As Seen on TV: The Portrayal of Women
A quantitative and qualitative analysis of the presence and portrayal of women in Dutch television fiction and non-fiction
University of Amsterdam Graduate School of Communication Research Master in Communication Science
Thesis Supervisor: Dr. Jeroen S. Lemmens Student Name: Violet Luif
Student: 5698316
Abstract
Until now, virtually all studies that focus on the representation of men and women in media content, conclude that the number of women who appear in the media is much lower than the
proportion of women in the overall population. In the current study, the presence and portrayal of women on television have been studied by using a mixed-method approach. The
first study applied a content analysis to quantify the portrayal of women on television while the second study used focus groups with women of different age groups and levels of education to obtain a greater insight into their perception on the portrayal of women on television. The representation of women on television differs strongly for fiction and non-fiction programs. In non-fiction women are slightly underrepresented, however, among younger
age groups men and women are equally distributed. Men are overrepresented in nearly all levels of employment. In non-fiction, women are strongly underrepresented. Women seem to
be underrepresented in more information oriented genres, functions, levels of expertise and topics. While in the more informal genres, functions, levels of expertise and topics, men and
women seem to be equally distributed. The qualitative dimension of the study has indicated that women have a realistic notion of the portrayal of women on television. Causes that were
mentioned for the underrepresentation of women were based on three principles: male dominance, female subordinance and a more critical approach towards women by the
audience. Women seem to be quite indifferent towards the actual and perceived underrepresentation of women on Dutch television. They rejected the thought of interference
of the government in the representation of women in television programs.
Introduction
In February 2013, the Dutch Public Broadcasting Organization (NPO) announced their intentions to have a greater female presence in news and talk shows (Volkskrant, 2013). According to the Dutch Media Act, the NPO should broadcast a balanced image of Dutch society through its programs (Mediawet, 2008). This is why an accurate representation of the Dutch society in television programs in the period of 2010-2016 is one of the NPO’s priorities (Nederlandse Publieke Omroep, 2010). This means, among other things, that the organization wants to portray different social groups in society in an appropriate and proportionate way. Since half of the Dutch population consists of women (Centraal Bureau voor de Statistiek, 2013), half of the people appearing on Dutch public television should be female. Based on its own research in 2010, the NPO found that only 40% of the people who are visible on Dutch television with speaking roles were female (Monitor Representatie, 2010). According to the NPO this share is disproportionately small. In the current study, the knowledge about the representation of women in contemporary Dutch television is updated. Both the frequency of women on television and the role in which women are portrayed are examined. Furthermore, the current study aims to provide more insight into the way in which women perceive their representation on television. In order to analyze the portrayal of women from a quantitative and a qualitative perspective, a mixed-method approach will be used. The first study will use a content analysis to quantify the portrayal of women on television while the second study will use focus groups to obtain a greater insight into how women perceive their portrayal on television.
Scientific research substantiates the importance of a representation of society on television that corresponds with reality. Previous research has shown that the orientation of viewers towards their own and other social groups in society is influenced by the way in which the relationship between these groups is represented on television (Gerbner, Gross,
Morgan & Signorielli, 1994). According to studies on cultivation, the image that people have of the demographic structure of society is linked to their television viewing behavior. People who watch a lot of television estimate the demographic composition of society to be more comparable to the way it is represented on television than to the actual structure of society (Gerbner & Gross, 1976). Moreover, cultivation research asserts that a cultivation effect is most pronounced in contexts with which the audience is unfamiliar (Gerbner & Gross, 1976; Morgan & Signorielli, 1990; Pfau, Mullen & Garrow, 1995). In addition, research from the field of social psychology has demonstrated that the media also play a role in establishing a picture of the relative status of social groups within a broader socio-cultural context. For example, research has identified a correlation between exposure to television content in which men and women are portrayed in a stereotypical manner and having a stereotypical image of men and women (Signorelli, 1990). In summary, it can be stated that what is being broadcasted on television is associated with the worldview of television viewers (Bandura, 2002). More specifically, television content is related to the maintenance and transfer of stereotypical images of males and females.
The representation of men and women in media content has been studied extensively in other countries. Particularly in the areas of advertising (e.g. Bartsch, Burnett, Diller and Rankin-Williams, 2000; Stern & Mastro, 2004) and fiction (e.g. Lauzen, Dozier & Horan, 2008; Signorielli & Kahlenberg, 2001; Davis, 1990). Virtually all studies that focus on the representation of men and women in media content, conclude that the number of women who appear in the media is much lower than the proportion of women in the overall population. In addition, it is clear from previous studies that women are not only underrepresented in comparison to men, they are also depicted in a very different light. Generally, prevailing gender stereotypes in the media are confirmed in these studies; women are depicted more often in the domestic realm while men are more commonly portrayed in a work-related
context (e.g. Bartsch et al, 2000; Lauzen & Dozier, 2005; Lauzen et al, 2008; Mastro & Stern, 2004).
In the Netherlands, very few studies on the image of different social groups in television programs have been conducted. The male-female ratio in Dutch television programs has been examined in research by Koeman, Peeters and D' Haenens (2007). They used content analysis on both fiction and non-fiction programs to test whether public and commercial broadcasters provided a representative picture of Dutch society in their programs. They found that men were, in both fiction and non-fiction programs, present more often than women. It was also found that men were portrayed as experts significantly more often than women. Furthermore, women were more often associated with themes like education, family and relationships. Summarizing, they found that neither the public nor the commercial broadcasters provide a representative picture of Dutch society in their programs. Regardless of the matters that women were associated with, the study by Koeman, Peeters and D' Haenens (2007) did not focus on the position that women fulfill within a program (e.g. presenter, assistant or guest). It is therefore interesting to examine whether the position within a program shows similar role-reinforcing patterns. In the present study, the different aspects from research introduced above are combined, updated and supplemented with contemporary data on the representation of women on Dutch television. Through a content analysis of a representative sample of the entire television offering, a complete and up-to-date picture of the representation of gender in Dutch fiction and non-fiction on television will be provided, focusing on frequency and distribution of roles and status.
Despite the fact that in the past, a relatively large amount of quantitative research on the representation of women in various media has been published, the research field lacks the combination of a quantitative and a qualitative dimension. This is why in the present study, in addition to a quantitative content analysis, the qualitative aspect of the representation of
women in Dutch television content is also discussed, with a view to answering the following: How do women perceive their own representation on television?
To date, several researchers have tackled the issue of how social groups experience their own representation on television. For instance Healey and Ross (2002) focused on the representation of age groups and used focus groups with elderly people to examine the perception of seniors of their representation in television content. Mendez-Mendez and Alverio (2002) focused on ethnicity and studied how Latinos perceived the manner in which they were portrayed on television. Both studies indicated that these groups not only felt underrepresented in television content, but also that they could not identify with the manner in which they are portrayed, they perceived their portrayal as negative and failing to correspond with reality. The current study will examine how women experience the representation of ‘the woman’ on television. The emphasis will be, as in the quantitative study, both on the frequency, the role and the status of women on television. Furthermore, this study will demonstrate how women perceive the influence of the current representation and if they think that a representative image of women on television should be actively encouraged.
Theoretical Framework
Distribution of Gender Representation in Fiction
Over the last fifty years, many researchers have studied the representation of women on television, especially in the United States, the representation of gender in fictional television shows has been studied extensively. In 1975, McNeil published a study on the representation of women in U.S. prime time television drama. It was found that in this genre, 32% of all the characters were female. These results were substantiated by a study by Greenberg (1980) who analysed three seasons of U.S. prime time and Saturday morning 6
television drama in the mid 1970’s and found female characters comprised 29% of the total characters portrayed. In 1989, Signorelli conducted a study on the evolution of the portrayal of women in American prime time fiction between 1969 and 1985. The percentages reported in this study were similar to those found previously; approximately 30% of the characters portrayed were female. Furthermore, Signorelli (1989) found a significant increase in female quota. In the 90’s the proportion of female characters in U.S. prime time had increased to 40% (Elasmar, Hasegawa & Brain, 1999; Lauzen & Dozier, 1999). More recent research by Lauzen, Dozier & Horan (2008) confirms the share of 40% female characters in television fiction. Overall it seems that the proportion of female characters in American fiction is increasing. However, the distribution between female and male characters is still not representative of the population.
The presence of women in television fiction has not been studied solely in America. In the Netherlands, various studies, conducted throughout the last three decades show that female characters in Dutch fiction are fundamentally underrepresented in relation to their proportion in the population, and in relation to their male equivalents (Bouwman, 1987; Eie, 1998; Koeman, Peeters, & D’Haenens, 2007; Nederlandse Publieke Omroep, 2008; Sterk & Van Dijk, 2003). Similar to their American counterparts, these studies found that the proportion of female characters has increased slightly over time. In 2010 the share of women present in Dutch television fiction had increased up to 43% (Nederlandse Publieke Omroep, 2008). However, the distribution between female and male characters is still not representative of the population.
Gender Role Representation in Fiction
Research on representation of gender on television does not focus solely on the frequency in which women are presented on television; the way in which they are presented
is also part of the representation. One aspect of the representation of gender on television is the role that characters play within a program. In a content analysis of U.S. fiction between 1953 and 1979, Dominick (1979) found that the percentage of female characters in a major role varied between 25% and 35%. In comparable research, Greenberg and Collette (1997) found that the percentage of female major characters, introduced between 1966 and 1992 was 35%. In the late 90’s, Lauzen and Dozier (1999) conducted similar research in which they found that 43% of all major characters were female. More recent research by Lauzen, Dozier & Horan (2008) confirms the share of approximately 40% female characters in television fiction. Women seem to be not only increasingly present in television fiction, the proportion of women in major roles also seems to increase. However, it is critical to be mindful of the fact that female characters remain significantly underrepresented
The role that characters fulfil within a program has also been investigated in American fiction. Several content analyses of American prime time television have shown that men are more often portrayed with a job than women (Glasscock, 2001; Lauzen et al, 2008; Signorielli & Kahlenberg, 2001). In addition to their labour status, women in American prime time television are more often portrayed in interpersonal roles related to household and care (Greenberg & Collette, 1997; Signorielli & Bacue, 1999). Men, on the other hand, are relatively more often portrayed in leadership positions and with clearer career goals than women (Lauzier & Dozier, 2004).
A similar representation the role of men and women within the household was found in Dutch prime time television fiction (Emons et al, 2010). A content analysis on prime time fiction programs, conducted by the Dutch public broadcasting (2008) found that male characters focus mainly on their career while female characters focus either on household and care, or they combine these activities with their work (Nederlandse Publieke Omroep, 2008).
Gender Representation in Non-Fiction
The above-mentioned studies focus solely on television programs fiction. However, portrayal of gender within a non-fiction television program is equally relevant when considering the overall portrayal of women on television. Previous studies on the representation of women in non-fiction television programs focused mainly on news broadcasts. Just as in fiction, the main conclusion in these studies is that women are quantitatively underrepresented. A content analysis by the Global Media Monitoring Project (1995) that was conducted simultaneously in 71 countries demonstrated that women are globally underrepresented in news broadcasts. This study showed that only 21% of the persons appearing in news broadcasts are women (Gallagher, 2006; Spears and Seydegart, 2000). The persistent underrepresentation of women in news broadcasts is remarkable. Through the last decades, women have been increasingly represented in leading positions in commercial and political domains. However, on television, they remain strongly underrepresented compared to men (De Swert & Hooghe, 2010).
Just as in fiction, women are not only quantitatively underrepresented in news broadcasts. The way in which they are portrayed is also stereotypical (Carroll and Schreiber, 1997; van Zoonen, 1998). Recent studies show that women in news broadcasts are often associated with feminine topics like family, childcare and health. Subsequently, women are underrepresented in news items on economy, foreign affairs and finance (Craft and Wanta, 2004).
Stereotypes
The above overview of content analyses on the representation of gender in both fiction and non-fiction does not only demonstrate an underrepresentation of women in the media but also a stereotypical image. Stereotypes are not necessarily negative. Stereotypes 9
are certain attributes that are assigned to a social group, without taking into account individual differences within the social group. In general, people have the tendency to evaluate social groups, other than the social groups in which they themselves belong, as homogeneous (Mullen and Hu, 1989). Stereotypes are intended to organize information in an orderly manner and to make it possible for people to make quick decisions in (complex) situations (Allport, 1954). Even though the media may not always portray a reflection of reality, research has shown that stereotypes are partly shaped by what people see and hear in the media (Babad, Birnbaum and Benne, 1983; Smith & Granados, 2009; Van Dijk, 1995). This implies that the media can also contribute to stereotypical images. Traditional gender stereotypes posit that men represent the ideal, or norm, against which women are judged. As such, women become the perpetual other, valued primarily in their relations to others, men in particular (Donelson, 1999). When multiple programs on television repeat these gendered roles, they will be perceived as increasingly valid and credible (Merskin, 2006).
Social role perspective
The social role perspective argues that the observed distribution of women and men in social roles such as interpersonal and work roles underlies gender stereotypes (Eagly & Steffen, 1984). When applied to television content, this suggests that the basic social roles assigned to female and male characters are important contributors to the construction and maintenance of gender stereotypes (Eagly, 1987). The contribution that both fictional and non-fictional characters on television have on viewers’ expectations and beliefs about gender illustrates the importance of increased insights on the current state of gender representation in the television landscape. This is why in the current study the representation of men and women in Dutch fiction and non-fiction is quantified.
Media as Reflection of Reality
The media does not only construct a reality, it can also be understood as a reflection of reality (Van Dijk, 1995). Television represents the contemporary view on society as a whole as well as (stereotypical) beliefs about various groups of people within society (Westerik, Douma and Konig, 2007). In the current labour market, men and women seem to be treated in an increasingly fair and equal manner. According to the Emancipation Monitor 2012, published by the Central Bureau for Statistics, a woman’s gross hourly salary was nearly 18 percent lower than for men. However, women between 25 to 30 years earned slightly more per hour than men of the same age (Central Bureau voor de Statistiek, 2012). Research on the current emancipation climate in the labour market did not only identify an increasing salary for young women, it also implied that although the situation in the labour market has worsened in recent years due to the economic crisis, women managed to maintain their position in labour participation. From 2008 to 2009, the net employment of women between 20 to 64 years, increased slightly to 64%. This percentage did not change in 2010 and 2011 (Central Bureau voor de Statistiek, 2012). The current study explores whether this trend is also reflected in Dutch television content.
Content Analysis
Summarizing, this research focuses not only on the quantity of men and women in Dutch fiction and non-fiction. Building on previous research, the current study examines whether female and male characters continue to inhabit traditionally gender-typed roles. To explore how men and women are portrayed on television, the specific functions or role of people in Dutch fiction and non-fiction programs broadcasted by public and commercial broadcasters will also be examined. By looking specifically at the role of women in fiction and the function of women in non-fiction television programs, this research is intended to contribute to the understanding of the representation of women on television.
Gender Representation - The Perception of Women
The focus in studies on the representation of gender on television is, as in the studies described above, often on quantitative content analyses of television programs. However, studies on the reception of gender representation have also been conducted. These studies focus mainly on the correlation between viewing behaviour and stereotypical sex-role beliefs. In 1975, a survey based study among children indicated a relation between the amount of television a person watches and traditional sex-role beliefs (Beuf, 1974; Freuh & McGhee, 1975). Gunter & Wober (1982) combined a media diary with a questionnaire and discovered that adult heavy viewers were also more likely to have stereotypical gender beliefs than non-heavy viewers. These studies focus mainly on the effect of gender representation. Previous studies on how women perceive women in media content have mainly been conducted in the field of advertising (Duker & Tucker, 1977; Leavitt, 1978; Ezekiel, Edegoh & Emmanuel, 2013; Ford & Latour, 1991; Ford & Latour, 1993). Recent research on the perception of women on the representation of women on Dutch television or an existing framework of reference on this topic is lacking. This is why focus group research is particularly suitable to explore this topic. Focus groups are suited to explore phenomena and to identify frames that people use to comprehend the world (Miller & Glassner, 2010). Similar to the quantitative study, the emphasis is both on the frequency, the role and status of women in Dutch television fiction and non-fiction.
Method Study 1
Sample for Quantitative Study
In this study, ten Dutch television channels were analysed. These included the public broadcasters (Nederland 1, Nederland 2 and Nederland 3) and the largest commercial broadcasters, RTL (RTL4, RTL5, RTL 7, RTL 8), and SBS (SBS 6, Net 5 and Veronica).
The measurement period was five weeks, beginning on September 2, 2013 and ending on October 6, 2013. The programs on each channel were examined for one composite week. This implies that for each channel, one day in every five days was monitored (so one Monday, one Tuesday, etc.). All Dutch fiction and non-fiction programs that were broadcasted completely or partly between 18:00 and 24:00 on one of the channels in the sample were analysed. In total, 593 programs were coded. In all programs within the selection, the first 20 people that were visible while speaking were coded. In total 4857 persons have been coded. 2486 persons appeared on the public broadcasting channels, 2371 persons appeared on the commercial channels. The focus in this study is solely on Dutch speaking people. Foreigners are also fairly often portrayed in Dutch programs (e.g. news). They may affect people's perception of the world, but they do not contribute to the representation of the Dutch population.
Procedure for Quantitative Study
Coding procedure. A team of five coders conducted the coding of the television content. The coders were trained in the coding scheme and then practiced coding together and on their own. The coding scheme went through a number of revisions to ensure that it was clear. For the retrieval of information, the coders based their substantiated coding both on the observed visible features of the person as well as written and spoken information within the program. Insinuation was not considered a solid basis for information retrieval. The coders randomly selected each program, with 8% of the programs coded by two coders to determine inter-coder reliability. Both Krippendorff’s Apha and Holsti’s Method were used to establish the inter-coder reliability. Inter-coder reliability was adequate for all variables. For Krippendorff’s Alpha, the observed agreement ranged between .85 and .99 where Krippendorff’s Apha inter-coder reliability above .67 is considered reliable. For Holsti’s
method, all observed agreements ranged between .85 and 1 where an inter-coder reliability above .80 is considered as reliable (Lombard, Snyder-Duch & Bracken, 2002).
Measures for Quantitative Study
Genre. For every program in the sample a genre was coded. Thirteen main genres have been coded; fiction, news, current events, opinion programs, religious programs, art & culture, human interest, education, games & quizzes, sport, entertainment shows, satire and music.
Features. For each person, the first variables that were coded were gender and age. An estimation of the age category would be made based on visible, written and spoken information. The age categories were child (0-12 years old), teenager (13-19 years old), young adult (20-34 years old), adult (35-49 years old), middle adult (50-64 years old) and senior (65 years or older).
Topic. For each person, one main topic of conversation was coded. The list of topics consisted of the following options: ‘art/culture/entertainment’, ‘criminality/legal processes’, ‘accidents/disasters/war/terrorism’, ‘economics/finances/work’, ‘environment’, ‘healthcare’, ‘politics’, ‘science/education’, ‘multiculturalism/integration’, ‘gender emancipation’, ‘sports’, ‘family/relations’ and ‘consumer matters/spare time’.
Roles (fiction). With respect to roles, the significance of the part a person played was the main consideration. In assessing the roles in fiction programs, a distinction was made between leading role, supporting role or extra.
Professional status (fiction). For fiction, the level of employment of the character was coded. The options for level of employment were higher educated profession, middle educated profession, lower educated profession, student, retired and unemployed.
Functions (non-fiction). When defining function, the main consideration is the part a person fulfils within a non-fiction program. In assessing function in non-fiction programs, a distinction was made between internal and external functions. Internal functions were fulfilled by people who are professionally involved in the creation of the program. The internal functions consisted of four categories: presenter, correspondent, assistant and judge. All the people who were not professionally involved in the program had an external function. The external functions were further divided into six categories: interview subject, the subject of portrait, discussion participant, game contestant, speaker or performer. Furthermore, the level of expertise was coded for each external function. The level of expertise consisted of four functions: layman, expert, prominent person and speaker from experience.
Method study 2
Sample for Qualitative Study
For the qualitative part of the study, focus groups were conducted with women from three different age groups and two levels of education. For the focus groups a total of 29 women were recruited through snowball sampling and institutions such as schools and elderly homes. In total, six focus group interviews were carried out; they ranged in size from four to seven participants. The three different age categories were 18-29 years, 30-49 years, and 50 years and older. The education level was divided into two categories, higher and lower education. General information on the group size, age and television viewing behaviour of the participants can be found in Table 1.
Table 1: Focus group size, age of participants and hours of TV per week
Age group Education level Group size Age Hours TV per week
Mean SD Mean SD 18-29 Lower educated 4 19.5 1.7 19.5 20.9 18-29 Higher educated 6 24.7 1.0 10.5 8.2 30-49 Lower educated 4 38.0 2.9 14.8 5.3 30-49 Higher educated 4 45.3 1.7 9.0 3.6 50+ Lower educated 7 80.6 3.6 25.2 4.1 50+ Higher educated 4 65.3 2.5 19.5 6.5
Procedure for Qualitative Study
At the beginning of each interview the moderator introduced herself, the procedures and guidelines were explained and the presence of an audio recorder was pointed out. The participants were then asked to introduce themselves to the group and to the moderator. All focus group meetings were conducted in a semi structured way. An interview guide was prepared beforehand to make sure that the same questions and issues were addressed in each group (see Appendix B). However, if the conversation would move into another direction, the moderator would follow the participants in their thoughts and return to the interview guide later on. This meant that the order of the topics could vary from one group to another. The focus group consisted of three main parts, the first part addressed the participants’ general television viewing behaviour, the second part concerned the perception of the participants on
the representation of women in fiction and in the third part the participants spoke about their perception of the portrayal of women in non-fiction.
To make sure that the participants agreed on the interpretation of the moderator, the moderator would summarize what had been discussed after each part of the focus group. This gave the participants the opportunity to alter their previous opinions or to correct the interpretation of the moderator. At the conclusion of the focus group, participants were asked to complete an anonymous questionnaire on their demographic information and the amount of hours they watched television per week. After this they were debriefed and thanked for their participation.
Each focus group was transcribed directly after the interview. Grounded theory as defined by Glaser and Strauss (1967) was applied in this research as a framework to study the qualitative data. The coding procedure in the current study occurred in accordance with Corbin and Strauss (1990) and has been carried out in three stages. The first stage is referred to as ‘open coding’ and entails the thorough analysis of the complete transcripts by allocating codes to relevant quotes or paragraphs to create an initial impression of the tendency of the text and its main concepts. In the second stage, the ‘axial coding’, the labels that are identified in the first stage were summarized under initial concepts. During the third step, or the ‘selective coding’, the core categories were identified and linked to other categories (Corbin & Strauss, 1990). If additional refinement was needed, supplementary coding has been conducted on the initial material. This process has been repeated until no more additional useful information could be extracted from the data (Bryman, 2008). The concept indicator model that has been retrieved from the analysis can be found in Appendix A, the interview guide can be found in Appendix B.
Measures for Qualitative Study
General viewing behaviour. Participants were invited to share their own television viewing behaviour with the group, they would discuss what, when, why and how they watch television. Furthermore they were asked to describe what makes a program either attractive or unattractive. After this, the participants were asked whether the persons that contributed in the program made a difference in the appeal of a television program. This question served as a bridge to the next part of the interview that was concerned with the perception of the participants on the representation of women in Dutch television fiction.
Fiction. In this second part the participants were asked to name as many titles of Dutch fiction programs as possible. After this the participants had to estimate how the quota of men and women in Dutch television fiction was distributed. Then the role of women, the perception of this role, the causes and consequences, the ideal situation and the possibility for active policy were discussed. After a short break, the third part of the conversation would start. The moderator would announce that the conversation would move to Dutch non-fiction television.
Non-fiction. The participants were first asked to write down as many Dutch non-fiction titles or genres as possible. When finished, they were asked to read what they had written out loud and the moderator would categorize these titles and genres into a genre overview. After this the participants were asked to indicate what they thought the ratio of males to females was in each genre. Similar to the discussion on fiction, the role of women, the perception of this role, the causes and consequences, the ideal situation and the possibility for active policy were discussed for non-fiction in general and per genre.
Results Study 1
Gender Distribution on Dutch Television
In total 4857 Dutch speaking persons have been coded; 2486 persons appeared on the public broadcasting channels, 2371 persons appeared on the commercial channels. The gender distribution among these people was not equal χ2 (1) = 421.61, p = < .01 (n = 4857). Just over a third (35.3%) of the people present on Dutch television were women (n = 1713). The distribution of gender did not differ significantly between commercial (35.7%) and public broadcasters (34,8%), χ2(1) = 0.42, p = n.s. (n = 4857).
Gender Distribution in Fiction
Within the total sample of 4857 persons, 838 fictional characters (17.3%) were coded. The gender distribution among these 838 characters was not equal, 44.4% was female χ2
(1) = 10.54, p = < .05. (n = 838).
Between the six age groups, the relative distribution of men and women in television fiction did differ significantly χ2
(6) = 30.08, p = < .01 (N = 838). However, the only age groups in which women were significantly underrepresented were adults (36.9%) and middle adults (31.5%). Within the age groups children, teenagers, young adults and senior adults no significant difference between the relative share of men and women was found. A complete overview of the distribution of gender across the different age groups can be found in Table 2.
Table 2: Gender Distribution across Age Groups in Fiction Men Women Share (%) N Share % N Children 56.3% 9 43.7% 7 Teenagers 46.3% 105 53.7% 122 Young Adults 46.5% 86 53.5% 99 Adults 63.1%* 166 36.9%* 97 Middle Adults 68.5%* 87 31.5%* 40 Senior Adults 63.2% 12 36.8% 7
Note: * indicates a significant inequality in gender distribution, at least p <.05
The relative distribution of men and women on television differed significantly per role type χ2
(3) = 24.01, p = < .01. (n = 838). Within the sample, 341 leading roles were coded. The relative share of men and women playing leading roles in Dutch television fiction did not differ significantly χ2(1) = 0.50, p = n.s. (n = 341). The largest group in the sample was the supporting roles (n = 360). Contrary to leading roles, supporting roles were slightly more often played by men (56.5%) than by women (43.5%) χ2(1) = 5.88, p = < .05. (n = 360). The largest difference between the sexes has been found within the gender of the extras. They were more often male (71.3%) than female (28,7%) χ2(1) = 23.45, p = < .01. (n = 129). If the extras on television are not taken into account and the focus is on the main and supporting roles, no significant difference between the distribution of men and women was found χ2(1) = 1.55, p = n.s. (n = 701).
For a total of 472 characters present in Dutch television fiction a level of employment could be deduced. The distribution of men and women did differ significantly per portrayed level of employment χ2(8) = 23.45, p = < .01. (n = 838). Women were underrepresented in the three main employment categories; higher (32.5%), middle (32.9%)
and lower (34.1%) educated profession. Within the largest level of employment category, students (n = 193), no significant difference between the relative share of men and women was found χ2(1) = 0.01, p = n.s. (n = 193). A complete overview of the distribution of gender across the different levels of employment can be found in Table 3.
Table 3: Gender Distribution in Fiction across Levels of Employment
Men Women
Share (%) N Share % N
Higher Educated Profession 67.5%* 83 32.5%* 40 Middle Educated Profession 67.1%* 55 32.9%* 27 Lower Educated Profession 65.9%* 27 34.1%* 14
Student 49.7% 96 50.3% 97
Retired Person 44.4% 4 55.6% 5
Unemployed 85.7% 6 14.3% 1
Note: * indicates a significant inequality in gender distribution, at least p <.05
Gender distribution in Non-fiction
Overall, 4019 people were coded within Dutch non-fiction. The gender distribution in non-fiction was not equal, 34.4% of the coded people were women χ2(1) = 444.78, p = < .01. (n = 4019).
Between the six age groups, the relative distribution of men and women in television non-fiction did differ significantly χ2 (6) = 113.81, p = < .01 (N = 4019). Except for the categories children and teenagers, in which the representation was equal, women were underrepresented in all age categories. The relative share of women in the group middle adults was lowest, not even a quarter of the persons in this group was female (23.2%). A
complete overview of the distribution of gender across the different age groups can be found in Table 4.
Table 4: Gender Distribution across Age Groups in Non-Fiction
Men Women Share (%) N Share % N Children 49.4% 44 50.6% 45 Teenagers 44.1% 49 55.9% 62 Young Adults 58.4%* 572 41.6%* 408 Adults 69.1%* 1187 30.9%* 530 Middle Adults 76.8%* 661 23.2%* 200 Senior Adults 63.1%* 164 36.9%* 96
Note: * indicates a significant inequality in gender distribution, at least p <.05
The relative gender distribution differed significantly per genre χ2(12) = 261.23, p = < .01. (n = 4019). A distinction can be made between male genres and genres in which the representation of men and women is equally distributed. Typical male genres are news (28.9%), current events (27.1%), opinion programs (26.5%), art & culture (20.0%) and sports (3.0%). Genres in which the distribution of men and women does not differ significantly are religious programs (50%), games & quizzes (51.7%), entertainment shows (44.9%), satire (48.5%) and music (44.2%). A complete overview of the distribution of gender across the different genres can be found in Table 5.
Table 5: Gender Distribution across different Genres Men Women Share (%) N Share % N News 71.1%* 647 28.9%* 263 Current Events 72.9%* 304 27.1%* 113 Opinion Programs 73.5%* 125 26.5%* 45 Religious Programs 50.0% 9 50.0% 9
Art & Culture 80.0%* 32 20.0%* 8
Human Interest 62.3%* 661 37.7%* 400
Education 61.4%* 62 38.6%* 39
Sports 97.0%* 292 3.0%* 18
Games & Quizzes 48.3% 138 51.7% 148
Entertainment Shows 55.1% 172 44.9% 140
Satire 51.5% 17 48.5% 33
Music 55.8% 43 44.2% 34
Note: * indicates a significant inequality in gender distribution, at least p <.05
A distinction was made between internal and external functions. Internal functions were fulfilled by people who were professionally involved in the creation of the program, all the people who were not professionally involved in the program were coded as an external function. The relative distribution of men and women on television differed significantly per internal function χ2(5) = 46.53, p = < .01. (n = 4019). Within the sample, 227 of the 789 internal functions were fulfilled by women (29.0%). Except for the category assistant, each type of internal function was significantly more often fulfilled by a man than by a woman. Compared to men, women were most underrepresented in the function of correspondent
(12.1%). A complete overview of the distribution of gender across the different internal functions can be found in Table 6.
Table 6: Gender Distribution in Non-Fiction across External Functions
Men Women Share (%) N Share % N Presenter 66.7%* 273 33.3%* 136 Correspondent 87.9%* 153 12.1%* 21 Assistant 60.7% 17 39.3% 11 Judge 77.8%* 49 22.2%* 14
Note: * indicates a significant inequality in gender distribution, at least p <.05
Within the complete sample, 3230 external functions have been coded. The relative distribution of men and women on television differed significantly per external function χ2
(7) = 67.76, p = < .01. (n = 4019). The most present internal function was the interview subject (n = 609). Women were underrepresented within this function (31.9%) The category discussion participants had the lowest ratio of women (18.5%). The two categories in which the distribution of gender was not significant were the categories game contestant (51.7%) and performer (47.9%). A complete overview of the distribution of gender across the different external functions can be found in Table 7.
Table 7: Gender Distribution in Non-Fiction across External Functions Men Women Share (%) N Share % N Interview Subject 68.1%* 1303 31.9%* 609 Subject of Portrait 65.4%* 306 34.6%* 162 Discussion Participant 81.5%* 44 18.5%* 10 Game Contestant 48.3% 112 51.7% 120 Speaker 79.6%* 74 20.4%* 19 Performer 52.1% 37 47.9% 34
Note: * indicates a significant inequality in gender distribution, at least p <.05
When looking at level of expertise, the relative distribution of men and women on television did differ significantly per level of expertise χ2
(5) = 256.14, p = < .01. (n = 4019). Men and women are equally distributed among the levels of expertise layman and speaker from experience. The level of expertise in which women were relatively most underrepresented was the expert (n = 1371). Only 19.3% of the experts present in Dutch non-fiction television were female. A complete overview of the distribution of gender across the different levels of expertise can be found in Table 8.
Table 8: Gender Distribution in Non-Fiction across Levels of Expertise Men Women Share (%) N Share % N Layman 52.5% 555 47.5% 503 Expert 80.7%* 1106 19.3%* 265 Prominent Person 60.4%* 151 39.6%* 99
Speaker from Experience 55.0% 176 45.0% 144
Note: * indicates a significant inequality in gender distribution, at least p <.05
Men and women in Dutch non-fiction programs talked about different topics. With three exceptions, all topics were relatively more often discussed by men. The four only topics on which men and women equally commented were healthcare (48.6%) family/relations (51.4%), gender emancipation (53.8%) and multicultural society (44.4%). The topics in which women were most strongly underrepresented were sports (9,0%) and politics (16.5%). The distribution of gender with respect to other topics of conversation can be found in Table 9.
Table 9: Gender Distribution in Non-Fiction across Topics of Conversation.
Men Women
Share (%) N Share (%) N Art/ Culture/ Entertainment 56.1%* 420 43.9%* 328 Criminality/Legal Processes 72.1%* 235 27.9%* 91 Accidents/ Disasters/War/Terrorism 70.6%* 151 29.4%* 63 Economics/Finances/Work 71.6%* 194 28.4%* 77 Environment 71.3%* 102 28.7%* 41 Healthcare 51.4% 128 48.6% 121 Politics 83.5%* 343 16.5%* 70 Science/Education 74.2%* 49 25.8%* 17 Multicultural Society/Integration 55.6% 10 44.4% 8 Gender Emancipation 46.2% 12 53.8% 14 Sports 91.0%* 403 9.0%* 40 Family/Relations 48.5% 175 51.5% 186
Consumer Matters/Spare Time 61.1%* 368 38.9%* 234 Note: * indicates a significant inequality in gender distribution, at least p <.05
Results Study 2
To examine women’s perception of their portrayal on Dutch television, six focus group interviews were carried out. Most participants seemed quite comfortable with discussing their evaluations of the portrayal of women on television. They made reference to a variety of different programs and genres to illustrate their perceptions. Even though women of different age groups and different levels of education approached the topic from differing social positions and with varying levels of eloquence, the perceptions of women on the
representation of women in both fiction and non-fiction were remarkably similar throughout the six groups.
Perception of the Portrayal of Women in Fiction
A clear picture on the gender distribution in Dutch television fiction was not prevalent in any of the focus groups. When participants attempted to estimate the relative share of women present in television fiction, they estimated that the gender distribution was approximately 50/50. However, the dominant thought was that this distribution was not similar among different program types or genres. The same principle applied with the roles that women play in fiction, these also differed per program type or genre. Participants could not identify a dominant role type. Although they did not identify dominant role types, the participants did accentuate that female characters in series are often one dimensional:
Reina (46 years old, higher educated): “Everything is very clearly portrayed according to concepts. (…) The bitch is always a bit heavy and not that pretty and the women that are fun are also always beautiful.”
However, they accentuated that a fictional character cannot have too many dimensions, to ensure that a character is coherent for the viewer. This applied both on female and male characters.
Moniek (23 years old, higher educated): “I don’t think there is room for that, because that makes it a very ambiguous character. I think that in series it is very important that it is clear which role it is and if you add different layers, and she is cold but also very sweet, the character wouldn’t be clear.”
The participants claimed that dependent and independent women are present in Dutch television fiction. In the two older groups a shift of the role of women over time into a more independent direction was mentioned. This was perceived as a reflection of a changing society.
Ans (79 years old, higher educated): “I think that the situation has changed a bit, I think that the traditional stereotype of a woman that is always in the kitchen with her apron on while taking care of her family, is slightly outdated.(…) Back in the day, that came across as normal, that was just how it was.”
Among all groups, women mentioned that they are not actively aware of the distribution of men and women in fiction. It is not something that they claim to notice while watching television. They only reflected on it because they were stimulated to do so within the focus group interview.
Reina (46 years old, higher educated): “It wouldn’t surprise me if female characters would be portrayed in a stereotypical way, this is the language that we have all learned and that we speak. However, it would be nice to break through this custom.”
Throughout all focus groups there was a consensus that the representation of women in television fiction did not influence the participants themselves. The younger age groups considered an unconscious influence as an option, whereas among the middle aged and older participants, an influence on the participants themselves was ruled out. Among each focus group, the representation of women in television fiction was deliberated as an influence on younger groups. This could either be a negative influence:
Lydia (35 years old, lower educated): “I think it always has an influence, I think we accept a lot of gender perception that reaches us, but if you think about it, you may not want your daughter to identify with those images.”
Or a positive influence:
Thea (82 years old, lower educated): “I think it could encourage the youth. Like hey, there are possibilities!”
The participants were also asked if stimulating or enforcing a representative image of women on television would be desirable. Throughout the focus groups the opinions were quite similar. Participants seemed to struggle with this question, because they were confronted with conflicting values. In each focus group there were participants that were more and less in favour of the active stimulation of an equal representation in fiction. However, the enforcement of a representative image through quota or legislation was seen as conflicting with the artistic freedom of the makers throughout all the groups. The main thought among the groups was that the social roles that women play in society and the way women are represented on television are interrelated. Additionally, the participants indicated that an equal representation of men and women will occur automatically, no active stimulation is necessary.
Liza (25 years old, higher educated): “I think an equal representation will occur automatically when a series is successful and it contains an equal representation. Because I think people can recognize this quite well. However, I don’t think you should force it.”
Perception of the Portrayal of Women in Non-Fiction
When discussing the perception of women in non-fiction television, a distinction between different genres was made. The participants wrote down as many non-fiction titles and genres as possible, after which the moderator combined the individual input of the participants into one genre list. It was remarkable to see how the perceived gender distribution throughout the different genres was similar for nearly all the groups. Some groups came up with more genres than others. However, the overall trend was that talk shows, current affairs, hard news, art & culture, comedy, sports and crime shows were genres that were perceived to be dominated by men. Genres in which men and women were perceived to be (nearly) equally represented were entertainment, education, human interest, reality, games and quizzes and soft news. Based on this distribution several groups conclude:
Merel (25 years old, higher educated) “Men are more present in serious programs while women are more often present in programs where something has to be brought with empathy or the atmosphere has to be pleasant.”
Several reasons for this distribution of men and women in programs were suggested within and between the different groups. It seemed as if this distribution between men and women in certain genres could not be attributed to one cause. While talking about this subject the participants seemed to be hesitant in their arguments. They seemed aware of the fact their arguments were based on speculation. The two main arguments for the perceived underrepresentation of women in information oriented television genres were; on the one hand, the dominant attitude of men, both on television and in society as a whole. On the other hand, a subordinate attitude of women was also mentioned quite frequently as a reason for the underrepresentation of women in more information oriented television genres. The first argument was mentioned in each of the focus groups but it has been framed in various ways:
Reina (46 years old, higher educated) “In general, men take themselves quite serious and we still live in a society where men are very dominant in terms of image formation, they set the norm to which women are judged.”
Duygu (19 years old, lower educated) “I think that men have more guts than women. You don’t see women fighting on the street, men just have the power, the guts and the attitude.”
The dominance of men seems to be interrelated with the perceived subordinate attitude of women. Men can be dominant because women give men the opportunity:
Bien (68 years old, higher educated) “I think that women have the feeling that there is a glass ceiling, I’m not sure whether that is actually the case but they give up their place easily to someone that can tell the same story, a man in this case.”
According to the participants, women have several reasons why they do not want to make an appearance in more information oriented television programs. All the reasons could 31
be traced back to one main concept: lack of confidence. According to the groups, this is demonstrated by the fact that women are more hesitant, they are more concerned with appearance, they do not want to be put on the spot, they do not want to embarrass themselves, they do not take themselves seriously and they do not have enough courage.
Moniek (23, higher educated) I guess that women are more concerned with; if I would ask this question, how would that come across?”
The participants did not only relate the underrepresentation of women in information oriented programs to male dominance and female subordinance. According to them there are other attributes that make men and women more suitable for different kinds of programs. One of the participants stated that men are funnier, more concise and direct than women. Women on the other hand are more emotional, soft and focused on a good atmosphere than men. This is, according to the participants, the reason that guests and presenters of more information oriented television programs are more often men. However, one participant stated that it depends on the type of man or woman.
Steffi (26 years old, higher educated) “He is such a softy and he presents several children shows.”
Some of the appointed reasons for women to not want to appear on television are related to the perception of the audience. The participants stated that women are more often judged on their looks than men and that it is harder for women to be taken seriously by the audience; women have to prove themselves more. Furthermore, the participants indicate that the audience is habituated to see men in certain types of roles or functions:
Merel (25 years old, higher educated) “It’s quite a done deal that men always present talk shows, so right now it would be a risk for the makers to choose a woman as a presenter. And if apart from that, women are also anxious and nobody applies, the decision is easily made.”
Moreover, in certain programs like talk shows and current events, the external functions are first and foremost determined by the issues of the day. The participants state that since higher functions are less often fulfilled by women, the people you see in these shows are a reflection of society.
Nina (25 years old, higher educated) “No matter how you put it, the higher positions in society are more often fulfilled by men than women. And well, to choose someone in a lower position, just because she is a woman, is a bit odd.”
Older women however stress the point that women are increasingly more present in influential positions, both on television and in society in general.
Loes (84 years old, lower educated) “I think that a turning point is coming, because women are more often fulfilling higher positions and study more often than during our time, so I think it will change automatically, slowly but steady.”
Similar to the perceived influence of fiction there was consensus throughout all focus groups that the representation of women in television non-fiction did not influence the participants themselves.
Sophie (24, higher educated) “I think that back in the days when I was younger, I used to think that I had to be as pretty as the women I saw on television. I think, however, that nowadays we are more critical about what we see.”
However, the women did consider the possible influence of the image that is projected on television on the self-image of other groups, mainly young girls.
Moniek (23, higher educated) “You rarely see bigger or less pretty women on television, so they might think that you have to be like fucking Yolanthe to become a bit successful in life.”
However, here it is mentioned as well that women on television can serve as role models for younger girls.
Women rejected the thought of interference of the government in the representation of women in non-fiction programs. The main argument among the focus groups was that people should be on television because they are qualified and not because they have a certain gender. It is all about the individual. One participant did state that even though there should not be too much interference from higher institutions on television content, the principle behind the inequality in representation is not right. While another participant stated that:
Steffi (26, higher educated) “The question is if we want women to be more prominent on television, we do want to have babies at a certain point in time. (…) Maybe if we have to make a choice, a lot of women choose for their families and then that glass ceiling exists for a reason.”
Discussion
In the current study, the presence and portrayal of women on television have been studied. In order to analyze this, a mixed-method approach has been used. The first study applied a content analysis to quantify the portrayal of women on television while the second study made use of focus groups with women to retrieve more insight into their perception on the portrayal of women on television. In order to analyze this, a mixed-method approach has been used. The first study applied a content analysis to quantify the portrayal of women on television while the second study made use of focus groups to retrieve more insight into the perception of women on the portrayal of women on television.
In line with previous studies, the current study demonstrated that women overall remain underrepresented on Dutch television (Elasmar, Hasegawa & Brain, 1999; Lauzen & Dozier, 1999; Lauzen, Dozier & Horan, 2008), just a little over a third of the people on Dutch television were women. However, it is important to make a distinction between the presence of women in fiction and non-fiction. In the current Dutch television landscape, women are relatively more present in fiction than in non-fiction. If the extras on television are not taken into account and the focus is on the main and supporting roles, no significant difference 34
between the distribution of men and women has been found. This finding is in line with previous American and Dutch studies on the representation of women in television fiction that indicated that the share of female characters increased up to 40% in 2008 (Elasmar, Hasegawa & Brain, 1999; Lauzen & Dozier, 1999; Lauzen, Dozier & Horan, 2008).
The present study indicates that in Dutch television fiction, women are not equally represented within each age group. In the adult and middle adult age group, women are underrepresented. However in all the other categories, no significant difference between the distribution of men and women is identified. Related to the underrepresentation is the representation across levels of employment. When looking at the gender distribution across levels of employment women are underrepresented in the three main categories; higher, middle and lower educated profession. This is in line with previous research that demonstrated that men are more often portrayed with a job than women (Glasscock, 2001; Lauzen et al, 2008; Signorielli & Kahlenberg, 2001). However, within the largest employment category, students, no significant difference between the relative share of men and women was found. This finding is in line with the finding mentioned above on the presence of young women on television. Harwood & Anderson (2002) conducted a study on the representation of gender and age on television and found comparable results; women tend to be overrepresented in younger adulthood, but underrepresented in later middle-age. They refer to this underrepresentation as a “double jeopardy” threat (Harwood & Anderson, 2002). This indicates that this group is even more underrepresented due to the combination of its underrepresented gender and age. The observational learning theory can be applied to understand how the underrepresentation of women in certain age groups on television implies a negative effect on society. This theory states that television provides viewers with the opportunity to observe persons on television in different phases of life. By watching television viewers may develop frameworks or expectation about what it means to be a
woman of a certain age. So the lack of female role models in middle and old adulthood can cause a negative image of this social group among television viewers (Harwood, 2002).
In line with previous research, this study found that the gender distribution in Dutch non-fiction television is not equal, 34% of the people present in non-fiction were women (Gallagher, 2006; Spears and Seydegart, 2000; De Swert & Hooghe, 2010). Except for the categories children and teenagers, in which the representation was equal, women were underrepresented in all age categories. A distinction has been identified between male genres and genres in which the representation of men and women is equally distributed. Genres in which women are overrepresented were not identified. Typical male genres are news, current events, opinion programs, art & culture and sports. Genres in which the distribution of men and women is equal are religious programs, games & quizzes, entertainment shows, satire and music. In non-fiction programs, women are underrepresented in all internal functions except for assistant. Within the external functions women are underrepresented in all external function except for game contestant and performer. Within the different levels of expertise women are underrepresented in the categories expert and prominent person, while in the categories layman and speaker from experience men and women are equally distributed.
Except for the topics family/relations, gender emancipation, healthcare and multicultural society, in which women and men are equally present, women are underrepresented in each topic. The topics in which women were most strongly underrepresented were sports and politics. This tendency is in line with a study by Craft and Wanta (2004), in which they found that women in news broadcasts are often associated with feminine topics like family, childcare and health and underrepresented in news items on economy, foreign affairs and finance.
All of the above mentioned differences between the portrayal of men and women in Dutch television non-fiction seem to follow the same tendency. Women are underrepresented
in more information oriented genres, functions, levels of expertise and topics. While in the more informal genres, functions, levels of expertise and topics, men and women seem to be equally distributed.
When looking at the results of the qualitative study it is remarkable how the perceptions of women on the representation of women on Dutch television are similar to the actual representation of women on Dutch television. Women seem to have a clear notion of the quantitative distribution of men and women in both fiction and non-fiction. Throughout the different groups, women estimated a (nearly) equal distribution of men and women in fiction. Also the impression that the older women in the qualitative study expressed on the age representation was quite accurate. They emphasized that they noticed and appreciated that young women are increasingly present on television.
For non-fiction programs, women identified masculine genres and genres in which men and women are (nearly) equally represented. Feminine genres were not identified. The distribution as identified by women throughout the different focus groups is remarkably comparable to the observed distribution of gender across the different genres. In line with this finding from the content analysis, the participants indicated that men were more often present in information oriented programs, while women are more often than men present in programs where empathy and a pleasant atmosphere play an important role. Since the Dutch Public Broadcasting wants to focus particularly on an equal distribution of men and women in news and talk shows (Concessiebeleidsplan 2010-2016; NPO), this underrepresentation of women in information oriented programs does not correspond with their ambitions. Furthermore, this underrepresentation apparently does not remain unnoticed.
The participants of the focus groups opted various possible reasons for the underrepresentation of women in information oriented television programs. Causes that were mentioned were male dominance, female insecurity or sub ordinance, a more critical
approach towards women by the audience, an underrepresentation of women in higher positions in society in general and the specific character traits of men and women that make them more suitable for certain types of programs.
Overall, women do not think that the presence or portrayal of women on television has an influence on them. Furthermore, the influence that the image of women on television might have on other people is also questioned. The group that is most often mentioned as being vulnerable to influence of a certain portrayal of women is the youth. The fact that the participants believe that they are not influenced by media content while other people might be, is called the third person effect and is in itself not surprising (Perloff, 2002). However, previous research has demonstrated that the observed distribution of women and men into social roles, such as interpersonal and work roles, underlies gender stereotypes (Eagly & Steffen, 1984). It can therefore be stated that the portrayal of women on television does have an effect on society as a whole (Gerbner, Gross, Morgan, Signorielli & Shanahan, 2002). This might also attenuate the opinion of these women towards the current representation of gender on television and their opinion on possible policy towards gender equality on television.
The most remarkable outcome of this study might be that women seem to be quite indifferent towards the actual and perceived underrepresentation of women on Dutch television; the fact that an underrepresentation of women on television on general is not perceived as something negative. Women rejected the thought of interference of the government in the representation of women in fiction and non-fiction programs. For fiction, the main argument among the participants was that interference would be in conflict with the artistic freedom of the makers. For non-fiction, the main argument was that people should be on television because they are qualified and not because they have a certain gender.
When the results of the quantitative and the qualitative study are combined, four main trends can be identified. First of all, younger women seem to be more visible in fiction and non-fiction than adults and older women. This result was found in the content analysis and confirmed by women in the focus groups. It is not certain if the presence of younger women on television can be understood as a trend that will carry on towards the presence of older age groups when these younger women get older. However, it is worth mentioning that the emancipation of younger women in The Netherlands is also increasing on other levels. In 2012, a woman’s gross hourly salary was nearly 18 percent lower than for men. However, women between 25 to 30 years earned slightly more per hour than men of the same age (Central Bureau voor de Statistiek, 2012). Future research could explore whether a correlation between financial emancipation of women and their presence on television exists.
Second, even though this finding was not confirmed in the focus groups, the content analysis found that in fiction, women are underrepresented in all levels of employment. Only female students are equally represented compared to their male counterparts. Research on the current emancipation climate implied that although the situation on the labour market has worsened in recent years due to the crisis, women managed to maintain their position in labour participation. From 2008 to 2009, the net employment of women between 20 to 64 years, increased slightly to 64%. This percentage did not change in 2010 and 2011 (Centraal Bureau voor de Statistiek, 2012). If this trend will be reflected in the representation on television fiction, it would imply an increase of employed women on television.
Third, in non-fiction both in the content-analysis and in the focus groups, certain gender specific tendencies were identified. Women seem to be underrepresented in more information oriented genres, functions, levels of expertise and topics. While in the more informal genres, functions, levels of expertise and topics, men and women seem to be equally distributed. The participants of the focus groups opted various possible reasons. Causes that
were mentioned were male dominance, female insecurity or subordinance, a more critical approach towards women by the audience, an underrepresentation of women in higher positions in society in general and the specific character traits of men and women that make them more suitable for certain types of programs.
The fourth and maybe most remarkable trend in this study might be that women seem to be quite indifferent towards the actual and perceived underrepresentation of women on Dutch television. An underrepresentation of women on television in general is not perceived as something negative. Two possible explanations for this can be deducted from the focus groups. On the one hand, older women claim that a lot has already improved in the field of women emancipation and the portrayal of women on television over the years. On the other hand, younger women claim that the emancipation of women that has been accomplished by previous generations could feel like a burden. This perceived burden might cause a shift from more independent women towards more family and care oriented women. However, this shift would not be enforced by external social pressure but it would be initiated by women themselves. This is in line with the post feministic trend in which new femininity with an emphasis on individual choice is central (McRobbie, 2004). These personal choices of women may inevitably also be traditional and conservative. This however should not be labelled as negative because this choice was not imposed by the media or patriarchy. Further research is necessary to explore whether this theory does indeed correlate with the indifferent attitude of young women towards underrepresentation on television.
Within the current study, certain shortcomings can be identified. First of all, conclusions about the distribution of men and women have be drawn very carefully. Since the total sample consists of more men than women, the relative distribution shifts easily into the direction of more men per sub category. It is however possible to compare the different categories with each other and base conclusions on this distribution. Furthermore, focus
