The Portrayal of Men and Women in British Television Advertisements: A Review of 7 Studies Published Over a 12 Year Period

A good deal of this literature was inspired by McArthur and Resko’s [2] study analysing American television commercials. The coding categories devised for the content analysis in this paper formed the basis of nearly all subsequent research and this modest study has to date over 70 citations.The essential methodology of the study was replicated in other countries [12,13]. Furnham and Mak [1] reviewed 14 studies done up to the end of the century. They found evidence of cross-cultural/national similarly in terms of stereotyping on such categories as credibility, age and argument-made while others like location showed much greater variety. By-and-large, sex-role stereotyping was more prevalent the more traditional the country in terms of social and religious values.


Introduction
Over the years there has been a lot of research into the portrayal of men and women in television commercials [3][4][5][6]. There have also been attempts to review these studies at various points [7][8][9]. The research area remains very active [10,11].
A good deal of this literature was inspired by McArthur and Resko's [2] study analysing American television commercials. The coding categories devised for the content analysis in this paper formed the basis of nearly all subsequent research and this modest study has to date over 70 citations.The essential methodology of the study was replicated in other countries [12,13]. Furnham and Mak [1] reviewed 14 studies done up to the end of the century. They found evidence of cross-cultural/national similarly in terms of stereotyping on such categories as credibility, age and argument-made while others like location showed much greater variety. By-and-large, sex-role stereotyping was more prevalent the more traditional the country in terms of social and religious values.
The latest review looked at 30 studies published since the year 2000 [14] which examined over 8000 advertisements shown in 20 countries. They noted that whilst it was possible to show many cross-national differences it was difficult to ascertain whether portrayals changed within countries over time. This study attempts just this In a very comprehensive review of 64 studies Eisend [15] found some evidence of a decrease in gender role stereotyping in television advertising over time but only in particular high masculine countries. More importantly in his study he was able to provide empirical evidence to test the mirror (advertising mirrors socio-cultural values) vs the mould (adverstising moulds gender beliefs and stereotypes) argument. He concluded that: "...gender stereotyping in advertising depends on gender-related developments and value changes in society rather than the other way around". This paper compares the data from seven studies done in Great Britain that compared advertisements on similar channels over a 10 year period. This sort of analysis allows for investigating trends in the area. There appears to be no other data set that allows for this sort of comparison across time. It is hypothesized that as societal and social attitudes change so too do television advertisements. That is, the more societies do not use sex-role stereotyping in everyday life around personal lifestyle and business issues, so television reflects this. Table 1 shows the data from 7 studies all using essentially the same methodology. They were published over a 12 yr period and most had another country comparison group. If it is true that gender role stereotyping throughout British society is in decline it may be expected that studies should show fewer differences; that is, television advertisements mirror what occurs in society.

Central figure
There was a tendency for there to be more men than women as the central figure in television advertisements in Britain. Only one study by Furnham and Saar [16] found the opposite effect with around 60% of the central figures being women. Most others were around the same amount, with one study by Skoric and Furnham [17] that had double the amount of men (N=122) compared to women (N=61) as central figures. This therefore showed essentially no change.

Mode of presentation
Most studies have found that men are more likely to be used as voice-overs in advertisements while women were used more visually. Skoric and Furnham [17] found 72% of men as voice-overs while only 41% of women. Furnham and Hussain [18] found 76.2% of women were presented visually while 46.7% of men were presented visually. Most studies have generally found about a 20% difference between the two sexes with males being more common as voice-overs and females more commonly presented visually. Only Furnham and Imadzu [19] found women to present end comments more than men and this was not a large difference (and not significant) as it was 59.2% of women and 55.9% of men. Britain, 1993 [21] Britain, 2000 [20] Britain end comments that were made by each gender. Furnham and Farragher [20] found that 13.6% of males and 13.4% of females made an end comment while Skoric and Furnham [17] found that 84.4% of males and 70.5% of females made end comments. There was considerable variability between which gender made more end comments than the other. Furnham and Bitar [21] found that 75.2% of males, compared to 32.6% of females, made end comments while Furnham and Hussain [18] found that 66.7% of males and 71.4% of females made end comments. However, when there was a difference between the amount of male and female end comments, there was usually a narrower range when females had more end comments and there was no significant difference. For example, Furnham and Saar [16] show that 42.1% of males made end comments while only 19.4% of females did and Furnham and Spencer-Bowdage [22] found that 59.5% of females and 53.4% of males made end comments. This was seen in the majority of studies in Britain that were analysed and this pattern has not changed much.

Credibility
In every study analyzed, women were always seen more in the role of a user of the product than men were. Most studies have shown a large and significant difference between men and women in the role of the user. Furnham and Spencer-Bowdage [22] found men were presented as users 32.8% of the time while women were in 67.6% of advertisements. Furnham and Saar [16] found this number jump for both men (63.2%) and women (93.5). Yet there was still a 30% difference between the two which has generally remained stable through time. Men were almost always seen more as an authority figure, except for two studies by Furnham and Farragher [20] and Furnham and Imadzu [19]. Furnham and Farragher [20] found men (11.1%) and women (13.4%) were shown as authorities approximately the same percentage of time in the advertisements that they analysed. Furnham and Imadzu [19] found similar results with males (39.9%) and females (39.2%). There does not seem to be a clear difference across time and there has generally been a 20-30% difference between men and women. This difference is found in the role of the user and the authority -women remain users, men in the role of experts or authorities.

Role
All studies showed women in dependent roles more often than males in advertisements. Furnham and Bitar [21] found a high difference between men (17.0%) and women (60.6%). This has been seen more recently as well in Furnham and Spencer-Bowdage's [22] study where men (27.6%) were also much less likely to be in a dependent role than women (62.2%). Men have also been shown to be in more autonomous roles and this hasn't gotten better through time. Furnham and Farragher [20] found only about a 10% difference between men (69.1%) and women (58.5%) while Furnham and Saar [16] found almost a 40% difference between men (57.9%) and women (19.4%). A higher proportion of men tend to be shown as 'interviewer/ narrator' but this gap has narrowed in Britain since 1993. The most recent studies have found the same proportion of men and women central figures in this type of role. Furnham and Hussain [18] found the closest results between men (64.4%) and women (64.3%) and Furnham and Saar [16] found similar results with men (31.6%) and women (32.3%).
It should be noted that this content analytic category has changed a little as well. Most studies have broken the 'autonomous' label down even further to separate options of 'interviewer/narrator' and 'professional.' There has been a narrowing of the gap, with regard to dependent roles, through the years from there being a 40% difference in Furnham and Bitar's [21] study to less than a 20% gap in Furnham and Hussain's [18] study, while Furnham and Imadzu [19] found under a 5% gap between men and women.

Age
Women in the advertisements analysed are more likely to be younger than older. This has changed over time. In earlier studies women were seen more often in "middle-aged" groups. Furnham and Bitar [21] showed 72.1% of women in this group, which is the highest it has been through all studies. Since then, women have become even younger when portrayed in advertisements. Furnham and Farragher [20]found women in the young-age group 34.1% of the time, Furnham and Spencer-Bowdage [22] saw this rise to 64.3% and Furnham and Saar [16] found similar results with 67.7%.
Men are more likely to be in middle-aged group than women and this is also the most common age for men that are portrayed in advertisements. Most studies have shown at least 60% of men in the middle-aged group. Furnham and Farragher [20] found 60.5% of men in this age group while Skoric and Furnham [17] found this rise to 91.8%. Across all studies and time in Britain, there are not many advertisements that include older men and women (above the age of 45), with most studies finding older central figures in less than 10% of all advertisements looked at. Furnham and Imadzu [19] have showed these figures at 5% for men and 2.5% for women. Most other studies show similar results.
All studies from all countries in fact show consistency in results (both across country and time) and is quite predictable from the results seen over the years. There is also not a drastic difference in sex of the individual portrayed if they are in the oldest age group category. Over time, there has also been a trend that the central figures, women specifically, have been younger than advertisements in some of the earlier studies.

Argument
Only one study has suggested that females are less likely to present an argument for purchasing or using the brand or service. Furnham and Spencer-Bowdage [22] found 56.8% of females, compared to 22.4% of males, did not make any type of argument. Generally there has not been a significant difference in the proportion of types of arguments that men and women present. Most studies have shown under a 10% difference and even more under a 5% difference with the type of argument made. Only two of the studies suggested possible differences between men and women's arguments. Furnham and Spencer-Bowdage [22] found 53.4% of men and only 29.7% of women made factual arguments while Furnham and Hussain [18] found 57.8% of men and 21.4% of women presented factual arguments. The results were similar for opinion arguments as well. Generally there was not a large difference but the two most recent studies found differing results. Furnham and Hussain [18] found 66.7% of women, compared to 33.3% of men, presented an opinion argument. Furnham and Saar [16] found the opposite effect where only 12.9% of women, compared to 31.6% of men, made opinion arguments. It appears not to be the case that at any point men gave rational, logical, scientific arguments while women gain a-rational, emotional argument for the product.

Reward type
Women have been associated with 'social/self-enhancement' rewards when compared to men. Furnham and Imadzu [19] found 35.8% of females and 14.7% of males were associated with this type of reward. All other reward types have shown mixed results (i.e. Practical and pleasure). Some studies suggest that pleasure rewards have been slightly more associated with males. While most showed near equal percentages, Skoric and Furnham [17] found males to be much more likely to be associated with pleasure rewards (51.6% vs. 8.2%). Furnham and Hussain [18] also found 68.9% of males, compared to 19% of females, who were associated with pleasure rewards. There were also gender differences with practical rewards. Furnham and Imadzu [19] found males (26.6%) to be more likely to be associated with practical rewards than women (15.8%). On the contrary, Skoric and Furnham [17] found women (62.3%) were more likely than men (40.2%) to be associated with practical rewards.

Product type
Women have been more likely to advertise body products than men. Skoric and Furnham [17] found 36.1% of women and only 10.7% of men advertised body products. This is also the type of product that women are advertising the highest proportion of time when compared to other products they advertise. Furnham and Imadzu [19] found 29.2% of advertisements with women to be for body products, almost double the other product types: 14.2% for home, 16.7% for food, and 15.0% for services. Women were also more likely than men to advertise home/domestic products. While 36.1% of females in Skoric and Furnham's [17] study advertised home products, only 13.9% of males did so. Men were more often shown advertising food products, when compared to women. The difference was not as noticeable as with body products but there was still an observable difference. Furnham and Spencer-Bowdage [22] found 22.4% of men and 10.8% of women advertising food. Not as many studies looked at auto/sports, however, the ones that did found that men were more likely to advertise them. Furnham and Farragher [20] showed 16% of men with auto/sports products, and only 2.4% of women. Sex differences in this area have remained constant even though some new products have emerged and others disappeared from television advertisements

Background
Women have been shown more often than men with children in the background. Women are also generally seen more often with mostly female backgrounds. Men are generally seen more than women with a mixed male/female background. However, both men and women are still shown with no background at all about 30-40% of the time. Furnham and Saar [16] have found this figure to be the highest with 47.4% of males and 45.2% of females with no background.

Location
Across all studies, women are generally shown in the home about 40-50% of the time while men only are about 15-20% of the time. Furnham and Saar [16] found the lowest percentage for men where 5.3% of men, compared to 41.9% of women, were presented in a private residence. Skoric and Furnham [17] found 12.3% of men and 36.0% of women in private residence settings, which was more typical across the other studies. Men are more likely to be shown in occupational settings but the difference was not as significant as in private residence. Furnham and Saar [16] found 21.1% of men and 6.5% of women in occupational settings. Leisure was also another setting that offered differences between men and women. While Furnham and Spencer-Bowdage [22] found women (2.7%) in a leisure setting more often than men (0%), all other studies found the men more likely to be in this setting. Furnham and Farragher [20] found 50.6% of males and 22.0% of females in leisure settings while others like Furnham and Hussain  [18] offered a smaller difference between males (46.7%) and females (33.3%). Overall it seems the largest and most stable differences are seen in private residence setting.

Discussion
The seven studies in this analysis were done over a 10 year period, which allowed for testing of the mirror/mould alternative hypothesis suggested by Eisend [15].The studies contained advertisements from different-channels and sent out at different times of day but from the same country. There is therefore a reasonable amount of 'noise in the system' to study sex role stereotypes over time. Yet there seemed very little evidence of changes in the way men and women were portrayed in television advertisements at least in Great Britain. Women tend to be portayed as younger than men, users not authorities and in the home. As has been the case cross-culturally for over 30 years men are still the central character in most television advertisements and they do more voice-overs. They are portrayed as more rational and authoritative than women. There seems little evidence of systematic linear changes in any of the twelve categories.
There could be many different explanations for this. First, little has actually changed over the past decade with respect to sex-role and gender stereotypes in society. Major changes have occurred in public beliefs and behaviours with respect to sex and gender but these have occurred gradually over the past fifty years. In other words this is too short a time period to expect to see changes. Second, it may be that advertisers tend to be rather conservative in the way the sexes are portrayed, not keeping up to date with changes that have actually occurred in society. That is, there is a time lag effect such that television advertising takes time to "catch up" on the changes in society.
Third, it may be the advertisements have changed in the use of humour, word-play or other subtle visual and verbal cues that are not picked up by the coding scheme used here. In this sense the problem lies in the out-of-date coding system used in these studies and that if this was updated and made more subtle, changes would become noticeable. Fourth, that there are systematic,consistent and explicable changes in sex role portrayal but these are related specifically to ether specific products or channels or time-of-day when those advertisements occur. In this sense only a very much finer grain analysis would yield any consistent findings.
There is also an issue with some coding categories like the product category or type. It is quite clear that there are universal patterns in how advertisers would choose to advertise products like soap or beer; fast foods or cars. Aronovsky and Furnham [23] looked specifically at advertisements for food products in Britain. They found similar sex differences in daytime and evening commercials with very clear and stereotypic differences in role, age, product appeal and end product. Furnham and Li [24] looked at food and beverage advertisements aired in Hong Kong. They found more evidence for sex role stereotypes in Western made as opposed to Chinese made advertisements but overall relatively little evidence of sex-role stereotypes. Certainly examining sex role stereotypes for advertisements for very specific categories of products or services seems the way to proceed in this area.
A related issue concerns how television has changed over the years and whether advertising has indeed changed but in other media. Perhaps the most obvious change has been the growth of channels available in most countries as well as relative ease through satellite connections to receive 'foreign' television from other countries. This has fragmented the national audience and seen the growth of very specific channels specialising in everything from news to sport, and weather to classical films. Most of these are commercial and are liberally interspersed with advertisements. The question is how advertisements differ as a function of channel. This is no doubt a past function of the size and demography of the audience which would no doubt relate to products advertised. However it remains of interest how sex-roles are portrayed perhaps on channels clearly aimed at quite different audiences such as shopping and cooking channels vs science, history or geography channels. This would be a most fruitful area for future research.
There continues to be sustained research interest in this topic with papers appearing regularly from many different countries [10,25,26] Most studies are descriptive rather than hypotheses testing. What the research area still lacks is the development of more subtle and upto-date coding categories as well as theoretical frameworks to make specific predictions about cultural or temporal differences. Changes in the television industry and in the development of new (and interactive) electronic media represent a new opportunity to study how the genders are potrayed.
This study made it possible to examine how gender roles in advertising changed over a 10 year period in Great Britain. The results showed very little change over this period despite changes in gender roles in society as a whole. It is probable that if Eisend [15] is correct in his analysis, namely that advertisements mirror society, subtle changes will occur in the way men and women are portrayed in television commercials but this will take longer than the decade of material examined in this study. Further, those changes will be subtle requiring researchers to regularly update their coding categories to understand subtle changes and nuances which are characteristic of short, expensively made television advertisements.