1
29.06.2017
Master’s Thesis
Reelistic vs. Realistic:
An exploration of Racial and Religious Minority
Representation in Hollywood films
Graduate School of Communication
Master’s Programme Communication Science Supervisor: Dr. Rachid Azrout
2 Author: Karishma Sholingur Kothandaraman Mohan Student ID: 11181486
Abstract
Recently, social media movements such as #oscarssowhite have defamed Hollywood for their underrepresentation and misrepresentation of minorities. Hollywood films epitomize the broader undercurrent in mass media that minorities lack accurate representation. However, due to current issues with race relations in the US becoming tenser, it is now more important than ever that minority issues and needs are addressed in the media. To analyse, to what extent Hollywood has an issue with racial and religious minority visibility and stereotyping over time, this study content analyses 1109 characters from 111 blockbuster films produced between 1980 and 2016. It is scientifically relevant as it seeks to analyse trends in Hollywood films over an extended period of time which is lacking in earlier literature. The findings of this study are that most minorities are proportionately underrepresented in Hollywood blockbusters compared to their demographic. The study also finds that stereotypical depictions of minorities is in decline and that the stereotypes used are prone to change.
3 Introduction
Recently in the US, there has been a lot of tension between society’s majority and minorities which stems from fear and a lack of understanding of the ‘other’. Fear mongering tactics are constantly being used by highly visible populist and conservative politicians. These tactics include suggesting a border wall between Mexico and the US, hatred towards Muslims arising over terrorism and anti-immigrant sentiments fuelled by the refugee crisis which now plague the US. Alongside internal grassroots movements such as Black Lives Matter, these sentiments elucidate a severe lack of understanding of minority needs and an inability to sympathise with minorities on an individual basis.
Part of the explanation for these tensions is that people are exposed to society’s outgroups through the media rather than through face-to-face contact. Several studies have already shown that in traditional mediums such as television news and newspapers, minorities are often misrepresented and lack visibility (Trebbe & Schoenhagen, 2011; Phillips, 2009; Dixon & Williams, 2014; Jacobs, 2016). With declining news consumption, especially amongst young people, there is a necessity to study alternative entertainment mediums. The effects of entertainment mediums such as soft news and infotainment have been shown to have an equal effect on viewers as traditional fact based mediums (Baum, 2003; Boukes et al, 2015). To expand the literature available on minority representation in entertainment media, in this study, minority representation in Hollywood movies is studied over time.
As contemporary society’s storytellers, Hollywood have the same agenda-setting capabilities and priming effects as more traditional media such as the news (Mulligan & Habel, 2011). As a consequence, their depictions of minorities can have a strong effect on the attitudes of film consumers. Social media movements such as #oscarssowhite suggest that there is a definite
4
issue with minority representation in Hollywood films and the public are increasingly aware of this issue. The inescapability of biased casting choices is best exemplified by the casting of Cuban descent actress, Eva Mendes as Will Smith’s romantic interest in the film, Hitch. When discussing the casting choice, Smith stated the following:
“There’s sort of an accepted myth that if you have two black actors, a male and a female, in the lead of a romantic comedy, the people around the world don’t want to see it… We spend $50 something million making this movie and the studio would think that was tough on their investment. So the idea of a black actor and a white actress comes up- that’ll work around the world, but it’s a problem in the US.”
In spite of Smith’s fame and likeability, he was incapable of escaping the constraints placed on minorities in Hollywood. This indicates a necessity to fix the issue of minority
representation in Hollywood films. Considering Hollywood films are a capitalist commodity, one way to increase minority visibility and encourage accurate representation is for viewers to transition from ‘viewer consumers’ to ‘viewer citizens’ (Kim, 2004). However,
Hollywood’s issue is difficult to rectify without information on the current state of affairs on minority representation on screen. As a result, there is a need to assess the extent of minority visibility in Hollywood and the stereotypes utilised to portray minorities over time. Trend analysis can offer an indication as to whether minority affairs are improving or deteriorating in Hollywood over time.
With all this considered, this study aims to explore racial and religious minority visibility in Hollywood over time to offer a trend analysis of Hollywood blockbuster films. Firstly, a theoretical overview of the current trends in minority visibility and stereotyping in media is offered. Methodologically, this study content analyses 111 blockbuster films over 37 years (1980-2016) to explore the state of minority representation and stereotyping in media and
5
how it changes over time. In turn, a quantitative analysis of the representation of minority characters and a qualitative analysis of the stereotypes they portray are presented in the following study.
Fictional framing and minority representation
Trebbe and Schoenhagen (2011) identify that in the vast amount of research conducted into minority representation in mass media, there are three broad patterns. While these trends are evident across a vast array of mediums, they are particularly evident in news media. The first of the three noticeable patterns is that minorities are generally underrepresented in news media. For example, a study conducted on television news in Australia reveals that minorities are generally underrepresented in news media (Phillips, 2009). Secondly, when minorities are represented in the media, they are often stereotyped. Dixon and Williams’ (2015) study perfectly exemplifies this trend in their finding that American cable news presents minorities in a stereotype laden manner such as the ‘Muslim Terrorist’ or the ‘Latino/a undocumented immigrant’. Lastly, when minorities are represented in news media, they are frequently framed in a negative manner. A longitudinal content analysis conducted by Jacobs (2016) on the crime section in newspapers in Flanders highlights that minorities in news media are frequently portrayed as more threatening, violent and anonymised. These studies reveal that there is a consistent issue with minority misrepresentation in multicultural Western societies globally. In this study, I will focus on two of the patterns identified by Trebbe and
Schoenhagen (2011): minority underrepresentation and stereotyped images of minorities presented in the media.
Indeed, general analysis of minority groups in the media reveal that minorities are largely underrepresented in media (Trebbe & Schoenhagen, 2011; Bleich, Bloemraad & Graauw, 2015). For example, a study by Santa Ana (2013) on Latino/a visibility in news media finds
6
that Latino/as representation in television news in the USA is considerably disproportionate to the percentage of the population they comprise. Similarly, Beick, Bloemraad and Grauw (2015) assess the frequency with which certain Indian origin communities and Vietnamese communities are mentioned in four regional North American newspapers. Once again, they find that these groups are underrepresented in the media when compared to their
demographic in population. Another study by Tukachinsky, Mastro & Yarchi (2015) content analyses 345 television show over the span of 20 years. They find that Latino/as, Asian Americans and Native Americans are underrepresented on television. Thus, in general, minorities lack visibility.
Issues of visibility are further complicated as certain minorities are represented more often than others in the media. Dixon and Linz (2000) find that Blacks are more likely to be overrepresented in television crime news than Latinos and Whites in relation to their
proportion in the population. Likewise, a content analysis on minority presence in magazine ads concludes that black models were the most visible minority group (Bowen & Schmid, 1997). In contrast, Asian and Hispanic models were barely used. This repeated trend in underrepresentation and discrepancies in visibility amongst minorities shows that certain minorities are prioritised over others.
Pertaining issues of minority visibility, several studies have been conducted on the representation of minorities in news media (Trebbe & Schoenhagen, 2011; Phillips, 2009; Dixon & Williams, 2015; Jacobs, 2016). However, as traditional news consumption is in decline, particularly amongst young people, it is important to understand how other media sources can impact attitudes (Casero-Ripolles, 2012). Certainly, several studies regarding minority representation in other mediums of entertainment, such as television programmes (Tukachinsky, Mastro & Yarchi, 2015), advertising and magazines (Bowen & Schmid, 1997) have already been discussed in this literature review and indicate the presence of a large body
7
of literature regarding minority presence in mass media. This study further seeks to expand the array of media analysed in the field of political communication by using Hollywood films.
Hollywood narrative films can have an equally persuasive effect as fact oriented media such as the news (Green et al, 2016). Mulligan and Habel (2009) argue that fictional media
partakes in agenda-setting and priming on an equal level to news media. In turn, this makes fictional media equally, if not more persuasive than news media. This is primarily attributed to individuals being more resistant to the persuasive impact of fact based media such as the news due to distrust in news media (Jones, 2004; Tsafti & Capella, 2005). The effects of fictional media are said to be more difficult for individuals to resist in spite of analysing films for bias due to narrative transportation (Mulligan & Habel, 2009; Green & Clark, 2013). Therefore, the content presented in fictional media can have an equally persuasive effect on forming people’s attitudes and world view as other media sources. Hollywood narratives are consequently insightful source of content to analyse when determining factors in the
formation of people’s attitudes towards minorities.
Several pre-existing content analysis studies of Hollywood films exist but are however, limited in scope and do not offer a comprehensive understanding of minority representation in Hollywood( Escholz et al., 2002; Dale, 1935; Shaheen, 2003). Escholz et al.’s study finds that minority representation is improving but still restricted to traditional stereotypical roles. While comparative, the study is narrow in scope as it is not longitudinal and instead, only films from a singular year were analysed. Another study by Dale (1935) finds that minority representation is frequently stereotypical and only existent within the confines of the genre, comedy. While Dale’s study (1935) is valuable and representative of the time, it is now anachronistic and no longer explains contemporary trends in films. There is also a tendency to focus on a singular minority group which does not examine the differences in
8
representation of various minorities. Shaheen (2003) for instance, provides a content analysis on the representation of Arabs in over 900 films. His findings are extensive and suggests that Arabs are often underrepresented and their stereotypical representation has remained static over time. However, few conclusions can be drawn on the nature of other minority groups from his literature. This study hopes to circumvent these pitfalls to provide a more
comprehensive understanding of minority representation in Hollywood films.
Representation within Hollywood films and the media generally are problematized as they are influenced by external factors such as history and changes over time. Kim (2004) highlights how following the Civil Rights movement in the USA, African Americans were seldom caricatured as maids and handymen. Instead, this role was increasingly fulfilled by Asian Americans and consequently, the visibility of Asian Americans in media increased. More recently, a study conducted by Tseligka (2016) following negative press about Greeks after the bankruptcy showed that there was a significant shift in both the frequency and nature of representation in the German press. These changes show that historical context is essential in understanding minority representation in films and media. As a result, it is important to study minority representation in media over time because the media plays an integral role in the production, reproduction and transformation of attitudes (Hall, 1995). Changes in representation over time can therefore change people’s attitudes towards minorities as well.
Whitewashing is also a serious issue regarding minority visibility in Hollywood films. Historically, there have been popular and overtly racist depictions of minorities on screen in genres such as Black minstrelsy (Perez, 2014). In these types of movies, white characters often adopt makeup and stereotypical mannerisms which make the character’s intended race to be easily identifiable. In spite of the belief held by many that we live in a post-racist society (Lee, 2013; Lum, 2009), in contemporary times, media still use white actors in order
9
to portray minority characters on screen. These representations are problematic because they reduce the visibility of minorities and exclude them from public discourse on their own race.
Another factor that can influence minority representation in Hollywood films is that certain genres provide a more conducive avenue for positive depictions of minorities and are consequently more likely to cast minority actors. Pertaining the film Rush Hour 2, Park, Gabbodan and Chernin (2006) contend that comedy as a genre helps alleviate differences between minorities as race-related jokes are frequently perceived as inoffensive. As a result, some genres are more likely to portray minorities with greater frequency and accuracy than others. Vice versa, consumer demand for certain genres that are more conducive to minority representation, including specific minority groups such as LGBT; albeit not a minority considered in this study, dramatically increases the production of media involving the certain genre (Roos & Schachar, 2014; Hart, 2005). Hence, this study also considers the impact of genres on minority representation in Hollywood.
Underrepresentation in the media may translate to the social and systematic oppression of minorities in several cases (Bleich, Bloemraad & Graauw, 2015). In many multicultural societies, the white majority and several other minority groups seldom come in direct contact with every minority group living in their country (Kidd, 2011). Therefore, mediated
representation may be the only exposure society’s majority may have to certain minority groups. Underrepresentation can rid Hollywood films of the normalising message that minorities belong to the society they currently live in. This effects both minorities and the majority (Kidd, 2011). Media representation may also control the relative importance
attached to the interests and needs of different minorities based on their appearance in media (Ungerleider, 1991). On a superficial basis, this leads to the majority being generally
uninformed about minorities and their lives. However, as the majority are often also those who serve as government representatives, biases created through the perpetuation of a media
10
stereotype can impact policy making. As a result, majority perception of minority stereotypes can prevent minorities from being accepted in society and deprive them of their needs on both a social and a systematic basis.
Stereotypes
Tropes are common symbols of understanding established and conventionally structurally embedded in most Hollywood films over time (Weaver, 2012). They act as visual and verbal cues that influence how the audience react to certain situations, objects and people presented in films subconsciously (Astrom, 2015). Hollywood has a tradition of using identity limiting and stereotypical tropes when addressing minorities. One of the earliest studies regarding minorities in Hollywood films is Edgar Dale’s content analysis of 115 motion pictures in 1935. Dale (1935) finds that minorities are either used for comedic purposes or portrayed in an unattractive manner in these films. Moreover the repetition of tropes of certain minorities has been worryingly recycled in spite of changing times (Sheehan, 2003). The recycling of stereotypical tropes can have several negative effects on minorities in multicultural societies.
Due to their control over tropes, storytellers usually possess a great deal of power in society as one of the key purposes of storytelling is to offer constructive insights into social life (Bennet & Edelman, 2003). As Hollywood play the role of storytellers in our current society, the representation of minorities in their stories can have a significant impact on how
minorities perceive themselves and are perceived. According to identity forming theory, accurate representation is necessary for minorities to perceive themselves as part of society (Kidd, 2011). Therefore, when minority groups are exposed to negative or inaccurate portrayals of themselves, their construction of their own identity is influenced by such portrayals. For example, a study conducted by Schmader, Block and Lickel (2015) finds that Mexican Americans who are exposed to even 5 minutes of footage with negative stereotypes
11
felt negative emotions such as shame, guilt and anger in response. Hence, minorities may take these negative and inaccurate portrayals into consideration as a subconscious measure of how valued they are in society as individuals and as a member of a minority group.
Ultimately, exposure to stereotyping effects the identity forming process of minorities.
Moreover, even positive stereotypes such as the intelligent Asian American can have an impact on minorities by limiting their identity to specific social niches. Park and Lee (2010), through interviews and observation, have found that Korean immigrant high school age students often reconstructed their identity to imitate the “model minority” stereotype of Asian Americans upon moving to the USA. As a consequence, when compared to their white peers, Asian Americans have a less malleable identity if they construct their identity based on media representations. It reiterates the notion that minorities are not allowed to embody their space as an individual but as a group of the ‘other’. Thus, both positive and negative stereotyping of minorities in Hollywood films can impact minority individuals’ construction of their own identity.
As a result, stereotypes invoked using tropes in Hollywood offer a great deal of insight into society’s perception of minorities and how the relationship between the privileged and minorities has transformed over time. In particular, American society has an endless history of establishing the other in society from the Bloodthirsty Native American warriors to the more recent Muslim terrorist (Ugor, 2006). Hollywood movies can serve as a tool for the West and the privileged to perpetuate this lopsided power relationship and maintain the image of minorities as the other (Said, 1978; Shaheen, 2003). Certain forward thinking attempts have been made to combat this constant need to establish the ‘Other’ such as the Star Trek franchise including in-depth and positive minority characters such as Uhura. Nevertheless, a few attempts are not an absolute solution.
12
Furthermore, much like the representation of minorities may have changed historically, the stereotype of a certain minority in Hollywood films can either remain static or evolve over time. Shimizu (2016) highlights that Hollywood films are a product of the time and society they were produced. Consequently, as different historical occurrences shape society,
stereotypes in films can also change. The way different minority stereotypes change or do not is exemplified by how Arabs are portrayed as violent hook-nosed men from the initial
development of Hollywood movies (Sheehan, 2003). In contrast, initially, Asians were portrayed as martial arts experts and they are now portrayed as the smart kids in class
(Holtzman & Sharpe, 2014). Therefore, different minorities may change overtime in different ways.
Research Questions
A number of research questions have arisen from the literature review of minority
representation and stereotyping in Hollywood films. Primarily, this study aims to analyse the extent of minority representation in Hollywood films along with representation issues such as whitewashing, change in representation and stereotypes overtime and the effect of genres on representation. However, generally speaking, there is no coherent way to make predictions about the representation of minorities and different minority groups and the stereotyping they experience. Therefore, a series of research questions are posed.
In order to investigate the nature of visibility of minority groups and problems encountered by minority groups regarding representation such as whitewashing and genres, the following research questions are considered:
RQ1: How visible are different minority groups and minorities in general in Hollywood films?
13
To investigate the stereotyping and changes in stereotyping in Hollywood over time, the next two research questions are posed for qualitative analysis based on the data:
RQ3: Which stereotypes are dominantly present in minority character representation in films?
RQ4: Do the stereotypes used to portray minorities in films change over time?
Method
This study uses content analysis to explore the representation and stereotyping of minorities in Hollywood blockbusters. In a similar fashion to other studies that content analyse
Hollywood films, Blockbuster films from several years were chosen for analysis. This is because Blockbuster films have mass appeal and are likely to be more culturally influential than less widely viewed films (Eschholz, Bufkin & Long, 2002).The unit of analysis used in this study are characters in blockbuster Hollywood films. The coding was carried out by a single coder. An additional coder also coded a subsample of 12 films to test inter-coder reliability. In total, the sample includes 1107 characters from 111 Hollywood blockbuster films.
Sample Due to the longitudinal nature of this study, the sample of movies chosen span from 1980 to 2016. The sample is comprised of three films from the top 10 blockbuster films during the time frame. The time frame was divided into three year cycles to reduce the number of films coded. In every three year cycle, for the first year, the 1st, 4th and 7th films were coded. For every second year, the 2nd, 5th and 8th films were coded. For every third year, the 3rd, 6th and 9th films were coded. Therefore, 111 films were coded over 37 years. These films were selected using the online database, ‘Box Office Mojo’.
From each blockbuster film, the top ten listed characters are selected for coding based on the film’s IMDb page. According to the IMDb contributor guidelines, the cast list is generated based on the actors who first appear in a film’s credit sequence. Therefore,
14
including the top ten characters in every blockbuster film ensures the inclusion of at least a substantial proportion of every film’s main cast. The coders then use all information available on the IMDb page to code information about the film itself. Character information is coded using IMDb actor biographies, character descriptions and plot synopsis. When insufficient information is available on the IMDb page for coding characters, coders were encouraged to substitute using other online movie databases such as Rotten Tomatoes and external sources such as Wikipedia. For a more detailed description of the IMDb subsections used to code each variable, check the codebook included in the appendix.
Key Variables The 111 films included in this study were coded for ‘Title’, ‘Year’,
‘Director’, ‘Main Production Company’ and ‘Genre’. All of the variables with the exception of ‘Genre’ were open variables. ‘Genre’ was measured using a nominal variable where multiple options could be selected.
For each of the first 10 characters in the IMDb cast list, basic information was coded such as ‘Gender’, whether the character was ‘Human or not’, whether the character was a ‘Minority or not’ and finally, whether the character was a ‘Main or Supporting’ character.
A shallow coding of actors and characters who were not minorities was conducted. These characters were coded for whether they were ‘Good guys or Bad guys’ and ‘Character Depth’ using ordinal variables. Good guy/ Bad guy was measured using a three point scale (Good, Neutral or Bad). ‘Character depth’ was measured using a five point scale (Yes, Somewhat, Neutral, Not Really and No).
Minority characters were coded in more detail. Along with the variables used in coding non-minority characters and actors, minorities were coded for ‘Race of actor’ and ‘Race of character’. The coding options for race were African, African American, Asian, Asian American, Indigenous, Indian, Hispanic, Arab, White, Mixed and Other. Similarly the
15
‘Religion of actor’ and ‘Religion of character’ were also coded. The coding options for religion were Christian, Muslim, Hindu, Buddhist, Jewish, Other and Uncodeable.
In terms of measuring the nature of minority representation the following variables were coded. The ‘Classification of Minority’, in other words, whether or not the minority character was played by a minority actor or not was coded for. In addition, whether or not the
character’s role was race specific or not was coded using the variable ‘Importance of Race’. Lastly, whether the character was stereotyped or not was measured using a three point scale (Yes, Somewhat, No) and which stereotype a minority character belonged to was coded using an open question.
The specificities for each variable and the exact criteria for coding them can be found in the codebook in the appendix.
Intercoder Reliability A subsample of 12 films was randomly selected for coding by a second coder to measure intercoder reliability. The results of the intercoder reliability using Kappa, Percentage of Agreement and Krippendroff’s Alpha are presented in the table below (Table 1).
The variables ‘Gender’, ‘Main or Supporting actor’, ‘Religion of actor’ and ‘Religion of character’ all have a KALPHA value of 1. These variables are therefore considered to have a high level of intercoder reliability.
The Kappa and KALPHA values for the variables ‘Inportance of race’, ‘Movie Genre’. ‘Human or not’ and ‘Classification of minority’ have been omitted from the table presented below as there was no variation between the responses of ‘Coder 1’ and ‘Coder 2’.
Nevertheless, as there is a 100% agreement between ‘Coder 1’ and ‘Coder 2’ on these variables, they are considered reliable.
16
‘Race of actor’ (KALPHA= .73) has a moderate level of reliability. However, when the options ‘African’ and ‘African American’ are combined into one option, ‘African_1’ and ‘Asian’ and ‘Asian American’ are combined into one option, ‘Asian_ 1’, the new recoded variable ‘Race of actor 2’ (KALPHA=1) has a high level of intercoder reliability. When the same recoding process is applied to ‘Race of character’ (KALPHA= .63), the new recoded variable, ‘Race of character 2’ (KALPHA=1) also has a high level of intercoder reliability. Therefore, the two recoded variables were used for analysis. For the variables ‘Good Guy/ Bad Guy’ (KALPHA= .37) and ‘Character Depth’ (KALPHA= .41), there is a low level of agreement between ‘Coder 1’ and ‘Coder 2’. This is probably because these variables are highly subjective. However, for the purpose of this study, they are still included in the analysis. The variable ‘Is the character stereotyped?’ (KALPHA= .26) has a low level of intercoder reliability. However, ‘Coder 1’ and ‘Coder 2’ agree on this variable 88.33% of the time. This is probably explained due to the option ‘No’ (N= 1048) having a substantially higher frequency than ‘Yes’ (N= 53) and ‘Somewhat’ (N=8). The extreme variation in
distribution is in general, incompatible with Krippendroff’s Alpha. Therefore, considering the percentage of agreement, the variable is retained for analysis.
The intercoder reliability of the open ended variables included in this study were not calculated as they were coded qualitatively.
Table 1: Intercoder Reliability Results
Variable Kappa Percentage of
agreement KALPHA Movie Genre - 100% - Gender 1 100% 1 Human or not - 100% - Main or Supporting character 1 100% 1 Minority or not .6 88% .6 Classification of - 100% -
17 minority Importance of race - 100% - Race of actor .71 83.33% .73 Religion of actor 1 100% 1 Race of character .6 66.67% .63 Religion of character 1 100% 1
Good guy/ Bad guy .25 54% .37
Character Depth .09 23.33% .41
Is the character stereotyped?
.21 88.33% .26
Results
The results section is divided into three parts corresponding to the research questions. The first section gives an overview of ‘Minority representation in Hollywood films’ and
corresponds with RQ1. The second section discusses the impact of genres on the visibility of minorities and corresponds with RQ2. Finally, the third section, ‘stereotypes dominantly present in minority representation and changes over time’ corresponds with RQ3 and RQ4.
Minority visibility in Hollywood films (RQ1)
To answer RQ1 out of the 11091 characters coded, 179 were minorities (N=179, 16.4%). The most visible minority actors were Africans (N= 76, 6.85%), white actors playing minority characters (N= 46, 4.15%) and Asians, including Indians (N=29, 2.61%). The race of the most visible minority characters were Africans (N=73, 6.58%), minority actors playing white characters (N= 27, 2.43%) and Asians including Indians (N=37, 3.37%). A detailed
description of all the racial minorities included in this study is offered in the table below (Table 2).
In order to compare the distribution of races in Hollywood with the distribution of race in the US population, a chi square goodness-to-fit test was conducted for the variables ‘Minority or
1 Of the intended 1110 characters, only 1109 were coded as one movie in the sample contained fewer than 10 characters.
18
not’. If the sample was distributed in a similar manner to the US population, we can expect 61.6% of the sample to be white and 38.4% of the sample to be minorities. Given this, the minimum expected frequency was 430. The test indicated that the minorities in Hollywood are not similarly distributed in the sample as they are in the general population (X2(3) = 238.20, p<.001). Overall, only 15.96% of the sample are minorities when compared to the demographic where 38.4% of the characters are minorities. Thus, minorities are represented 22.44% less often in Hollywood than they are in the demographic.
A chi-square goodness-to-fit test was also conducted for the variable ‘race of character’ to see if there was a difference in representation of individual races. Only Africans, Hispanics, Arabs and Asians, including Indians were considered individually in analysis as the other races coded had low frequencies. The test was statistically significant (X2(4) = 333.01, p<.001). Africans were present 6.58% less frequently in Hollywood films than they are in reality. Asians, including Indians were underrepresented by 2.26%. Hispanics were the most proportionately underrepresented group in the sample and were underrepresented by 16.52% when compared to the demographic. Meanwhile, Arabs are more likely to be represented than their demographic by .49% in the sample.
Table 2: Race of minority characters and actors coded.
Racial Groups Race of Character and % out of sample
Race of Actor and % out of sample Racial minority percentage in the population (Census data, 2015) African 73 (6.58%) 76 (6.85%) 13.3%
Asian & Indian 37 (3.37%) 29 (2.61%) 5.6%
Indigenous 7 (.63%) 4 (.36%) 1.2% Arab 11 (.99%) 6 (.54%) .5% Hispanic 12 (1.08%) 10 (.90%) 17.6% White* 27 (2.43%) 46 (4.15%) 61.6% Mixed Race 1 (.09%) 4 (.36%) 2.6% Other 9 (.81%) 4 (.36%) .2% Total Minorities 177 (15.96%) 179 (16.4%) -
19
Missing 932 (84.04%) 930 (83.86%) -
Total 1109 1109 100%
*Note 1: White characters and actors involved in this table are considered minorities because they are religious minorities
*Note 2: The inconsistency in total characters and actors analysed is due to minority voice actors playing animals in animated films in two instances.
In terms of religion, most characters were areligious (N= 154, 13.89%).The most prominent religious characters were Muslims (N=9, .81%), followed by Jews (N=7, .63%) and
Christians (N=5, .45%). The religion of most actors was uncodeable (N= 139, 12.53%). The most visibly religious actors were Jews (N=26, 2.34%), Muslims (N=6, .54%) and Christians (N=4, .36%). A detailed description of all the religious minorities included in this study is offered in the table below (Table 3).
Comparisons similar to that of race in the sample were not made to the general population as the frequency of religious minorities is low in the sample. Nevertheless, the proportion of religious minorities in the population are displayed in the Table 3.
Table 3: Religion of minority characters and actors coded.
Religious groups Religion of character
Religion of actors Religious minority percentage in the population (Census data, 2015) Christian 5 (.45%) 4 (.36%) 70% Muslim 9 (.81%) 6 (.54%) 0.9% Hindu 1 (.09%) 1 (.09%) 0.7% Jewish 7 (.63%) 26 (2.34%) 1.9% Other 1 (.09%) 1 (.09%) 26.5% Uncodeable 154 (13.89%) 139 (12.53%) - Total number of minorities 177 (15.96%) 177 (15.96%) 100% Missing 932 (84.04%) 932 (84.04%) - Total 1109 1109 100%
20
A logistic regression was conducted to find out the effects of ‘Year of Release’ on the likelihood that a character is a minority or not. Linearity of the variable ‘Year of Release’ in relation to the logit of the dependent variable ‘Minority or not’ was tested using the Box-Tidwell (1962) procedure. Based on this assessment, ‘Year of Release’ was not found to be linearly related to the logit of the dependent variable (p= .483). No outliers were found in the data. The logistic regression model was not statistically significant X2(1)= .493, p= .483. The model explained .1% (Nagelkerke R2) variance in minority status and correctly classified 85.0% of cases. Due to the insignificance of the regression model, it is determined that ‘Year of Release’ does not have an effect on the minority status of a character.
Therefore, there is no pattern pertaining to the visibility of minorities in Hollywood films over time. Although in 2016 (N=16, 53.33%), more minority characters than non-minority characters were cast in the coded films, both the years 1988 (N= 16, 53.33%) and 1992 (N= 16, 53.33%) also cast the same number of minorities. On average, every year approximately 5
out of 30 Hollywood characters were minorities (M= 4.97).
Figure 1: Graph of minority distribution in Hollywood blockbusters over time (1980-2016)
0 5 10 15 20 25 30 35 40 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Year of release
Minority count in Hollywood blockbusters
(1980-2016)
21
A crosstab was conducted to analyse whether there is a statistically significant difference in minority and non-minority characters and how often they play the main character in films. Minorities played the main character in films 20.7% (N= 37) whereas, non-minorities play the main character 29.4% of the time (N= 273). A Chi-square test was conducted and the difference between a minority and non-minority character playing the main character was found to be statistically significant (X2 (1, N = 1109) = 5.66, p =.017).
Table 4: Crosstab of frequency of minorities and non-minorities playing the main and supporting characters with column percentages
Minority Not Minority Total
Main character 37 (20.67%) 273 (29.35%) 310 (27.95%) Supporting
character
142 (79.33%) 657 (70.65%) 799 (72.05%)
Total 179 (100%) 930 (100%) 1109 (100%)
A crosstab was also conducted to analyse the prevalence of whitewashing in Hollywood films over time. ‘Year of release’ was treated as the continuous independent variable and
‘Classification of minority’ was treated as the dependent variable. When an Eta coefficient test was conducted, it revealed that 22.66% of variation explained in the ‘classification of minority’ variable was explained by variance in time (η2
=.23). Therefore, there is a medium level of association between time and the prevalence of whitewashing in Hollywood films. According to the crosstab, whitewashing has decreased over time.
Impact of genres on the visibility of minorities (RQ2)
Since every film coded was coded under several genres, the counts exceed the original total number of characters. Therefore, the counts used to answer RQ2 do not add up to 100% according to the sample and N= 2940 when answering this question. The genres Historical
22
and War were excluded from analysis because they occurred too infrequently for meaningful analysis.
A crosstab was conducted to analyse whether there was a higher difference in frequency of minorities represented in each genre. A Chi-square test revealed that minorities are more likely to be featured in Crime, Sci-fi and Thriller than they are in all of the other genres of film (X2(9) = 35.47, p<.001). Minorities constitute 34% of the Crime genre, 22.35% of the Sci-fi genre and 20% of the Thriller genre.
For a more nuanced analysis, on whether there was a statistically significant difference between minority visibility in the genres where minorities are most visible compared to others, independent sample t-tests were conducted. Proportionately, minorities are most represented in Crime films (N= 17, 34%). However, when minorities (M= 236.5, SD= 310.42) and non-minorities (M= 1218.5, SD= 1676.55) were compared in Crime films and non-Crime films, the difference in minority representation in Crime films was not statistically significant to minority representation in non-crime films (t (2)= .81, p = 0.501). Sci-fi (N= 38, 28.79%) was the second most likely to feature minorities. When minorities (M= 236.5, SD= 280.72) and non-minorities (M= 974.5, SD= 1191.47) were compared in Sci-fi and non-Sci-fi films, no statistically significant difference in frequency of representation was found (t (2)= .85, p = 0.484). Thriller (N=30, 20%) was proportionately, the third most likely to feature minorities. The difference in minority (M= 236.5, SD=292.04) and non-minority (M=1218.5, SD= 1553.51) representation in Thriller and non-Thriller films was also not found to be statistically significant (t (2)= .88, p = 0.472). These results suggest that the genres that there is no difference in minority visibility in genres that are proportionately most likely to include minorities and other genres. A graph offering basic insight into the distribution of minorities in varying genres of Hollywood blockbuster films is included below (Figure 2).
23
Figure 2: Graph of minority visibility in Hollywood movie genres
Africans were the most prominently featured minority in the genres Thriller (58.6%), Adventure (34.1%), Action (56.6%), Fantasy(35.5%), Drama (42.2%), Sci-fi (52.6%), Romance (80.8%), Crime (82.4%) and Comedy (49.5%). Meanwhile, in Animation films, Arabs were the most prominently featured minority (42.9%).
Stereotypes dominantly present in minority representation and changes over time (RQ3, RQ4)
In total, 61 minority characters coded were stereotyped (N= 61, 5.50%). Out of these, 53 were definitely stereotyped and 8 were only somewhat stereotyped.
15,86% 34% 19,29% 22,35% 14,38% 7,73% 17,86% 15,59% 10,78% 20% 87,14% 66% 80,71% 77,65% 85,62% 92,27% 82,14% 84,41% 89,22% 80% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Comedy Crime Romance Sci-fi Drama Fantasy Action Adventure Animation Thriller
MINORITY VISIBILITY IN HOLLYWOOD MOVIE
GENRES
24
To answer RQ3 and RQ4, the data was treated qualitatively as the variable ‘Which stereotype does the character belong to?’ was coded as an open and qualitative question. Furthermore, the frequency of occurrence of each stereotype is low in the sample.
The most prevalent minority stereotype is thug (N= 9, 14.8%). The second most predominant stereotype is Wise (N=6, 9.8%) and Magical (N= 6, 9.8%). These stereotypes were followed by Maid (N= 5, 8.2%) and Arabs in the Dessert (N= 5, 8.2%). The exact count of all the other stereotypes per minority group is presented in Table 4. The dominant stereotype used to portray African characters in films is Wise (N= 5, 6.85%). Asians were mostly portrayed as Martial Arts experts (N= 3, 17.65%). Indians were mostly represented in Filler (N= 3, 15%) roles such as Taxi driver or convenient store owner. For Indigenous characters, the most frequent stereotypes used are Magical (N= 2, 28.57%) and Beautiful Maiden (N= 2, 28.57%). Arabs were stereotyped as Arabs in the Dessert (N= 5, 45.45%). Finally, Hispanics were dominantly stereotyped as Thugs (N= 5, 41.67%).
Concerning religion, Muslims were mainly stereotyped as Barbaric Arabs (N= 2, 22.22%) and Arabs in the dessert (N= 2, 22.22%). There was also one instance of Muslims being stereotyped as terrorists (N=1, 11.11%). The only stereotypical representation of Hindus was a filler character (N= 1, 100%) and Sikhs were represented as Thugs (N= 1, 100%).
Christians were stereotyped as Wise (N= 1, 100%). Jewish characters were not stereotyped.
Table 5: Frequency of occurrence of each stereotype per minority group
Minority Stereotype Years Frequency and
percentage of sample stereotyped
African Wealthy African 1988 4
Black Best Friend 1983, 2005 2
Maid 1988, 1992 4
Thug 1985, 2006 1
Wise 1982, 1984, 1985,
1999
25
Martial Arts 2016 1
Magical 1999 2
Filler 2011 1
Total 19 (1.7%)
Asian Martial Arts 1984, 1986, 2005 3
Immigrant 2005, 2010 2 Nerd 1985 1 Total 6 (.54%) Indian Filler 1997, 2010, 2016 3 Wealthy 1983 1 Thug 1983, 2006 2 Total 6 (.54%) Indigenous Magical 1995, 2012 2
Blood thirsty warriors 1995 1
Beautiful maiden 1995 2
In the wilderness 1988 1
Total 6 (.54%)
Arab Arabs in the dessert 1992 5
Barbaric Arab 1992, 1993 2 Terrorist 1994 1 Total 8 (.72%) Hispanic Thug 1984, 1990, 1994, 2007 5 Martial arts 2004 1 Maid 1992 1 Immigrant 1984 1 Latin lover 2011 1 Total 8 (.72%)
Muslims Arabs in the dessert 1992 2
Barbaric Arab 1992, 1993 2 Terrorist 1994 1 Total 5 (.45%) Hindus Filler 2016 1 Total 1 (.09%) Christians Wise 1982 1 Total 1 (.09%) Sikhs Thugs 1983 1 Total 1 (.09%) Total number of minorities stereotyped 61 (5.50%)
RQ4 was answered by manually assessing the open variable ‘Which stereotype does the character belong to?’. Timelines were created to assess the progression of minority stereotypes over time and are presented below the discussion of every minority included.
26
Africans are stereotyped less frequently now than ever before. The 1980s (N=11) had the most instances of African stereotyping. This number reduced in the 1990s (N= 9) and since 2000 (N=4) there have been few instances of African stereotyping. African stereotyping patterns also indicate that Africans are now portrayed in a more individual manner. For example, in the 1980s and 1990s, Africans are portrayed primarily as wise individuals, black best friends or magical people. These are all highly reliant on offering, fairly often, the white lead character guidance. However, since 2010, the two stereotypes used for African
characters have included Filler and immigrant. This indicates that they are part of wider society and their individual stories are being heard more frequently.
Figure 3: Changes in African stereotypes over time
Asians are only stereotyped in 5 instances. In the 80s, Asians were most likely to be stereotyped as martial arts experts (N=3). However, while this stereotype disappears in the 1980s, it returns in the early 2000s. In the 80s, the stereotype of the Asian nerd was also used (N=1). The only other stereotype used in reference to Asians was immigrant (N=1) in the 2000s. There were no instances of Asian stereotyping since the year 2005 which indicates that Asian characters are less frequently stereotyped on screen now than earlier.
27 Figure 4: Changes in Asian stereotypes over time
Comparatively, Indigenous stereotypes have remained static over time indicating no interest on behalf of Hollywood to present integrated images on indigenous individuals. In the 80s, Indigenous (N=1) people are only portrayed in the Wilderness. During the 90s, Indigenous representation increases but indigenous people are stereotyped as magical (N=1), beautiful and sexually viable maidens (N= 2) and blood thirsty warriors (N=1). In the 2000s, the stereotype of magical indigenous people is once again seen (N=1).
Figure 5: Changes in indigenous representation over time
As for Indian characters, in the 80s, Indians were stereotyped as Wealthy (N= 1) and Thug (N=1). In general, Indian characters have played a filler role (N= 3) since the 90s. This suggests that while they are represented as members of the community in multicultural societies, Hollywood has deemed their narratives unimportant for the general public. The notable exception to the filler role, is the 2006 representation of Indians as Thug again (N=1).
28 Figure 6: Changes in Indian stereotypes over time
All stereotyped representations of Arabs were from the 90s (N= 8). The year 1992 seems to be a mark a significant shift in stereotypes from Arabs being dominantly represented as being dessert dwellers (N= 5) to the creation of a more violent image of Arabs. Since 1992, Arab stereotypes used include, Barbaric Arab (N= 2) and Terrorist (N= 1). Therefore, Arabs are portrayed as foreign or violent.
Figure 7: Changes in Arab stereotypes over time
Hispanics are consistently portrayed as thugs through all the decades included in this study (N=5). Other Hispanic stereotypes in the 80s were Immigrant (N=1) and in the 90s were Maid (N=1). In the 2000s, Hispanic representation extends and enters into the territories of Martial arts experts and Latin Lovers (N=1). In general, the stereotyping of Hispanic characters has dramatically reduced with only two instances since the 2000s. However, they are nevertheless portrayed as violent, unassimilated and the women are sexually viable.
29 Figure 8: Changes in Hispanic stereotyped over time
Similar to Arabs, all stereotypical representation of Muslims coded in this study were from the 90s (N=5). Often times, Muslim was synonymous with Arab. Muslims were once again represented in a passive dessert dweller frame in 1992 (N=2). However, this stereotype became increasingly violent through the 90s with the stereotype of Barbaric Arab (N=2) and Terrorist (N=1) introduced.
The other religions stereotyped were Hindus (N=1), Christians (N=1) and Sikhs (N=1). Christians were stereotyped positively using the stereotype Wise. Meanwhile, Sikhs were stereotyped as thugs and Hindus were stereotyped as filler characters which are fairly negative portrayals of the religions.
Discussion
In this study, minority visibility, issues with visibility such as the influence of genres and whitewashing along with the existence and evolution of minority stereotyping were explored. By using a sample of 111 blockbuster films over 37 years, this study offers a more
comprehensive and improved picture of changes in ethnic and religious minority
representation over time which was previously lacking in the literature. The findings of this study are in accordance with the trends in media portrayals of minorities that are highlighted by Trebbe and Schoenhagen (2011). Indeed, minorities in Hollywood are underrepresented
30
when compared to the demographics in the US population and misrepresented through stereotyping. The two most striking findings of this study are that:
(1) Minority representation in Hollywood blockbuster films follows similar trends to minority representation in other mediums such as news and advertising.
(2) However, on a more optimistic note, there is a trend of decline in stereotyping of minorities in Blockbuster films for most racial and religious minorities.
Several findings in this study suggest that minority visibility in Hollywood blockbusters is similar to minority visibility in other mediums; that minorities are less visible than non-minorities. Africans are the most represented minority group. This finding is consistent with many other comparative studies that Africans are the most represented minority group (Dixon & Linz, 2000; Bowen and Schmid, 1997). Considering they are the largest minority group in the US, Hispanic visibility in Hollywood appears to be the lowest proportionately when compared to their demographic. This is also consistent with Santa Ana’s (2013) findings. Asians also experience underrepresentation however, to a lesser extent than Hispanics. Regardless, not all minorities are underrepresented.
Arabs seem to be the rare exception of an overrepresented minority in the sample. As Shaheen (2003) suggests, there have been few changes in Arab representation over time. Arabs have consistently remained the ‘other’ in Hollywood. Upon appearance, they are stereotyped, often negatively, 72.73% in this sample. Interestingly, unlike what most of the literature suggests, Arab representation did not increase and negative stereotypes were not visible in the sample post 9/11 (Nacos & Torres-Reyna, 2003). Perhaps the reason there was no instances of post- 9/11 stereotyping in this study’s sample is due to the fact that anti-Arab sentiments in the US preceded 9/11. Shaheen (2012) claims that in the 1990s, 45% of
31
fewer changes in entertainment media’s representation of Arabs post 9/11 than initially thought. However, the extensive negative stereotyping shows that overrepresentation of a minority group does not necessarily result in accurate representation.
In contrast to racial minorities, religious minorities are barely visible in Hollywood. When compared with the rest of the population, Hollywood characters are overwhelmingly areligious. Only Muslims are visible in Hollywood and their representation is stereotype laden and often synonymous with Arab. A possible explanation for this phenomenon is that the subject of religion is too taboo and dividing as most people are personally involved in their religion (Smith, 2001). As blockbuster films often intend to reach the widest possible audience, extensive religious depictions can have an alienating impact on the audience. Therefore, the capitalistic pressures of the movie industry could be a reason as to why religious minorities are virtually invisible.
Furthermore, stereotype analysis of minority characters confirms Holtzman and Sharpe’s (2014) idea that minority stereotypes do undergo changes in the media. African and Hispanic stereotyping trends suggest that these races are now being seen in a different light.
Previously, Africans were often caricatured as the Black best friend or as Wise. While these stereotypes are not negative, they did make depictions of black characters in Hollywood dependent entirely on their white counterparts. Similarly, Hispanics were mostly featured as thugs. However, for both minorities, recent stereotypes have expanded into the realms of ‘martial arts’. This suggests that representations have become individuated allowing these minorities to occupy their space in the public sphere outside of considerations of their race. However, stereotypes are not necessarily changing for the better for all races. Asian and Indian stereotyping have become increasingly homogenised. In the 2000s, Indians mostly played filler characters and Asians mostly played immigrants. These roles diminish the identity of members of these groups as individuals and force them to occupy the peripheries
32
of society. As a result, while there is definitely evidence of changes in stereotype over time, these are not necessarily entirely positive changes.
Upon evaluating the longitudinal sampling of this study, two suggestions for improvement can be made to assist trend analysis. The only evidence of change over time in quantitative analysis was that there was a decrease in whitewashing in Hollywood over time. Perhaps due to the manually coded sample size of this study being relatively small (N=30 per year) and time constraints, trends in minority stereotyping were less apparent. Most minority groups had too few instances of stereotyping to make generalisations about trends regarding them. Possible future research on this subject matter could be done using automated content analysis to derive a larger sample capable of analysing trends. Furthermore, using secondary source analysis such as IMDb, may have also impacted the accuracy of character stereotypes. Watching the films would have allowed the coder to improve the accuracy in stereotype coding.
Findings suggest evidence of a selection effect amongst the audience. Regarding the effect of genres on visibility, findings indicate that minorities are less likely to be featured than non-minorities in every genre of film. Furthermore, the most popular genres of blockbuster films, Comedy, Action and Adventure are less likely to include minorities in their cast than films of other genres. This finding is similar to that of Park, Gabbodan and Chernin (2006) indicating that some genres definitely create a more conducive environment for on-screen minority visibility than others. Therefore, society’s in-group feels more comfortable seeing depictions of minorities in certain contexts than others. These findings show that Hollywood create films for capitalistic purposes and respond to audience preferences over artistic purposes. However, this could be a limitation in only sampling Blockbuster films. Perhaps due to the selection effect mentioned, minorities are less visible in the most popular films. Minorities are likely to
33
be featured more in independent films or films created by minority directors which were not abundantly present in the sample.
When contextualising the findings of this study within the literature, two of Kidd’s (2011) concerns for minority underrepresentation and misrepresentation become apparent. Firstly, the underrepresentation of minorities means that society’s ingroup and other minorities are inherently unaware of individual minority needs. Secondly, underrepresentation and
misrepresentation leads to minorities feeling like outsiders in society. To minimise the effects of these consequences, further research should also be conducted into operationalising
analysis criteria similar to the Bechdel test. This could be used as an indication for
moviegoers to increase awareness on minority misrepresentation and underrepresentation in films. It will ultimately empower moviegoers with the ability to transition from ‘viewer- consumers’ to ‘viewer- citizens’ as it informs them on which media they are supporting (Kim, 2004).
In conclusion, minorities are proportionately underrepresented and frequently misrepresented. While these trends are prone to change over time, these changes are not always positive. Moreover, the presence of selection bias could potentially impact minority presence in different genres. Ultimately, as Hart (2005) suggests, minority visibility and accuracy in representation increases when consumer demand for minority visibility in Hollywood increases. As consumers, we must be more conscious of the choices we make in relation to films. The more aware we are as individuals on the individuality of minorities in society, the lesser the likelihood that outgroups are ignored in the public sphere.
34
Bibliography
1. Baum, M. A. (2003). Soft News and Political Knowledge: Evidence of Absence or Absence of Evidence?, Political Communication, 20(2), 173.
2. Bennett, W.L. & Edelman, M. (2003). Toward a New Political Narrative. Journal of Communication, 35(4), 156-171.
3. Bleich, E., Bloemraad, I., & Graauw, E. D. (2015). Migrants, Minorities and the Media: Information, Representations and Participation in the Public
Sphere. Journal of Ethnic and Migration Studies,41(6), 857-873. doi:10.1080/1369183x.2014.1002197
4. Boukes, M., Boomgaarden, H.G., Moorman, M., & de Vreese, C. H. (2015). At Odds: Laughing and Thinking? The Appreciation, Processing, and Persuasiveness of Political Satire. Journal of Communication, 1(5), 721.
5. Bowen, L., & Schmid, J. (1997). Minority Presence and Portrayal in Mainstream Magazine Advertising: An Update. Journalism & Mass Communication
Quarterly,74(1), 134-146. doi:10.1177/107769909707400111
6. Casero-Ripolles, A. (2012). Mas alla de los diarion: el consume de noticias de los joyenes en la era digital (ENGLISH). Communicare, 20(39), 151-158
7. Dale, E. (1935). The Content of Motion Pictures, New York: Macmillan Company.
8. Dixon, T. L., & Williams, C. L. (2014). The Changing Misrepresentation of Race and Crime on Network and Cable News. Journal of Communication,65(1), 24-39. doi:10.1111/jcom.12133
9. Eschholz, S., Bufkin, J., & Long, J. (2002). Symbolic Reality Bites: Women and Racial/ Ethnic Minorities in Modern Film. Sociological Spectrum, 22(3), 299.
35
10. Green, M. C., & Clark, J. L. (2012). Transportation into narrative worlds:
implications for entertainment media influences on tobacco use. Addiction,108(3), 477-484. doi:10.1111/j.1360-0443.2012.04088.x
11. Green, M. C., Garst, J., Brock, T. C., & Chung, S. (2006). Fact Versus Fiction Labelling: Persuasion Parity Despite Heightened Scrutiny of Fact. Media Psychology, 8(3), 267-285.
12. Hall, S. (1995) The Whites of their Eyes: Racist Ideologies and the Media, in G. Dines & J.M. Humez (Eds.), Gender, Race and Class in Media, pp.18-22, London: Sage Publications.
13. Hart, M. (2005). Gay markets meet growing demand: editors seek engaging stories on both general and genre-specific issues. The Writer, (7).
14. Holtzman, L., & Sharpe, L. (2014) Media Messages: What Film, Television, and Popular Music Teach Us About Race, Class, Gender and Sexual Orientation. Hoboken: Taylor and Francis.
15. Jacobs, L. (2016). Patterns of criminal threat in television news coverage of ethnic minorities in Flanders (2003–2013). Journal of Ethnic and Migration
Studies,43(5), 809-829. doi:10.1080/1369183x.2016.1217152
16. Jones, D. A. (2004). Why Americans Don’t Trust the Media: A Preliminary Analysis. The Harvard International Journal of Press/Politics,9(2), 60-75. doi:10.1177/1081180x04263461
17. Kidd, D. (2011). Representations of Race, Gender, Sexuality, and Disability in Reality Television. Conference Papers -- American Sociological Association, 182. 18. Kim, L. (2004). Be The One That You Want: Asian Americans in Television
Culture, Onscreen and Beyond. Amerasia Journal,30(1), 125-146. doi:10.17953/amer.30.1.356381678qj14622
36
19. Lee, C. (2013). Making race salient: Trayvon Martin and implicit bias in a not yet post-racial society. North Carolina Law Review, 91(1), 101.
20. Lum, L. (2009, February 04). The Obama Era: A Post-racial Society? Retrieved June 28, 2017, from http://diverseeducation.com/article/12238/
21. Mulligan, K., & Habel, P. (2011). An Experimental Test on the Effects of Fictional Framing on Attitudes. Social Science Quarterly (Wiley-Blackwell), 92(1), 79-99.
22. Nacos, B. L., & Torres-Reyna O. (2002 ). Muslim Americans in the News before and after 9-11. Paper presented at Restless Searchlight: Terrorism, the Media and Public Life Symposium, Cambridge, MA.
23. Park, G.C. & Lee, S.J. (2010). The Model Minority Myth and Stereotype and the Underachiever: Academic and social struggles of underachieving Korean
Immigrant High school Students. In R. Saran & R. Diaz (Eds.) Beyond
Stereotypes: Minority children of immigrants in urban schools (pp.13-27). Boston, MA: Sense Publishers
24. Park, J. H., Gabbadon, N. G., & Chernin, A. R. (2006). Naturalizing Racial Differences Through Comedy: Asian, Black, and White Views on Racial Stereotypes in Rush Hour 2. Journal of Communication,56(1), 157-177. doi:10.1111/j.1460-2466.2006.00008.x
25. Perez, R. (2014). Brownface Minstrelsy: "Jose Jimenez," the Civil Rights Movement, and the legacy of racist comedy. Ethnicities,16(1), 40-67. doi:10.1177/1468796814548233
26. Phillips, G. (2011). Reporting Diversity: The Representation of Ethnic Minorities in Australias Television Current Affairs Programs. Media International
37
27. QuickFacts. (n.d.). Retrieved June 27, 2017, from https://www.census.gov/quickfacts/
28. Roos, J. M., & Shachar, R. (2014). When Kerry Met Sally: Politics and
Perceptions in the Demand for Movies. Management Science,60(7), 1617-1631. doi:10.1287/mnsc.2013.1834
29. Said, E. W. (1978). Orientalism. New York: Pantheon Books.
30. Schmader, T., Block, K., & Lickel, B. (2015). Social Identity Threat in Response to Stereotypic Film Portrayals: Effects on Self-Conscious Emotion and Implicit Ingroup Attitudes. Journal of Social Issues, 7(1)l, 54-72.
31. Shaheen, J.G (2003). Reel bad Arabs: how Hollywood vilifies a people. The Annals Of The American Academy of Political and Social Science, 171.
32. Shimizu, C.P. (2016). Sex for Sale: Queens of Anal, Double, Triple, and the Gang Bang: Producing Asian/American Feminism in Pornography. In Dave, S.
Nishime, L. & Oren, T.G. (Eds.) Global Asian American popular cultures, New York: New York University Press. 235.
33. Smith, J. A. (2001). Hollywood Theology: The Commodification of Religion in Twentieth-Century Films. Religion and American Culture: A Journal of
Interpretation,11(2), 191-231. doi:10.1525/rac.2001.11.2.191
34. Trebbe, J., & Schoenhagen, P. (2011). Ethnic Minorities in the Mass Media: How Migrants Perceive Their Representation in Swiss Public Television. Journal of International Migration and Integration / Revue de lintegration et de la migration internationale. doi:10.1007/s12134-011-0175-7
35. Tseligka, E. D. (2016). Becoming theOther– stereotyping of Greeks by the German press. Continuum,30(6), 627-635. doi:10.1080/10304312.2016.1186604
38
36. Tsfati, Y., & Cappella, J. N. (2005). Why Do People Watch News They Do Not Trust? The Need for Cognition as a Moderator in the Association Between News Media Skepticism and Exposure. Media Psychology,7(3), 251-271.
doi:10.1207/s1532785xmep0703_2
37. Tukachinsky, R., Mastro, D., & Yarchi, M. (2015). Documenting Portrayals of Race/Ethnicity on Primetime Television over a 20-Year Span and Their
Association with National-Level Racial/Ethnic Attitudes. Journal of Social Issues,71(1), 17-38. doi:10.1111/josi.12094
38. Ugor, G. (2006) Demonizing the African Other, Humanizing the Self: Hollywood and the Politics of Post-Imperial Adaptations, Atenea, 26(2), 131-150.
39. Ungerleider, C.S. (1991). Media, minorities and misconceptions: the portrayal by and representation of minorities in Canadian news media. Canadian Ethnic Studies, 23(2), 158.
40. Weaver, H. A. (2012). Object lessons: a cultural genealogy of the dunce cap and the apple as visual tropes of American education. Paedagogica Historica,48(2), 215-241. doi:10.1080/00309230.2011.560856
41. Åström, B. (2015). The Symbolic Annihilation of Mothers in Popular
Culture:Single Fatherand the death of the mother. Feminist Media Studies,15(4), 593-607. doi:10.1080/14680777.2014.994100
39 Appendix 1:
Codebook
C1: Selecting the sample movies
The sample is selected using the website, Box Office Mojo
(http://www.boxofficemojo.com/). The highest grossing movies from the years 1980 to 2016 are searched on the website for every year. The time frame is then to be divided in a three year cycles. In every three year cycle, for the first year, the 1st, 4th and 7th films are selected. For every second year, the 2nd, 5th and 8th films are selected. For every third year, the 3rd, 6th and 9th films are selected.
C2. Selecting the sample actors and characters
After selecting the film, IMDb (http://www.imdb.com/) is used to identify the characters and actors to be coded from the film. The top 10 characters and actors from the IMDb cast list are selected for coding. For example, in the movie Star Wars: The Force Awakens, only the characters/ actors included within the red box are to be coded from the IMDb cast list.
40
The following questions are to be answered about every film on the list.
V1: Movie Name
The movie name is an open ended question. It should be answered according to the Box Office Mojo list included at the end of this codebook.
41
The film’s year of release is coded using an open ended question. The year of the film’s release is determined through the Box Office Mojo list rather than the IMDb page in order to maintain consistency.
V3: Main Production Company
This question is open ended. The answer for this question can be found on the film’s IMDb page under the subheading, Company Credits. Since most films include several production companies, only the first film production company is to be included as an answer for this section. For example, for the film, Star Wars: The Force Awakens, the image below shows that the answer for Main Production Company would be Lucasfilm in spite of the inclusion of Bad Robot and True North Productions in the Production Company list.
V4: Director
This is an open ended question. The answer for this question is to be determined using the film’s IMDb page. The director’s name is easily available directly underneath the primary image used in the film’s page. The image below offers an idea on where to find information about the director on a film’s IMDb page. For example, the director of the film, Star Wars: The Force Awakens would be J.J. Abrams.
42
V5: Genre of the movie
The film’s genre is coded using an open ended question and determined using the IMDb page. The film’s genre can be identified under the sub section of the IMDb page named, Storyline. In this section, a separate category for genre is included. Only the first three genres listed on IMDb should be coded as most movies are categorised as multiple genres. So, for the film, Star Wars: The Force Awakens, by applying the aforementioned rule to the image below, we can determine that the genre of the film is Action, Adventure and Fantasy.
43
For the top 10 characters listed in the IMDB cast list, fill out the following questions.
Note: When insufficient information is available about individual characters on IMDb the coder is permitted to supplement information from websites such as Wikipedia and Rotten Tomatoes.
V6: Gender
The character’s gender is determined by how the coder perceives the gender of the character. The following options are offered for the coding of gender.
- Male: A character is to be coded as male if they appear as male in the film. - Female: A character is to be coded as female if they appear as female in the film. - Transgender: A character is to be coded as transgender if they appear as transgender
in the film.
- Other: In case it is difficult to determine the gender of characters in films, the other category is offered for gender fluid and agender characters. For example, in some animated films, it may be difficult to be determine a characters’ gender.
