Gender and Race in TV Commercials

Television commercials have long been studied with a look at how they portray different races and genders (Furman and Mak, 1999). Unlike most programming, advertisements are produced with the specific goal in mind of trying to influence viewers' opinions on a product or service, either directly or by subtle associations. Commercials can also be a snapshot of cultural attitudes at a given time, and may even have a slight hand in shaping those attitudes by the way they portray different people (Coltrane & Messineo, 2000).

As society - and more importantly to advertisers, the market - begins to recognize the contributions from all segments of the population, the content of commercials will continue to evolve. However, the marketing principles of targeting and positioning will likely mean certain people will always be more aggressively advertised to, for the product that person is most likely to buy. So, while the true attitudes of society in general and the effects of commercial advertising can not realistically be known, the type of messages being sent out and their frequency can be studied.

Background

An hour of television programming can contain about 15 minutes of advertisements. In as little as 15 seconds each, effective television commercials are designed to influence viewers' opinions on an advertised product. Though it's impossible to know the extent of commercials' influence, repeated exposure to certain messages may have some affect on people's views, not only of the product but also their view of people and society (Coltrane & Messineo, 2000).

Most studies of gender and race portrayals in TV commercials have found that women and minorities were underrepresented in television commercials until recently, when representations have become closer to the actual composition of the population, with the exception of non-black minorities (Furnham & Mak, 1999).

However, most of the studies also show that women are still shown in mostly dependant roles with men as authority figures (Craig, 1992), while minorities are frequently portrayed according to stereotypes (Coltrane & Messineo, 2000).

Studies usually focus on commercials shown during prime-time programs, though studies on different times of day and different target-audience programming yielded more varied results.

The difficulty with most of these studies comes from the way the commercials are coded; traditional coding methods only acknowledged a single spokesperson and couldn't accurately represent supporting characters (Bartsch, et al., 2000). Also, studies focused on the setting and roles of the characters as indicators of their portrayals, without regard to actual content. According to these methods of classification (Kim & Lowry, 2005), a commercial for a cleaning product could show an incompetent man not knowing how to use a product and a women being the authority, but the woman would be considered as dependant since she would end up doing the chores.

This aspect of competence has never been studied, but it certainly seems that the majority of men in commercials today are portrayed as childish or ignorant. It also seems that women might now be overrepresented, as are blacks over other minorities. Another shortcoming of earlier studies is the reliance on major network channels, while cable channels with specific target audiences were never included. The purpose of this study is to add these factors into an up-to-date look at how gender and race are portrayed in television commercials.

Review of literature

A 1999 study by Adrian Furnham and Twiggy Mak examined 14 studies that had been done over a 25-year period from multiple countries, including three studies from the United States. Those studies were from McArthur & Resko (1975), Bretl & Cantor (1988), and Craig (1992). The categories that could be compared to all the studies were Role, Location, Age, Argument, and Product Type. The results of the comparison of the American studies revealed that women were shown as the primary users of products, dependant, and part of a family, while men were shown as the primary authorities of products, on the job, and more likely single. The gaps between gender portrayals were found decreasing in the most recent study in the comparison.

Craig's 1992 analysis of television commercials specifically examined the question of gender portrayals in regard to the time of day the commercial aired, to see if women were portrayed differently during daytime programs aimed at women versus general programming or shows aimed at a male audience.

Craig's method took daytime samples from 2-4 p.m. to represent women's programming, primetime between 9-11 p.m. to represent general programming, and weekend sports programs to represent men's programming.

However, in labeling these time frames as such, the study itself is acknowledging that daytime TV watchers are a primarily female audience. Women who work were obviously not considered in the sample target audience. The major flaw is that study's conclusion - that commercials for household products aired during that time are sexist, since they're only selling cleaning products to women; in reality, the target audience is simply those who are staying at home, so it only stands to reason that domestic products would be advertised more during the day. While the method of sampling different times of day is valuable to get a better sample of commercials, drawing any sort of conclusions based solely on the supposed target audience is pointless.

Scott Coltrane and Melinda Messineo tried to overcome this problem in their 2000 study by using Arbitron ratings in southern California to find the highest rated television programs by demographic and then choosing shows that would appeal to a certain age, sex, or race. This method has serious flaws as well, not the least of which was choosing the Superbowl and coding it as a male-audience program. The reasoning and justification behind the programs they chose for what audience is faulty at best and rudely ignorant at worst.

Another study in 2000 (Bartsch, Burnett, Diller, & Rankin-Williams) concluded that new complex types of commercials would require a more refined coding system than what had been use up to that point, as background characters and other influences were not taken into account, and a main product representative was often difficult to pick out. As the structure of commercials change, so must the methods to study them.

The solution to finding the correct target audiences is simple: Take samples from cable channels that specifically target certain audiences from multiple times of day, and combine them with the main four broadcast channels to get a representative sample of what's being shown across the television spectrum.

By updating the coding method and categories, this study will also present a more accurate look at modern commercial content than previous studies. The findings are expected to show the gap closing between men, women and minorities, and perhaps even tipping the other way.

Hypotheses

H1: Female and black characters will be represented equal to or greater than their ratio in the U.S. population.

H2: Women will be shown more often as professional businesspeople than as housewives.

H3: Men will be shown more often than women as incompetent, ignorant or foolish.

Method

This study took eight different channels to gather all the commercials that would represent modern advertising. The four broadcast channels were ABC, CBS, FOX and NBC, and the four cable channels were FX, USA, Lifetime and Spike TV. FX and USA were chosen for their general appeal and original programming, while Lifetime and Spike TV were chosen to represent women's and men's targeted programming, respectively.

Three time segments were used, with three hours in each segment. 7 a.m. to 9 a.m. was coded as morning programming, 1 p.m. to 3 p.m. was coded as daytime programming, and 7 p.m. to 9 p.m. was coded as prime-time programming. From each three hour segment, two hours were randomly chosen for each channel.

To ensure a complete representative commercial sample from each of the eight channels, eight days of samples were collected, so that no two hours of the same segment were taken from the same day, until six different hours were recorded for each of the eight channels. This would yield 48 hours of programming, more than the usual amount sampled in most studies of this kind.

Programming announcements, movie advertisements, public service announcements and purely text ads were disregarded. Local commercials were also discounted since the purpose of the study is to analyze national trends and the possible perpetration of stereotypes; besides being overly localized, local commercials are often made out of necessity, employing either the founding family or, even if actors are involved, lacking the time, money, and foresight that go into national marketing campaigns. Also, as in Kim and Lowry's (2005) and Craig's (1992) studies, duplicate commercials were included, since examining total frequency in relation to exposure is more important to the study than specific commercials.

Commercials were then coded on five categories, with two categories later discarded. The three useful categories are:

Representation: Characters in each commercial were classified as male or female; white, black, Asian, Hispanic/Latino or mixed/indeterminate; and as main characters (determined by relationship to the product, screen time, and speaking lines) or background characters.

Portrayal: Characters could be portrayed as intelligent (by showing knowledge of the product/service or correcting another character), ignorant (by being corrected, proven wrong, mocked or injured), or neutral.

Product type: The 14 categories of products are domestic (cleaning products, furniture, appliances), body (bath, hair, toiletries), clothing, fast food, restaurant, grocery items (including drinks), autos, equipment (including tools, parts, non-clothing sports gear, etc.), business products/services, electronics/toys, financial/real estate, drugs/medical, alcohol, and entertainment/vacation services.

Results

Two of the categories were coded for, but did not yield useful data.

Appearance: Modeled similar to Bresnahan et al. (2001), this category coded appearance on physical attractiveness, health/activity, projected confidence/aggressiveness, and social/economic status. These factors turned out to be mostly neutral, and even when they were present, value judgments were so subjective that no useful conclusions could be made.

Role: Adapted from Craig (1992), the purpose was to identify roles such as professional, homemaker, spouse, parent, etc. Although H2 would have been proven correct with the amount of coding completed, most commercials lacked any context to determine such roles; characters were often seen in leisure activities or simply in an empty white space with the product. The majority of roles were being coded as undetermined, which devalued the data on roles that could be quantified.

Complex coding questions also arose over multiple roles: if a man was working at a desk in the garage, did that make him a professional or a homemaker? If a woman is wearing a suit but holding a baby, does that means she's primarily a professional businessperson or a mother? Even if it were possible to cross-code so many variables, this study lacked the time and resources necessary.

After coding 1,130 qualifying commercials, a total of 1,829 females and 1,995 males were identified, including voiceovers. This puts men as making up 52 percent of characters in commercials examined, although the split is 50/50 when only considering visible actors, since males are more often just voices (see table 1).

Table 1 also shows the greatest disparities by the type of product the commercial is selling. Women were overwhelmingly featured in commercials for clothes and body products, while men were predominant in commercials for automobiles and equipment. Grocery items and drug or medical products were a large percent of both sexes' appearances.

Excluding voiceovers, men and women appeared almost the same amount overall as main characters, though women were more likely to be secondary characters and men were more likely to be background characters (table 2). Minorities of both sexes were more likely to be secondary, rather than main, characters. Comparing minority appearances to a 2005 Census Bureau estimate, black characters were represented by 12-13 percent of characters overall, just about equal to the black population percentage in the U.S. All other minorities were represented far below their actual ratio in the U.S. population.

Table 3 shows that while both sexes were portrayed as intelligent a near-equal amount of the time, men were over four times as likely to be portrayed as ignorant, foolish or incompetent. Table 3 shows that while both sexes were portrayed as intelligent a near-equal amount of the time, men were over four times as likely to be portrayed as ignorant, foolish or incompetent.

Discussion

Domestic products have generally been used to measure how women are portrayed, and this study found the gap narrowing between men and women, with women only making up 56 percent of actors in commercials for domestic products. Women still were the primary target of body care products, but one could easily argue that it's because they, for a fact, purchase the most body care products; there's nothing sexist about targeting the most likely consumer.

The portrayal of men as more often ignorant or foolish than women was not a surprise, especially if one watches the programs that surround these commercials. The foolish man is fast becoming one of the most prevalent stereotypes used for comedic effect in media today. For example, an ad for a trivia game showed a group of guys versus a group of girls answering trivia questions, with the men betting they would wax their legs if they lost the game. A montage of shots then shows the men floundering on questions while the women answer confidently, ending with a shot of the men making good on the bet. While the humiliation isn't as often at the hands of women as originally thought, if someone needs to act foolish, take a pratfall or be the brunt of any other humiliation, it will be very likely that this person will be male.

On the racial front, only black characters appear as often as they do in the population, although many women that were leaning toward being coded as black were so much lighter skinned than their male counterparts that many were coded as mixed race. Even when women were coded as black, they generally were still of lighter skin than black men.

The U.S. Census Bureau's official race statistics count people of Hispanic/Latino origin as white, which may also explain a lack of representation in this survey - it's simply not always obvious just by looking at someone what ethnicity that person would claim. However, even if all those coded as mixed/indeterminate were of Hispanic/Latino origin, it would still be an under-representation compared to the population. Asian persons were easier to identify and were also underrepresented, more so on the men's side than women's, and even when present were mostly in secondary or background roles.

Conclusions

It would be an impossible goal to expect exact gender and racial ratios in correct proportion to the population of the country - and unless buying practices were constant within those boundaries, such equal representation wouldn't be desired by advertisers anyway. Those who most often buy and use a product will always be the foremost ones targeted by commercials. With women having more financial independence than in times past, women will have more products targeted at them, but not necessarily the same types of products aimed at men, since women can have different buying practices. In that context, this snapshot of commercials shows a near-equality between the sexes as far as representation.

However, non-black minorities are the fastest growing segment of the population (U.S. Census Bureau, 2005) and were far underrepresented by the commercials in this survey. Worse still, what's not reflected in the numbers is that many minority characters were either portrayed as stereotypes or for specific products often associated with that group, like a Hispanic family enjoying tortilla chips or an Asian man using a new electronic device.

It's also of note how often men of any race are portrayed as foolish and incompetent, no doubt in an effort to counteract decades of authoritative men dominating the airwaves, but it's beginning to tip the scales too far the other way.

Future studies should continue to monitor these trends, focusing less on what types of products are being sold, and more on how the characters are portrayed, especially as many commercials are becoming abstract narratives with less connection to the product. Also, this study did not take into account the length of commercials, but a background character in a 60-second commercial may have more screen time than a main character in a 15-second spot. Therefore, a future comprehensive study of media portrayals should take character screen time into account, perhaps more so than counting the number of characters.

There's only so much that can be shown in a commercial, but taken as a whole, they can broadcast a lot of messages and subtle ideas. Whether commercials influence how people think of others or merely reflect those feelings, studying the content of commercials can provide interesting insight into our society, and should be further examined.

References

Allan, Kenneth and Coltrane, Scott. 1996. Gender Displaying Television Commercials: A Comparative Study of Television Commercials in the.... Sex Roles 35 (3-4), 185-203.

Bartsch, R.A., Burnett, T., Diller, T.R. and Rankin-Williams, E. 2000. Gender Representation in Television Commercials: Updating an Update. Sex Roles 43 (9-10), 735-743.

Bresnahan, M.J., Inoue, Y., Liu W.Y. and Nishida, T. 2001. Changing Gender Roles in Prime-Time Commercials in Malaysia, Japan, Taiwan, and the United States. Sex Roles 45 (1-2), 117-131.

Coltrane, Scott and Messineo, Melinda. 2000. The Perpetuation of Subtle Prejudice: Race and Gender Imagery in 1990s Television Advertising. Sex Roles 42 (5-6), 363-389.

Craig, Stephen R. 1992. The Effect of Television Day Part on Gender Portrayals in Television Commercials: A Content Analysis. Sex Roles 26 (5-6), 197-211.

Furnham, Adrian and Mak, Twiggy. 1999. Sex-role Stereotyping in Television Commercials: A Review and Comparison of Fourteen Studies. Sex Roles 41 (5-6), 413-437.

Kim, Kwangok and Lowry, Dennis. 2005. Television Commercials as a Lagging Social Indicator: Gender Role Stereotypes in Korean Television Advertising. Sex Roles 53 (11-12), 901-910.

Schneider, Kenneth C. and Schneider, Sharon B. 1979. Trends in Sex Roles in Television Commercials. Journal of Marketing 43 (3), 79-84.