The main intention of this study was to consider whether and how gender portrayals in advertisements aimed at men differ from those aimed at women, by using specific day parts (daytime, evening primetime and weekend afternoon sports) as a framework for the supposed target audience (women, family and men respectively). Schiebe and Condry (1984) examined advertisements according to product type and found major gender differences in the values promoted in advertisements. For example, advertisements aimed at women stressed beauty and youth while those aimed at men valorised ambition and physical strength.
Craig set about comparing 2,209 television advertisements, taken from three major network channels in the USA, between 6 and 14 January 1990. His main intentions were to 'fill in the gaps' left by similar studies, which tended to neglect the portrayal of men in favour of studying women. He also emphasised that the concept of 'day part' was a major independent variable in this study. Indeed, advertisements are not randomly scattered throughout the broadcast day, but vary according to the age, sex and social situation of the audience the advertiser intends to reach. By definition, 'day part' is the industry term used to describe a period of the day when advertisers consider the audience to be relatively stable demographically. This certainly provides very insightful clues as to how manufacturers and advertisers visualise the consumer.
Method
Craig simultaneously recorded the programming and advertising on the three network channels, which resulted in a total of 30 hours of daytime, 30 hours of primetime and 39.5 hours of weekend television. A total of 664 advertisements were obtained, with each cluster containing between one and seven different advertisements making a total of 2,209 separate advertisement 'types'. Through a series of editing procedures, there emerged a collection of advertisements, randomised by cluster, with all the clues removed as to their original programme context and time of broadcast, The actual coders were not told about the study's focus on day part differences. Craig also emphasised that any repeats of advertisements were coded, since the aim of the study was to examine exposure to stereotypes rather than to specific advertisements.
Categories for Coding and Results
Craig studied a number of features including: characters, primary visual characters, role, product, setting and primary narrator. The results proved very interesting and Craig listed his findings according to the coding categories he used, as follows:
Age and gender of all characters In this instance, each advertisement was categorised in terms of the mixture of age and gender in all characters, irrespective of their importance to the advertisement or their time on-screen (p.202). Craig found significant differences. While the actual percentages of all-male and all-female advertisements were about equal during evening primetime, advertisements for all-adult male characters were more than twice as likely to be found during the weekend (33%) as daytime (14%) broadcast. In the same way, advertisements with only adult females made just 5% of the weekend sample, but 20% of that during daytime hours (p.202). Advertisements with only children or teenage characters made up 6% of the daytime but just 1% of the weekend advertisements (p.203).
Sex and age of the primary visual character Craig found that 27% of the overall sample had no identifiable primary visual character, but an absence of codable primary visual characters was more prevalent in weekend (33%) and evening (32%) advertisements than in daytime advertisements (20%). Children or teens were found to be primary visual characters in only a small percentage of advertisements in this sample. They were more likely to be found in daytime (11%) than in evening (7%) or weekend (5%) advertisements (p.203). During daytime hours, only 40% of the adult primary visual characters were men, but during the weekend 80% were men.
The role of the primary visual character There were found to be large differences in the roles played by primary visual characters in different day parts. In summary, those characters in weekend advertisements were more likely to portray celebrities, professionals and workers than those in daytime advertisements, but were less likely to portray children/teens, home-makers, parents or spouse/partners. Also, these characters were about equally likely to portray 'sex objects/models' in all three day parts (p.302).
Analysis of products advertised While virtually all daytime advertisements fell into either the 'body', 'food' or 'home' categories, only 30% of the weekend advertisements did so. The weekend advertisements were 29% 'automative' and 27% 'business products/services'. For daytime advertisements 44% of the advertisements fell into the 'body' category, while only 15% of the weekend advertisements did so (p.205).
Analysis of primary setting Of the categories depicting a specific location, most of the daytime advertisements were set in the kitchen, 'other room' or 'outdoors away from home'. An analysis of sex differences revealed that women primary characters were proportionately more likely than men primary characters to appear in advertisements in all three 'indoors at home' categories. Daytime men were proportionately more likely to appear in advertisements set in the two outdoor categories as well as at business locations (p.206).
Analysis of primary narrator While male narrators were heard on virtually all weekend advertisements with narrators (97%), a somewhat smaller percentage (91%) were found in the evening advertisements, dropping even further (86%) for the daytime. Also, when women were heard as narrators, it was overwhelmingly for products classified as 'body' (62% in evening and 60% in daytime) (p.207).
Discussion
Overall, Craig concluded that the results of his study supported the hypothesis that advertisements aimed at one sex tend to portray gender differently from advertisements aimed at the other sex. The strategy of day part targeting was seen to go beyond the relatively simple practice of matching a particular product with a particular broadcast time, as it also meant matching a particular image of gender with a particular audience (p.208).
Gender portrayals during primetime are different from those of either daytime or weekends. During primetime, women were more likely to be shown in positions of authority and in settings away from the home than they were during daytime. Men, in contrast, were more likely to be portrayed as a parent or spouse and in home settings during primetime than they were at weekends. Primetime can therefore be referred to as the 'marked' category, since it does not over-employ the use of gender stereotypes. Indeed, the primetime advertisements in this sample were found to represent a more sophisticated and balanced portrayal of gender (p.209).
The magnitude of day part differences in this study was highly significant (at the 0.05 level; p<0.001), consequently suggesting that the television day part cannot be ignored as a variable in studying gender portrayals.