure (user vs authority), the role of the central figure (dependent vs narrator/celebrity), and location (home vs occupational setting). Hurtz & Durkin (1997) also found significant sex differences for credibility basis, role and location, very similar to Furnham & Schofield (1986). This study yielded significant differences in only two categories (25%) from the same radio station, namely Capital, and these were on different categories than the ones found by Furnham and Schofield (1986). In Furnham & Schofield (1986), 83% of central figures were male, while in the study on the same radio station, 78% of the central figures were male. Using this as a criteria, there seems to be relatively little decline in gender role stereotyping over this period.

It also seems that the sex-roles stereotypes that existed may have shifted, rather than significantly declined. Although sex differences in the role of the central figure were found in advertisements on LBC, none of the other differences found were the same as those found by Furnham and Schofield (1986) on Capital. The stereotyping that does exist appears to have moved from the nature of the central figure and the location of the advertisement to the arguments used to buy the product. Certainly the latter seem more subtle than the former.

It was predicted that the amount of gender role stereotyping in advertisements would vary between radio stations because the target audience was different. More specifically, it was predicted that LBC would have more advertisements with gender role stereotypes than Capital because of their audience is generally older and more conservative. However, the evidence was inconclusive. On one hand, on advertisements on LBC women were significantly more likely than men to be shown in dependent roles, whereas advertisements on Capital did not portray this gender role stereotype. On the other hand, the only significant difference in portrayal between the two radio stations was that females were more likely to use scientific arguments in advertisements on LBC than on Capital, showing that Capital has more advertisements imposing gender role stereotyped in this category. This may be due to the fact the LBC is a more intellectual, political and business news station so is more likely to have advertisements that use scientific arguments to sell their product no matter which sex the central figure is. This is, however, purely speculative, and warrants empirical confirmation. The conflicting evidence leads us to reject the hypothesis that LBC broadcasts more sex-role stereotyped advertisements than Capital. Hurtz & Durkin (1997) note that research such as this on the media portrayal of gender roles does not necessarily demonstrate an effect of the medium on society, as there are inevitably many other social, economic and political factors that influence gender role development, portrayal and understanding. Further, audiences are selective in terms of when they listen to which station/channel, and why.

However, there is evidence from time-of-day studies that suggest that advertisers are acutely aware of their audience and may portray gender roles differently as they think appropriate (Craig, 1992; Furnham, Abramsky & Gunter, 1997; Harris & Stobart, 1986; Hurtz & Durkin, 1997). The very slight reduction and change of gender role stereotypes on British radio advertisements may be due to several reasons. It could be that market demands and advertisers' preferences have changed, or that different products are now advertised on the radio for which sex-role stereotypes are less relevant. Certainly, sex role stereotyping in television seems on the decline and it may be that radio is lagging behind in these changes (Furnham & Skase, 1997). However, one really needs comprehensive longitudinal data to substantiate these claims.

REFERENCES

Bretl, D., & Cantor, J. (1988). The portrayal of men and women in US television commercials: A recent content analysis and trends over fifteen years. Sex Roles, 18, 595-609.

Courtney, A. E., & Whipple, T W. (1974) Women in TV commercials. Journal of Communication, 24, 110-118.

Culley, J., & Bennett, R. (1976). Selling women, selling blacks. Journal of Communication, 26, 160-174.

Craig, R. (1992). The effect of television's daily part in gender portrayals in television commercials: A content analysis. Sex Roles, 26, 197-211.

Dominick, J. R., & Rausch, G. E. (1972). The image of women in network TV commercials. Journal of Broadcasting, 16, 259-265.

Durkin, K. (1985). Television and sex roles acquisition: 1. Context. British Journal of Social Psychology, 24, 101-113.

Ferrante, C., Hayes, A., & Kingsley, S. (1988). Image of women in television advertising. Journal of Broadcasting and Electronic Media, 32, 231-237.

Furnham, A., & Bitar, N. (1993). The stereotyped portrayal of men and women in British television advertisements. Sex Roles, 29, 297-310.

Furnham, A., & Mak, T. (1998). Sex-role stereotyping in television commercials: A review and comparison of twelve studies done on five continents. Paper under review.

Furnham, A., & Skae, E. (1997). Changes in the stereotypical portrayal of men and women in British television advertisements. European Psychologist, 2, 44-51.

Furnham, A., & Schofield, S. (1986). Sex role stereotyping in British radio advertisements. British Journal of Social Psychology, 25, 165-171

Furnham, A., & Voli, V. (1989). Gender stereotypes in Italian television advertisements. Journal of Broadcasting and Electronic Media, 33, 175-185.

Furnham, A., Abramsky, S., & Gunter, B. (1997). A cross-cultural content analysis of children's television advertisements. Sex Roles, 37, 91-99.

Gilly, M. (1988). Sex roles in advertising: A comparison of television advertisements in Australia, Mexico and the United States. Journal of Marketing, 4, 75-85.

Gunter, B. (1985) Attitudes to broadcasting in 1984. London: Independent Broadcasting Authority.

IBA. (1985). Attitudes to broadcasting in 1984. London: Independent Broadcasting Authority.

Hurtz, W., & Durkin, K. (1997). Gender role stereotyping in Australian radio commercials. Sex Roles, 36, 103-114.

Harris, P., & Stobart, J. (1986). Sex-role stereotypes in British television advertisements at different times of the day. British Journal of Social Psychology, 25, 155-164.

Kolbe, R., & Langefeld, C. (1993). Appraising gender role portrayals in TV commercials. Sex Roles, 28, 393-417.

Lobban, G. (1975). Sex roles in reading schemes. Educational Review, 27, 202-210.

Lovdal, L. (1989). Sex-role messages in television commercials: An update. Sex Roles, 21, 715-725.

Manstead, A. S. R., & McCulloch, C. (1981) Sex-role stereotyping in British television advertisements. British Journal of Social Psychology, 20, 171-180.

Mazzella, C., Durkin, K., Cerini, E., & Buralli, P. (1992). Sex-role stereotyping in Australian-television advertisements, Sex Roles, 26, 243-259.

McArthur, L. Z., & Resko, B. G. (1975). The portrayal of men and women in American television commercials. Journal of Social Psychology, 97, 209-220.

Murray, J. R, Rubenstein, E. A., & Comstock, G. A. (1972). Television and social behavior, Vol 12. Television and social learning. Washington, DC: Government Printing Office.

Mwangi, M. (1996). Gender roles portrayed in Kenyan television commercials. Sex Roles, 34, 205-214.

O'Donnell, W. J., & O'Donnell, K. J. (1978). Update: Sex role messages in TV commercials. Journal of Communication, 28, 156-158.

Potkay, C., & Potkay, C. (1984). Perceptions of female and male comic strip characters: II Favorability and identification and different dimensions. Sex Roles, 10, 119-128.

Rak, D., & McCullen, M. (1987). Sex role stereotyping in television commercials: A verbal response mode and content-analysis. Canadian Journal of Behavioural Science, 19, 25-39.

Weitzman, L., Eiffer, D., Hokada, E., & Ross, C. (1972). Sex-role socialization in picture books for preschool children. American Journal of Sociology, 77, 1125-1150.

Adrian Furnham(1) and Louise Thomson University College London

(1) To whom correspondence should be addressed at Department of Psychology, University College London, 26 Bedford Way, London, WC1E 6BT.

COPYRIGHT 1999 Plenum Publishing Corporation
COPYRIGHT 2000 Gale Group