This study examines differences in gender portrayal in television advertisements between the predominantly Christian, Western, United Kingdom and the overwhelmingly Muslim, Eastern Malaysia through content analysis.
There is an extensive and ever growing literature on sex stereotyping and gender role portrayal in advertising (Kay & Furnham, 2013; Knoll, Eisend, & Steinhagen, 2011) . Indeed there are now well over 100 studies that have investigated gender portrayal in national television advertising (Furnham & Mak, 1999; Furnham & Paltzer, 2010; Furnham & Lay, 2017) . Moreover a recent review considered gender-role advertising in international advertising (Shaw, Eisend, & Tan, 2014) .
However, Eisend (2010) pointed out that prior research predominately examined short-term attitude changes. His meta-analysis of 64 studies regarding gender portrayal in advertising found that advertising primarily depended on the widely accepted gender attitudes and values in a society with no significant lasting effects on viewer’s attitudes. This study has been the first to provide definitive evidence to the debate regarding the influence of advertising on perception of gender roles in society. This corroborates the assumptions of earlier researchers regarding how stereotyped advertisements were simply “lagging social indicators”, i.e. the media perpetuating outdated perceptions which are no longer aligned with the current state of society rather than attempts to impose gender role beliefs.
Although gender stereotyped advertisements appeal to more traditional people, Morrison and Shaffer (2003) found that traditional individuals viewed non-conventional advertisements more favourably when prompted to self-reference. Additionally, researchers such as Caballero, Lumpkin and Madden (1989) also noted that stereotypical images may possibly dissuade consumers; this is potentially due to these perceptions being viewed as outdated and perhaps offensive by more progressive individuals. In combination with the progress of gender equality, it may be in the interests of advertisers to rely on less stereotypical advertising particularly when the advertisements attempt to relate to the consumer.
Women tended to be used to advertise home and body products while services, technological and automotive products were primarily advertised by men (Furnham & Bitar, 1993; Furnham & Mak, 1999; Bresnahan, Inoue, Liu, & Nishida, 2001; Uray & Burna, 2003) . However, there appeared to be mixed results for food advertising, e.g. between
Coinciding with the association between women and home products, female primary characters tended to be portrayed as being at home while men were more likely to be shown outdoors (Bretl & Cantor, 1988; Gilly, 1988; Furnham & Bitar, 1993; Uray & Burnaz, 2003; Furnham & Mak, 1999) . Furnham and Paltzer (2010) noted, in a meta- analysis of 30 studies performed in this area in the last decade, that the level of stereotyping in regard to location was lower in countries such as Kenya and New Zealand though women were still far more likely to be pictured at home in Mauritius, the United States and Malaysia. Bresnahan et al. (2001) observed that Malaysian men were overall more likely to be portrayed in the workplace as expected of a country with a patriarchal society. There were no significant differences found for British, Mexican and French advertisements in a meta-analysis conducted on 14 studies across 11 countries (Furnham & Mak, 1999) . On the other hand, females were portrayed far more frequently at home while men were more likely to be depicted in occupational settings in other studies of British advertisements (Furnham & Bitar, 1993; Furnham & Paltzer, 2010) . Therefore, it is hypothesised that both Malaysian and British female primary characters would be more likely to be depicted at home than males.
The tendency to portray males as an authority with expertise while relegating women to the role of users and models appeared to be sustained across time in Britain (Manstead & McCulloch, 1981; Furnham & Mak, 1999; Uray & Burnaz, 2003; Furnham & Paltzer, 2010) . However, Bretl and Cantor (1988) have noted an increasing number of males in the “user” role thereby suggesting that the media may be slowly acknowledging changes in the status quo and reflecting it in their advertising. This study will explore the credibility associated with men and women in
The research regarding credibility coincides with the way that men tend to consistently be portrayed as employed professionals in autonomous while women tend to be unemployed or familial roles in countries ranging from the US to Japan to Mauritius (Gilly, 1988; Bresnahan et al., 2001; Uray & Burnaz, 2003; Kim & Lowry, 2005; Furnham & Paltzer, 2010) . British advertisements tended to portray men in autonomous roles and women in dependent roles even though an increasing percentage of women have been depicted in the role of interviewers (Manstead & McCulloch, 1981; Furnham & Skae, 2015) . Tan et al. (2002) found that males were still more likely to be portrayed in professional roles than females in
There were generally two types of arguments used in prior studies: factual and opinion based. Furnham and Mak (1999) and Furnham and Paltzer (2010) found that men primarily gave factual arguments for a product in a majority of the samples while women tended to give opinions (e.g.
Furnham & Mak (1999) observed that female characters in British commercials were more likely to make factual arguments in comparison to male characters; this could possibly be in order to appeal to more female consumers since the advent of gender equality movements. Therefore, it is hypothesised that British female primary characters will be depicted using factual arguments more frequently than males. It is also hypothesised that females will tend to be shown using opinion-based arguments while males are shown use factual arguments in Malaysian advertisements due to the traditional nature of the country wherein women are viewed as being less knowledgeable than men.
There was a noticeable difference the type of rewards marketed with the general trend being pleasurable rewards for men who use the product and self-improvement as motivators for women in the UK as well as multiple other countries (Manstead & McCulloch, 1981; Furnham & Mak, 1999; Furnham & Skae, 2015; Furnham & Paltzer, 2010) . However, there appears to be a mixed result regarding the use of practicality as a motivator. There was also within country differences found (Furnham & Paltzer, 2010) in Hong Kong whereby the usage of practicality as a reward increased when the advertisement was in Chinese instead of English. There is a possibility that advertisements in the native tongue of the country are more reflective of their current standards than those in a foreign language as those who are multilingual (and thus more likely to be exposed to foreign cultures) may be more receptive to unconventional gender portrayals. Therefore, this study will focus mainly on advertisements produced in
There has been little research done regarding the temperament of the primary characters in television advertising. Gilly (1988) observed that Mexican women were more likely to be portrayed as frustrated while this was true for men in Turkish advertisements (Uray & Burnaz, 2003) . As frustration is one of the affective by-products of helplessness (Seligman, 1975) , women may be more likely to be depicted as frustrated in more traditional cultures wherein women are viewed as being less autonomous and capable than men. Thus, it is predicted that Malaysian female primary characters will be significantly more frustrated than the males while no gender differences are expected for frustration in British advertisements.
Similarly, few studies have incorporated humour as a factor despite the efficacy of humour in advertising (Furnham, Gunter, & Walsh, 1998) . Furnham and Skae (2015) posited that sex-role reversals may occur mainly in humorous circumstances, e.g. the inept male in a homemaker role rather than a serious depiction of non-traditional roles. This study will explore the general usage of humour in British and Malaysian advertising with the prediction that “humour” will be utilized more frequently in the presence of “role reversals”.
Finally, the bulk of prior research in this area has neglected the degree of sexualisation in televised advertising. Past studies (Gilly, 1988; Furnham & Mak, 1999; Furnham & Paltzer, 2010) have shown that female primary characters have been consistently younger than males even across cultures and the decades since the original study. Although earlier studies had initially found that a greater proportion of married female primary characters (Schneider & Schneider, 1979; Cagli & Durukan, 1989) , a recent study by Kim and Lowry (2005) noticed that women tended to be shown as unmarried in an analysis of Korean adverts. This may indicate an increasing use of sex appeal and a decreasing focus on the importance of social hierarchy whereby women were traditionally perceived as being more “respectable” due to their married status. Thus, this study aims to explore the levels of sexualisation of primary characters in a conservative, religious country (
Furnham and Mak (1999) noted that gender role stereotyping was more prevalent in more traditional countries such as Indonesia and Hong Kong even exceeding the levels found in the United States in the original studies from the 1970’s. They argued that the high levels of stereotyping found in Asian countries such as
This study expects findings from the Malaysian sample to be similar to those of in Turkey (Uray & Burnaz, 2003) and Hong Kong (Furnham & Li; 2008) , i.e. a significant gender gap though with more equal proportions of males and females distributed within the variables; due to a shift from collectivist to individualistic systems of thought as a result of increased Westernization. Malaysia is an Islamic country which adheres highly to traditional Malay culture with Islam being a strong foundation for many of the cultural norms and laws. A census conducted by the CIA in 2000 had found that 60.4% of the population were Muslim, 19.2% were Buddhist, 9.1% were Christian, 10.4% followed other religions while only 0.8% surveyed had no religion. It has also been noted that there has been strong gender stereotyping and linguistic sexism in Malaysian textbooks such as associating fathers with decision-making authorities with financial and occupational responsibilities while mothers were associated with childcare, cooking and cleaning the house (Hamid, Yasin, Bakar, Chee, & Jaludin, 2008) . This is in contrast to the
Women comprised of only around 35% of Malaysia’s labour force from 1975 to 2002, with small variations depending on work sectors. Furthermore, only about 10% of
This template, created in MS Word 2007, provides authors with most of the formatting specifications needed for preparing electronic versions of their papers. All standard paper components have been specified for three reasons: 1) ease of use when formatting individual papers, 2) automatic compliance to electronic requirements that facilitate the concurrent or later production of electronic products, and 3) conformity of style throughout a journal paper. Margins, column widths, line spacing, and type styles are built-in; examples of the type styles are provided throughout this document and are identified in italic type, within parentheses, following the example. Some components, such as multi-leveled equations, graphics, and tables are not prescribed, although the various table text styles are provided. The formatter will need to create these components, incorporating the applicable criteria that follow.
The sample of advertisements analysed in this study were obtained by recording prime- time television programming in three hour slots (19:00-22:00) during Monday to Fridays for a span of two weeks. The Malaysian sample consisted of advertisements aired on TV3 and Astro Ria during the weeks of 7th September and 14th September 2009. The British sample consisted of recordings from ITV and Channel 4 for the weeks of 9th November and 16th November 2009. Similar to Bretl and Cantor (1988) and Craig (1992), all public service announcements, programming related advertisements and duplicates were excluded from the final samples. Purely narrative advertisements, those containing more than one primary character and those solely containing gender neutral animated characters were also removed. 136 Malaysian advertisements and 100 British advertisements were obtained for the final samples.
2.2. Coding Procedure
Following Bretl and Cantor (1988) , an individual coded the entirety of the samples based on the criteria listed below and a second individual coded a random subset of 33 advertisements. Both coders were bilingual and the second coder was an independent party unaware of the experimental hypotheses of the study. Inter-coder reliability was found to be at 81.5% with agreements on 242 out of 297 coded items.
2.3. Coding Criteria
Primary characters were coded based on gender and defined as characters with the most screen-time and/or dialogue during the advertisement (McArthur & Resko, 1975; Schneider & Schneider, 1979; Bretl & Cantor, 1988; Tan et al., 2002) . In advertisements where there were multiple characters, the more prominent was selected as the primary character (Uray & Burnaz, 2003) . The advertisements were further coded according to nine other variables relating to the general attributes of the advertisement and the attributes of the primary character. The variables chosen were based on those frequently used by prior content analyses but some categories have been altered or collapsed to accommodate for the smaller sample sizes in this study. The definitions of these variables and their sources are listed in Table 1 and Table 2.
Chi-square tests were performed comparing the frequency distributions for male and female primary characters for UK and Malaysian advertisements. Cross-cultural comparisons were also performed, i.e. comparing depictions of one gender in the
Hypothesis 1: Product Types. In regard to British advertisements, females were more likely advertise health and beauty products than males (43.1% vs. 17.0%) and were also more likely to advertise household products (12.1% vs. 2.8%). Male characters tended to advertise services more frequently than female characters (23.8% vs. 3.4%). Similarly, women in Malaysia were more likely to be shown promoting health and beauty products (52.6%) in comparison to men (25.6%) while men were more likely to be shown advertising food products (53.8% vs. 18.6%). Significant gender differences were found for both the
Table 1. Variables related to the attributes of the advertisement.
Table 2. Variables related to the primary characters.
Hypothesis 2: Credibility. For both countries, males were more frequently shown as authorities (
Hypothesis 3: Roles. British male primary characters were depicted almost equally in professional (44.7%) and non-professional occupations (55.4%) while female characters were far more often shown as non-professionals (89.3%), Χ2 = 14.18, df = 1, p < 0.001. A comparable trend was also found in Malaysian advertisements whereby men were shown almost equally as professional and non-professional (43.2% vs. 56.8%) while 15% of women were shown in professional roles, Χ2 = 11.85, df = 1, p = 0.001. This supports the hypothesis that females would be more likely to be depicted in non-professional roles in both Malaysian advertisements and British advertisements
Hypothesis 4: Location. There were no statistically significant gender differences found in location of the primary character for both the UK (Χ2 = 5.28, df = 2, p > 0.05) and Malaysia (Χ2 = 5.92, df = 2, p = 0.052); although the Malaysian sample was close to significant with 23.7% more males being portrayed outside.
Hypothesis 5: Arguments. Hypothesis 5a: There was no significant gender difference found in the type of arguments used in Malaysian advertisements, Χ2 = 1.83, df =
Table 3. Comparisons of characteristics related to male and female characters in British and Malaysian commercials.
2, p > 0.05, therefore the hypothesis was not supported. Hypothesis 5b: Males were more likely to use opinion based (64.3%) rather than factual arguments (19.0%) while females were equally likely to use either arguments, Χ2 = 6.24, df = 2, p < 0.05. The results contradict the hypothesis that females would be more likely to use factual arguments in British advertisements.
Hypothesis 6: Rewards. No significant gender differences were found in either country for the type of rewards used by the primary character to promote the product despite a general trend for women to use self-improvement as a reward, (Χ2)UK = 6.44, df = 3, p > 0.05, (Χ2)Malaysia = 6.68, df = 3, p > 0.05.
Hypothesis 7: Frustration. Hypothesis 7a: As predicted, there was no significant difference found between genders for frustration in British commercials, Χ2 = 0.23, df = 1, p > 0.05. Hypothesis 7b: Females (14.4%) were shown to be frustrated more often in Malaysian commercials in comparison to men (2.6%), Χ2 = 3.99, df = 1, p < 0.05.
Hypothesis 8: Humour. There was no gender difference in the presence of humour in the
Hypothesis 9: Sexualisation. Females were found to be more frequently depicted as sexual objects than males in both Malaysian and British advertisements but only statistically significant differences were found in Malaysia ((Χ2)UK = 2.83, df = 1, p > 0.05; (Χ2)Malaysia = 3.88, df = 1, p < 0.05).
Hypothesis 10: Cross-cultural Comparisons of Male and Female Primary Characters. The qualities of British male primary characters were compared with Malaysian male primary characters and British female primary characters with Malaysian female primary characters in order to identify similarities and differences between countries. and British males, Χ2 = 15.96, df = 5, p < 0.05. Malaysian male primary characters were more likely to advertise food, and health and beauty products while British male primary characters were more likely to advertise electronics and services.
British female primary characters were found to rely on significantly different types of arguments than Malaysian female primary characters, Χ2 = 6.65, df = 2, p < 0.05. British females were equally likely to be shown to using either factual and opinion based arguments while Malaysian females were observed to used opinions (61.5%) more frequently than factual arguments (30.2%) in advertisements.
Another cross-cultural difference was that humorous advertisements were twice more common in British advertisements than Malaysian advertisements for both males (Χ2 = 10.86, df = 1, p = 0.001) and females (Χ2 = 8.10, df = 1, p < 0.005). As this was seen for both males and females, this would imply that neither gender is particularly associated as being a source of humour and is simply a cultural difference in reliance on humour in advertising.
There were no significant differences found between the portrayal of British men with Malaysian men and British women with Malaysian women in the remaining categories.
Similar to prior studies (Furnham & Bitar, 1993; Furnham & Mak, 1999; Bresnahan et al., 2001; Uray & Burnaz, 2003) , female primary characters tended to be used for health and beauty products more than males in both countries; they also showed similar distributions for types of products advertised. On the other hand, significant differences were found for male primary characters across the two countries. Malaysian male primary characters were more likely to promote food, household and body products than their British counterparts who focused on services and electronics. This may imply cultural differences regarding the importance of various goods and services. For example, Malaysians may be slightly more family oriented and invested in self-enhancing goods than the British who may be more interested in convenience goods such as technological devices and financial services. Furthermore, it may also be a reflection of audience demographics. For instance, the collectivist nature of Asian cultures would suggest that television watching may be a more family oriented activity than in individualistic cultures such as the UK and therefore, Malaysian advertisers would seek to market more family oriented goods during prime-time television slots.
Surprisingly, there were no significant gender differences in location of primary characters in both countries with women being equally likely to be shown at home or outside. Although an earlier study had found a lack of gender differences in
However, it should be noted that women were still far more likely to be portrayed as non-professional individuals, i.e. without occupations, in both the
In addition to this, both UK and Malaysian female primary characters were more likely to be depicted as users (90.4% and 83.5%) of a product rather than authorities. While this follows the general trend of previous studies (Furnham & Mak, 1999; Uray & Burnaz, 2003; Furnham & Paltzer, 2010) , this study had found to the contrary that male primary characters were also more likely to be shown as users instead of authorities in both countries. It had been initially suggested by prior researchers that being an authority placed the person at a high-credibility or expert status; with the stereotypical associations being that women were of low-credibility while men had high-credibility. However, Bretl and Cantor (1988) had noted early on that males were becoming more commonly depicted as users since the original studies. Evidence for the male shift from authority figures to user figures has been found mainly in studies performed in the last decade (Furnham & Farragher, 2000) . Therefore, this could be a sign that the perceptions of users and authorities have changed over time.
Significant gender differences for types of arguments used were only found in UK advertisements while both male and female Malaysian primary characters relied heavily on opinion based arguments. British male primary characters were more likely to present an opinion-based rather than a factual argument; female primary characters were equally likely to present either argument. Comparisons with prior studies performed in the US (Bretl & Cantor, 1988; Manstead & Mcculloch, 1981; Furnham & Skae, 2015) would indicate that there has possibly been a gradual increase in the use of opinion based arguments by men since McArthur and Resko (1975) . Gilly (1988) noted that Australian advertisements, which were considered far less stereotypical, tended towards opinion-based arguments as well. Therefore, there might have been a progression from males being associated with factual arguments to gender neutrality in Westernised countries for both these sub-categories. On the other hand, Malaysian female primary characters were significantly more likely to give opinions in comparison to
The lack of significant differences within and between cultures for type of rewards utilized by the primary character would suggest that expectations of men and women have been levelling out over time. For example, there appears to be an increase in the use of self-enhancement rewards by male primary characters in comparison to previous studies. However, there has been conflicting evidence from a number of studies regarding the types of reward thereby complicating the predictive quality of this category (Furnham & Mak, 1999; Furnham & Paltzer, 2010) .
Sexualisation was not particularly prominent in both countries although females were portrayed more frequently as objects of sexual attraction than men in Malaysian commercials. Although females were also more likely to be sexualised in British commercials, men were occasionally used as models as well. There is a stark difference with
In regard to temperament, the only significant difference found was between the presence of frustration in Malaysian male and female primary characters; with women being depicted as frustrated more frequently as predicted by the hypothesis. Furthermore, the findings support the hypothesis that there would be no gender differences found for frustration in British characters. The results would confirm the notion that women in countries typically more prone to gender stereotyping are perceived as being more helpless and emotionally volatile when faced with challenging situations.
Humour was observed to be used more frequently in British advertisements than Malaysian advertisements although there was no significant gender portrayal differences found within each country. The lack of significant differences in distribution was sustained even when the humorous advertisements were further sorted in relation to professional and non-professional roles. This would contradict the suggestion that role reversals (non-professional male and professional female primary characters) mainly occur in humorous circumstances (Furnham & Skae, 2015) . It is more likely that advertisers prefer to use humour in more casual situations regardless of gender.
The lack of significant gender differences in about half of the categories investigated would lead one to conclude that the gender gap in portrayals has been decreasing over time. As the findings show a more equal distribution for women than men within a category, e.g. location and arguments, there is a possibility that it is more acceptable for women to possess traditionally masculine traits than for men to possess effeminate traits. Aside from that, the lack of significant differences for a majority of the cross- country comparisons would suggest that the perceptions of gender roles in the
Like all others this study had limitations, the most important of which was the sample. Although we had a relatively large sample of television advertisements compared to other studies in this area, it was inevitable limited both in number and the channels sampled. Inevitably the larger the sample the more representative the results. Further this data was collected a few years ago and trends may have changed.
Future research could concentrate on additional content categories as well as advertisements played at different times of day. More importantly as young people watch television less and less, it would be interesting to content analysing advertisements in the new social media to see if the trends continue. Further, given the growth of many Asian economies, it is important to see how economic growth and development have an impact on advertisements over time.