Sex Roles (2010) 62:798–809 DOI 10.1007/s11199-009-9722-5 ORIGINAL ARTICLE Crime Scene Investigation: The Chief Inspectors’ Display Rules Juliane Finger & Dagmar C. Unz & Frank Schwab Published online: 29 December 2009 # Springer Science+Business Media, LLC 2009 Abstract Television as one factor in the “cycle of role division” (Eagly 1987; Shields 2002) can aid to perpetuate traditional role attitudes as well as set new norms. The aim of the study is to examine gender differences in facial expression of emotion on TV as well as changes of these differences over time. A content analysis of 12 episodes of a prominent German crime series from two time samples (1979–1981 / 2005) was done by employing the Emotional Facial Action Coding System (EmFACS, Friesen and Ekman 1983). Results show that there are gender differences in stereotypically male emotions: Male actors express these emotions significantly more frequently than female actors, while the latter show more “social smiles”. These differences are reduced over time. Keywords Emotions . Gender . Television content Introduction “Television is the source of the most broadly shared images and messages in history. It is the mainstream of the common symbolic environment into which our children are born and in which we all live our lives” (Gerbner et al. 1994, p. 17). Gerbner and colleagues attribute this important position to television due to some of its specific features: Television penetrates nearly all social and age J. Finger Department of Psychology, University of Hamburg, Von-Melle-Park 6, 20146 Hamburg, Germany D. C. Unz (*) : F. Schwab Media Psychology, Saarland University, PB 15 11 50, D-66041 Saarbruecken, Germany e-mail: d.unz@mx.uni-saarland.de groups, it is available nearly everywhere. The compliance with requirements to use it is almost non-existent (no intricate literacy, low cost, no mobility needed) and it is mostly used unselectively. Furthermore, a lot of programs that aim for a large audience are very similar in their main features. These programs show a “general adherence to common notions of justice and fair play, clear-cut characterizations, tested plot lines, and proven formulas for resolving all issues” (Gerbner and Gross 1976, p. 182). Simultaneously their content communicates a certain closeness to reality. Cultivation theory assumes that these characteristics lead to “massive, long-term, and common exposure of large and heterogeneous publics to centrally produced, mass distributed, and repetitive systems of stories” (Gerbner et al. 1994, p. 20), and as a consequence, these invariant cultural patterns in television content cultivate real-world beliefs. The purpose of this study is to examine the portrayal of men and women on German TV, specifically, gender differences in the facial expression of emotion of male and female actors. On the basis of arguments of the social role theory (Eagly 1987), which posits that gender differences are largely due to different social roles occupied by men and women, it is examined whether gender differences in emotion expression which are staged in TV series persist even if men and women act in the same work role. Furthermore, it is examined whether the differences in the portrayal of the genders on German TV have diminished over the last 30 years, taking into account the societal change towards a more egalitarian division of labour between the genders in Germany (e.g., United Nations Economic Commission for Europe [UNECE] Statistical Database 2009). To address these questions, a content analysis of a German media sample from two periods (1979–81 and 2005), employing the EmFACS (Friesen and Ekman 1983), was designed. The EmFACS measures emotion expression as shown via facial behavior. Sex Roles (2010) 62:798–809 Media images and portrayals have a mentionable effect on the thoughts and attitudes of the spectators (Gerbner et al. 1994). A meta-analysis of Shanahan and Morgan (1999) on 97 samples with 5,799 separate findings overall reports a small but consistent positive relationship of .10 between television viewing and beliefs about the world. This effect is not moderated by the content of the dependent variable, that is, it does not matter if the study is about gender roles, violence, policy or other themes; even demographic variables like age, education, sex of the participants do not show any significant effect. Although the assumptions of cultivation theory (e.g., Gerbner et al. 1994; Shanahan and Morgan 1999) minimize genre differences (e.g., Gerbner et al. 1978), there are findings that different genres may influence perceptions differently (e.g., Grabe and Drew 2007; Potter and Chang 1990). As an example, a study by Segrin and Nabi (2002) found evidence for the influence of viewing romantic programs, not TV viewing in general, on unrealistically idealistic expectations about marriage. Plus, according to the drench hypothesis of Greenberg (1988) noteworthy or striking media examples can be more influential than the sheer frequency of TV characters or behaviours that are viewed. This means that exposure to specific genres or programs may have a stronger influence on people’s attitudes and beliefs than overall TV viewing (cf. Covert and Dixon 2008; Harrison 2003). Thus, we argue that favorite television programs can reinforce gender stereotypes. According to theories and research in social cognition, stereotypes are defined as the beliefs about the attributes and characteristics of a group of people (e.g., Hilton and von Hippel 1996). Stereotypes represent categorical knowledge associated with social groups and may include important assumptions about how group attributes are related to one another; and therefore reflect the perceiver’s inferences and causal attributions based on the social categorization (Wittenbrink et al. 1998). Accordingly, gender stereotypes are the beliefs about the attributes and characteristics of men and women established by their gender. On the basis of different gender roles, gender stereotypes include differences between men and women in the expression of emotions (e.g., Alexander and Wood 2000). Gender stereotypes are quite stable across different cultures, “with cultural factors producing minor variations around general themes” (Best and Williams 2001, p.198). When choosing favorite television characters, preschoolers, adolescents and adults show same-sex preferences (Chory-Assad and Cicchirillo 2005; Hoffner and Buchanan 2005; Hoffner 1996; Wilson and Drogos 2007). According to social cognitive learning theory, individuals can learn from observing others, including characters in the media (Bandura 1986). Models that are attractive or that are similar to the viewer herself/himself are likely to be 799 imitated. Therefore, media characters that are attractive and alike to the self should be potent role models (cf. Wilson and Drogos 2007). This means that television can be one factor that interacts in the “cycle of role division” (Eagly 1987; Shields 2002). Images shown on TV can aid to perpetuate traditional, conservative (gender) role attitudes as well as set new norms for appropriate behavior (e.g., Calvert and Huston 1987; Rivadeneyra and Lebo 2008), particularly, favorite programs may have a strong impact (Calvert et al. 2003). Regarding the significant role of TV in socialization, which means, that television has cultivation effects (e.g., Nabi and Riddle 2008; Shanahan and Morgan 1999), gender-related images and content of TV programs have to be explored. If we are interested in the socialization of gender and in the development of gender-related stereotypes, it has to be asked if and how the staging of the genders in popular television programs is changing over time. Gender Biased Media Content Few studies have dealt so far with the portrayal of men and women on German TV. Research has been concentrated on the picture on American TV. It has been shown that the portrayal of men and women in the media does parallel the above mentioned stereotypes when it comes to the roles occupied as well as behaviors (for prime time television in North America e.g., Ganahl et al. 2003; Gerbner 1997, 1998; Glascock 2001; for computer games, e.g., Dietz 1998, North America; Smith 2007, review mainly North America; von Salisch et al. 2007, Germany). It has to be noted that, when comparing media content from the 70ies to our present days, the differences in the way women and men have been portrayed are reduced. For example, women in a North American media sample from the late 90ies are also shown working outside the house, in more prestigious work roles than in the 70ies (Signorielli and Bacue 1999). However, even though differences in quality and quantity of women’s and men’s roles and representation on TV have been diminished over the past 20 years, differences and inequalities persist until now (Bartsch et al. 2000; Coltrane and Messineo 2000; Thompson and Zerbinos 1995). All of these research papers dealt with changes of media content in North America. As mentioned above, very few researchers have examined the picture on TV in other parts of the world, such as Western Europe (Europaeische Kommission/ Referat Chancengleichheit fuer Frauen und Maenner 1999). Cross-cultural research indicates that despite the crosscultural similarity of gender stereotypes (Best and Williams 2001) there is some influence of culture models on gender differences in the expression of emotion, e.g. on the experience and expression of anger (Fischer and Manstead 2000; Fischer et al. 2004) or in smiling (LaFrance et al. 800 2003, meta-analysis). Similarly, a 14-country study conducted by Williams and Best (1990b) found intercultural differences in beliefs concerning appropriate relationships between the genders (modern/egalitarian vs. traditional/ male-dominant). The most modern beliefs were found in European countries (including Germany, the Netherlands, Italy, Great-Britain and Finland) while the United States were only in the middle of the distribution. The most traditional beliefs were found in Asian and African countries. These differences can be attributed to differences in women’s status and roles in the respective countries (Fischer et al. 2004). Thus, it seems necessary to look at the portrayal of gender and display roles in different countries. By examining the picture on German TV, this study intends to fill the existing gap when it comes to the analysis of gendered television content (cf. Europaeische Kommission/Referat Chancengleichheit fuer Frauen und Maenner 1999). Gender Differences in Emotions With the start of the women’s movement in the seventies, gender roles and role expectations have been the topic of numerous studies and books, in popular press as well as in scientific publications (e.g. Brody 1999; Mealey 2000; Pease and Pease 2001). Since emotions are essential parts of human life (e.g., Cosmides and Tooby 2000), especially, differences in emotionality are a popular topic for discussion. The following section introduces the social role theory (Eagly 1987) and its explanation for gender differences in emotion. After that, previous research on gender differences in emotions is presented. The social role theory (Eagly 1987) explains gender differences as evolving from the ongoing division of roles between men and women in society. To describe the roles, Eagly (1987) uses the terms “agency” (men) and “communion” (women) first used by Bakan (1966). The traditional female role of the caregiver requires interpersonal skills and an ability to care well for others, which is reflected in the term communion. Stereotypes ascribed to women on the basis of their gender role describe them as being caring, affectionate, gentle, cheerful, having a concern for others and striving for harmony (cf. Bakan 1966; Brovermann et al. 1972; Eagly 1987; Williams and Best 1982). The male role of the provider, which Bakan (1966) describes with the term agency requires skills that enable to be successful in a supposedly competitive job environment; it is characterized by attributes of competence, instrumentality and assertiveness. Further traits that are seen as stereotypically male include men being dominant, unfriendly, aggressive and task-oriented (cf. Brovermann et al. 1972; Eagly 1987; Williams and Best 1982). These attributes are also reflected in women’s and men’s self-descriptions (e.g., Feingold Sex Roles (2010) 62:798–809 1994) and their behavior (e.g., Hundhammer 2008). On the basis of the different gender roles, there are differing rules, scripts and norms that apply to men and women (Eagly 1987). The social roles also have an impact on the expression of emotions. The exertion of different roles does require different skills as well as the display of different emotions (e.g., Alexander and Wood 2000; Hochschild 2003). These norms for showing an emotion are also called display rules (Ekman and Friesen 1975). Display rules are the culturally shared norms and expectations of how, when and where to show emotions. Their content is mostly analogous to stereotypical beliefs about emotional expressivity (Brody 1999). A lot of research has been generated concerning the topic of gender roles and emotions (e.g., Aldrich and Tenenbaum 2006, on gendered patterns in emotion talk; Brody 1985, a literature review; Fischer 2000). The results of previous studies show that for Western countries (like Western Europe and North America) emotional expressivity of men and women is in line with the role-based expectations or stereotypical beliefs. As expected in social role theory, women show more emotions that are linked to caretaking activities, that is positive emotions to establish and maintain relationships with others, and that reflect their lower social status. Men, on the other hand, show more emotions that are linked to their agentic role, which enable them to compete in an aggressive and achievement-oriented job environment, e.g. in management positions (cf. Alexander and Wood 2000; Eagly 1987). In the following, evidence from Western countries is presented, and the countries in which the respective studies were conducted are explicitly mentioned. There is evidence that women report being more emotionally expressive (Bronstein et al. 1996, North America) than men do. Women smile significantly more than men, especially in public settings (e.g., Hall 1984, meta-analysis; Hall et al. 2000; LaFrance and Hecht 2000, meta-analysis; Lee and Wagner 2002, UK). Women express the emotions fear, sadness and shame more often (e.g. Grossman and Wood 1993, North America). These emotions are categorized as “powerless”, since they are related to submission and low self-esteem. They also express prosocial emotions like joy more often (Fischer and Manstead 2000, Western cultures). Men’s emotion expression reflects their role as the provider, adopting agentic behaviour (Bakan 1966). For men, studies have shown that they show assertive behavior more often than women, and conceal emotions such as shame or sadness (Brody 1999; Jansz 2000, Western cultures). They exceed women in expressing emotions of anger, contempt and pride (Averill 1983, North America; Fischer and Manstead 2000; Tomada and Schneider 1997, Italy), categorized as powerful, that is having higher social status, and related to aggression and high self-esteem. Sex Roles (2010) 62:798–809 Above that, there is some dissent regarding the expression of different smiling types in men and women. Emotion research makes a difference between two types of smiling (e.g., Hess et al. 2003; Thibault et al. 2009): The “Duchenne smile” (Duchenne 1862/1990) involves the movement of both the zygomaticus major muscle (raising the lip corners) and the orbicularis oculi muscle near the eyes producing crow’s feet-like crinkles(. A “social smile” then again is produced without the orbicularis oculi activation. The Duchenne smile is mostly interpreted from a vis-a-vis as an expression of truly experienced joy (Frank et al. 1993; in the FACS terminology of Ekman and Friesen the Duchenne smile is called “felt happiness”) and requires the recruitment of deliberately hardly controllable muscles in the upper face. The social smile (in the FACS terminology “unfelt happiness”) is a consciously controllable facial action that is shown primarily for communicative reasons in social interaction, for example as an indication of politeness or shyness or to display positive affect in public situations (Ekman and Friesen 1975, 1982; Frank and Ekman 1997; Hess et al. 1995; Hess et al. 2003). Hecht and LaFrance (1998) found gender differences for social smiles, but not for Duchenne smiles. Societal Changes in Germany from the Late 70ies Until Today (2005) Eagly (1987) argues that gender roles and the corresponding expectations can not be changed if the distribution of work is still gender-segregated and women don’t have access to traditional male domains. In Germany, several societal changes have occurred from the late 70ies until 2005 (the year from which the media sample for period 2 was taken). The male role of the provider has been eroding, as the unemployment rate, especially for less educated men, has been on the rise (Bundesministerium für Familie, Senioren, Frauen und Jugend 2005). Women’s level of education has been continually rising, which has helped them gain more job opportunities: 45.3% of the workforce in 2005 are women, as compared to 39.7% in 1980 (UNECE Statistical Database 2009). The number of women in decision making positions has risen, e.g., the number of female university professors in Germany has doubled from 6% in 1980 to 12.8% in 2004 (CEWS Statistik 2009). However, the so called glass ceiling (e.g., Osterloh and Littmann-Wernli 2000) still exists: women are still by far less represented in decision making positions than men (21% women, 79% men in 2004) and paid less than men: Women earned 22% less than men in 2002, as compared to 28% less than men in 1977 (Bundesministerium fuer Familie, Senioren, Frauen und Jugend 2004; cf. Müller 2000). 801 Research Questions and Hypotheses The objective of the study is to identify whether (1) the portrayal of emotions of male and female actors on German TV differs even if they occupy the same professional role, and whether (2) there has been a change in the portrayal of male and female actors from the late 70ies until now. The portrayal of emotions is operationalized by analyzing the facial expression of emotions. Facial expression is the most frequently mentioned form of emotion expression (Scherer and Wallbott 1994). Further reasons for focusing on the study of facial expressions of emotions include: First, gender differences are largest when looking at the nonverbal behavior (Hall 1984). Second, since facial expression is, in contrast to for example verbal statements, less prone to intentional “fraud”, or falsification (Ekman and Friesen 1975), facial expressions of movie actors may be credible expressions. Third, in movies facial expressions are particularly suitable to convey emotions of the actors (Wallbott 1988). Work roles vs. gender roles: The social role theory (Eagly 1987) states that everyday roles (e.g. at the workplace) can overlay gender roles; hence diminish stereotypical behavior of men or women. However, evidence in that respect is mixed — recent studies have found gender differences in some leadership behaviors even though men and women were doing the same job or occupying the same role (Eagly et al. 2003; Hess et al. 2005; Van Engen and Willemsen 2004). Regarding these results and taking into account the saliency of a persons’ sex and gender role, it is predicted that professional roles do not totally overlay gender roles. It is furthermore predicted that on TV, gender differences in emotional expression can be found even though women and men play the same role. Hypotheses for both negative and positive emotions are made on the basis of the gender-emotion-stereotypes mentioned above. As outlined above, in most Western countries, including Germany, the male stereotype is closely associated with the emotions anger/aggression and contempt. Women, in contrast, are believed to send more positive social signals which include showing more joy by means of smiling (cf. Alfermann 1996; Brovermann et al. 1972; Williams and Best 1982, 1990a). Hypothesis 1a: Male actors are expected to show significantly more negative, stereotypically male emotions (anger, contempt) than female actors. Hypothesis 1b: Female actors, on the contrary, are expected to send more positive social signals, that is, to show more smiles than male actors. Since emotion research identified two types of smiles that have a different social significance, a separate analysis for Duchenne smiles and social smiles is conducted. Social 802 Sex Roles (2010) 62:798–809 smiles are shown primarily for communicative reasons in social interaction, thus are more socially constructed. Duchenne smiles are mostly interpreted as an expression of truly experienced joy. Therefore, it may be argued that gender differences are greater for social smiles than for Duchenne smiles (LaFrance 2003). Hypothesis 1c: Female actors are expected to show more social smiles (unfelt happiness in the FACS terminology) than male actors. Changing gender roles with time: According to Eagly (1987) the cycle of role division can only change if women gain more access to traditionally male domains, especially at the workplace. She furthermore postulates a strong role of the media in the perpetuation of traditional roles. Thus, the (second) goal of this study is to examine possible changes in the portrayal of men and women on TV over time. As a matter of fact, since the late 70ies, in Western societies such as Germany, an increasing number of women have gained access to traditionally male domains, in real life (cf. Bundesministerium für Familie, Senioren, Frauen und Jugend 2004; Eagly 2003; UNECE Statistical Database 2009) as well as on TV’s fictional narratives (e.g., Signorielli and Bacue 1999; Süss 1993). Accordingly, Diekman and Eagly (2000) found evidence for a change in attitudes concerning stereotypical traits and behaviors, comparing stereotypical images of men and women of the past, present and future (cf. Diekman et al. 2004). But on the other hand, a number of scientists see stereotypes as relatively stable across time (Ashmore and Del Boca 1979; Williams and Best 1990a). In this contribution it is expected that “modern” TV takes into account a changing society. Thus it is predicted that equality in the expression of emotions by TV characters has been increasing over time (from the 70ies until today). Hypothesis 2: It is expected that gender differences are diminished in period two (2005) as compared to period one (1979–81), this refers to gender differences in the display of negative, stereotypically male emotions (Hypothesis 2a) as well as gender differences in the display of the positive, stereotypically female emotion joy, as shown via smiling (Hypothesis 2b) resp. social smiling (Hypothesis 2c). was made for several reasons: First, women and men occupy the same role as a chief inspector. That made it possible to examine sex differences in emotional expression without the influence of other social roles. Second, the analysis of just one genre was preferred, thereby diminishing the chance that differences in emotion expression are partly or solely due to particularities of a certain genre. Third, “Tatort”, as a very popular TV series, reaches a large audience, the mean market share is above 20%, and is distributed not only in Germany, but in two other German speaking countries, Austria and Switzerland. From the first broadcast of “Tatort” in 1970, over 700 episodes have been shown on TV. Thus, “Tatort” can be taken as a critical TV program that may have a strong effect on impression-formation and image-building according to the drench hypothesis. Since the goal was to examine gender as well as time differences, “Tatort”-episodes from two periods of time were selected. The starting point for the analysis was the year of the first appearance of a female chief in “Tatort”, this was in the year 1978. The next woman played the leading part as a chief in 1981. The options for the choice were unfortunately limited to two female chiefs in the late 70ies. An analysis of two episodes with the same actress was accepted because it was the only possibility to enlarge sample size. A random selection was drawn from all the episodes with the two female chiefs in the 3 years 1978, 1979, 1981. Accordingly, a random sample was drawn from all “Tatort”-episodes with male chiefs for each corresponding year — 1978, 1979 and 1981. For period one, the overall number of episodes from which samples were drawn was n=37 (for each year, number of episodes broadcasted was for 1978 n=12, 1979 n=13, 1981 n=12). The sample for period two was drawn from the year 2005. At the time the study was carried out, this was the last “complete” year of “Tatort”-broadcasts. For 2005, just like for the “old” “Tatort”-series, a random selection was made of three episodes with male, three with female chiefs. The total number of episodes broadcasted from which the sample for period two was drawn was n=34. (The slight difference in sample size for the two periods can be attributed to changes in the frequency of broadcasts of “Tatort” episodes per year, that is, due to its popularity, an increased number of “Tatort” episodes were produced over time.) On the whole, n=12 “Tatort”-episodes were examined, six from the late 70ies/early 80ies (n=3 male chiefs, n=3 female chiefs), the other six from the year 2005 (n=3 male chiefs, n=3 female chiefs). Method Procedures Sample Analysis of the facial expressions of emotions was done by the Emotion Facial Action Coding System (EmFACS, Friesen and Ekman 1983). EmFACS is a special application of the Facial Action Coding System (FACS, Ekman and Friesen 1978): Only those facial movements are scored For the investigation of gender roles as portrayed on German TV, the oldest and most popular German crime series (GfK, 2004), “Tatort”, was examined. That choice Sex Roles (2010) 62:798–809 which are relevant for emotion detection. The advantage of this coding method is its objectivity, hence an avoidance of biases in the analysis. The coding of so-called action units as visible appearances of facial muscle activities and interpretation of expressions (relying on an emotional dictionary) are separated processes. The overall reliability of EmFACS has been repeatedly demonstrated (Rosenberg and Ekman 1998). All Coders involved in the study passed the final examination administered by Ekman’s laboratory, establishing their personal reliability (>.80). The mean intercoder-reliability is above .80. The coding was restricted to scenes in which facial expressions were sufficiently visible, because only those scenes can transport the emotions of the actors. This means: Only shots, in which the face of the actor was sufficiently visible, were coded, that is at least American shot, medium close up or close up shot (see also Wallbott 1988). Scenes, in which a shadow or poor light obscured more than 50% of the face or in which accessories such as hat, sunglasses etc. obscured important parts of the face, and scenes, in which the actor turned the face away from the camera in a way that less than 55% of the face could be seen, were excluded from analysis. For each movie, the same amount of frames should be coded. The smallest amount of codable material in an episode was 6000 frames (= 4 min), so 6000 frames were set as a limit for the coding amount of all other episodes. For all episodes except one, the amount of 6000 frames was accomplished after two thirds of the episode (app. 60 from 90 min of overall duration). The analysis comprised four steps: First, all the episodes were digitized and converted to the same format with the same compression, making sure all were on the same level of quality. Second, the episodes were viewed and coded simultaneously. The coding units were the so-called apexes of action units as visible appearances of facial muscle activities, generating a time series of event points. In EmFACS the coder uses FACS-codes, but describes only those facial events that include action units or combinations of action units which are relevant for certain emotion expressions (e.g., Rosenberg 2005). The coded action units respectively the coded combination of action units (AUs) occurring at the same time were the units of observation (or units of recording; collected as momentary events). Parallel to this event-based EmFACS coding, the beginning and end frame of each coded scene (coding onset and offset times) were captured in a separate document, as well as the scene type (e.g., “conversation among men”, “finding a dead body”, etc.). In the third and last step, the coded data (documented as action unit combinations along the time line) were “translated” into emotions using the “Emotional Dictionary” (Wagner 1986) as a computerized algorithm. The Emotional Dictionary uses the originally coded EmFACS data and searches for AUs or combinations of 803 AUs which can be related to an emotion. Only those AUcombinations that are reliably identified as an emotional expression in different cultures are assigned to an interpretation (Wagner 1986). Negative, stereotypically male emotions were operationalized as those EmFACS events that could reliably be identified by the Emotional Dictionary as either anger or contempt. Positive, stereotypically female emotions were operationalized as those EmFACS events that could be reliably identified by the Emotional Dictionary as happiness (both felt und unfelt happiness). Emotions other than anger, contempt and happiness identifiable by the Emotional Dictionary are disgust, surprise, fear and sadness. As these were not part of our predictions they were not included in the statistical analysis. Statistical Procedures The unit of analysis were n=12 “Tatort”-episodes, six from the late 70ies/early 80ies (n=3 male chiefs, n=3 female chiefs), six from the year 2005 (n=3 male chiefs, n=3 female chiefs). The Randomization Test was used as a statistical test for small sample sizes. It is an exact test that has — in contrast to commonly used tests for small samples such as the Mann-Whitney-U-Test or the Kruskal-Wallis-H-Test — the advantage not to be based on the rank-order method. Thus, it is not vulnerable to information loss and more sensitive for effects. Results In all the twelve “Tatort”-episode samples, 533 EmFACS events (coded action units respectively the coded combination of actions units occurring at the same time) were coded, with a mean of 44.42 events per episode, or 11.1 events/minute in sequences fitting the mentioned selection criteria. On average, 27.42 of these 44.42 events per episode were interpreted, 17 were not interpretable. Only those events that were unambiguously interpreted by the Emotional Dictionary (Wagner 1986) were included in the statistical analysis. On average, of the 27.42 interpreted events, 2.8 were interpreted ambiguously and thus excluded from analysis. In the end, a mean of 24.58 events per episode sample could be statistically analyzed, on the whole 285 events over all episodes. Of these, 85 events were interpreted as emotions other than anger/contempt and happiness: disgust, surprise, fear and sadness. As mentioned above, these were not part of our predictions. Except for surprise, these emotions were hardly ever shown (less than 20 times per specific emotion in all episodes). The remaining 210 events were statistically analyzed. See Table 1 for detailed information about means and standard deviations for the respective analyzed emotion. 804 Sex Roles (2010) 62:798–809 Emotions Mean SD Time Gender Anger Contempt General smiling Felt happiness Unfelt happiness 1.50 3.08 12.92 4.50 8.42 1.68 3.45 6.60 3.15 5.30 .04* .01** .86 .66 .54 .97 .51 .67 Means and standard deviations for the unambiguously interpreted emotion expressions for which hypothesis were made, as well as the results of the statistical testing for differences between time period and gender, respectively. * p<.05, **p<.01. Negative Emotions Regarding negative emotions, we predicted that men would show anger and contempt more often than women, even though they are acting in the same role (H1a) and that these differences would be reduced over time (H2a). On the whole, male actors showed the stereotypically male, negatively connoted emotions of anger and contempt 17 times during the data collection of period one (1979–1981) and 21 times during period two (2005), whereas females showed these emotions three times during period one and 14 times during period two. In Hypothesis 1a, we predicted that men would show significantly more negative emotions stereotypically associated with the male role (anger and contempt), this was tested with a Randomization test. As predicted, a main effect of gender could be found (p=.01, one-sided). Above that, a main effect of time could be found (p=.04, one-sided), with both men and women showing more negative, stereotypically male emotions in the new episodes at period two. In Hypothesis 2a, it was predicted that gender differences in the display of negative, stereotypically male emotions were diminished during period two (2005) compared to period one (1979–81). Since the sample did not fulfill the requirements to statistically test the interaction effects of period and gender (H2a), the results can only be reported descriptively. Figure 1 shows the interaction of the factors period and gender. For both men and women, negative, stereotypically male emotions have increased from period one to period two; the increase for women has been sharper than for men. During period two compared to period one, the discrepancy in the frequency of these specific negative emotions has decreased substantially. Positive Emotions For the positive emotion joy, we predicted that women would show it more often than men, even though they act the same role (for general smiling, Hypothesis 1b and for social smiles, Hypothesis 1c) as well as a reduction of these gender differences over time (Hypothesis 2b and c). General Smiling was shown more often than every other emotion, on the whole 155 times. For male actors, this meant a total of 56 displays of smiles (28 during period one; 28 during period two). For female actors 99 displays of smiles were noted (61 during period one; 38 during period two). The prediction of H1b, women showing more positive social signals, that is showing (general) smiles more often, was tested with a Randomization test. H1b could not be confirmed: gender was not significant as a factor (p=.97). Above that, there were no significant differences in the display of (general) smiles between the two periods (p=.86). In detail, the Duchenne smile (felt happiness) was shown 24 times by male actors (15 during period one; 9 during period two) and 30 times by female actors (13 during period one; 17 during period two). For Duchenne smiles, the results of a Randomization test did not show any effects: Gender as a main effect was not significant (p=.51). Neither was period as a main effect (p=.66). Hypotheses 1c predicted gender differences in social smiles. Male actors showed social smiles (unfelt happiness) 32 times (13 at period one; 19 at period two), female actors 69 times (48 during period one; 21 during period two). Again, the randomization test did not confirm our hypothesis (H1c), gender as a main effect was not significant (p=.67), neither period as a main effect (p=.54). To further explore the data, the distributions were tested for uniformity with a Chi Square test. For Duchenne smiles, neither the distributions among, nor between gender were significantly different. Note that not all actors showed this emotion, and thus a few cases were not included in the analysis. In the case of social smiles, there was a significant main effect of period in the group of women (χ² (5, n=6)= 14.04; p=.02). Thus, female actors in the “old” episodes (1978–81) showed significantly more social smiles than in “modern” episodes (2005). This was not observed in the group of male actors (χ² (5, n=6)=6.63; p=.25). Additionally, during period one, for social smiles, there was a highly total number of stereotypically male negative emotions Table 1 Means and standard deviations for unambiguously interpreted emotions in all episodes. 25 20 15 10 5 0 1979-1981 period male actors 2005 female actors Fig. 1 The expression of stereotypically male negative emotions (anger and contempt) by male and female actors at different time periods. Sex Roles (2010) 62:798–809 805 significant deviation from equipartition between male and female actors (χ² (5, n=6)=21.92; p=.001). During period two, no significant deviation from equipartition could be found. These results can be seen as a hint to existing differences in the emotional expression of male and female actors when it comes to communicate social smiles. These differences tend to diminish over time, due to women showing less social smiles. Again, the interaction of the factors period and gender (H2c: reduction of gender differences in positive, stereotypically female social smiles in period two compared to period one) can only be reported descriptively. Figure 2 illustrates the findings of the statistical testing: Female actors exceeded men in displaying social smiles, a distinctness that is reduced during period two. Discussion total number of social smiles In sum, the results confirm the first hypothesis — gender differences in specific emotions — in the case of negative, stereotypically male emotions. The male chief inspectors in the crime series “Tatort” express these emotions significantly more frequently than female chief inspectors. For Duchenne smiles, no significant difference could be found, whereas for social smiling, a difference between male and female actors could be found. The results for hypothesis 2, the reduction of gender differences over time, could solely be reported descriptively. For the examined negative emotions, as well as for the social smiling, the reported gender differences seem to diminish over time. In Western societies, the emotions anger and contempt are more closely associated with men than with women (e.g., Williams and Best 1982). The same picture seems to apply for their portrayal in crime series on German TV. Male “Tatort” chief inspectors were found to facially express negative, stereotypically male emotions significantly more often than female chief inspectors. The differences 60 50 40 30 20 10 0 1979-1981 2005 period male actors female actors Fig. 2 The expression of social smiles by male and female actors at different time periods. tend to lessen over time. These results confirm observations of previous studies conducted in other Western countries: Gender roles seem to be stronger and more relevant for social behavior than other social roles, e.g. roles at the workplace (see Eagly et al. 2003; Hochschild 2003). Eagly’s (1987) assumption — specific roles can overlay the more general gender roles — does not seem to apply here. Interestingly, there was an increase in the quantity of expressions of stereotypically male negative emotions for men and women over time. This might be due to a change in movie plots and the extremity of (crime) scenes shown on TV from the 70ies until today, or to a general change in societies’ acceptance of these emotions, or to both of it. One can imagine that more exciting scenes such as the sight of a mutilated corpse can arouse more anger and contempt in staged reaction shots of the chiefs. Alternatively, a shift in the general acceptance of these negative emotions could have occurred from the 70ies until now. It is imaginable that the open display of these emotions is more accepted nowadays — for both men and women — than it was 25 years ago. Smiling was shown with the highest frequency of all emotions by all actors, but even more so by the female ones. Though the distribution of the data by trend confirms hypothesis 1 and 2, neither the effect of period nor of gender was statistically significant. A closer look at the data, that is a partition of the general “smile” in the two kinds “Duchenne smile” and “social smiling” gave more insight. In contrast to smiling in general, the variability among the groups of male and female actors is smaller when looking at the two kinds of smiling. Gender differences were largest for social smiles, especially at period one. On the basis of general knowledge about smiling and the postulates of social role theory (Eagly 1987) this picture can easily be understood. Different types of smiles have a different social significance: Whereas a Duchenne smile is mostly shown when truly feeling happy, social smiles are used primarily for communicative reasons in social interaction (e.g. Hess et al. 2003) and furthermore social smiling is one of the most common methods of masking other emotions (Ekman and Friesen 1975). Thus the display of social smiling by the female actors in crime series on German TV can be seen as conform to their gender role. The reduction in the amount of social smiles in episodes of 2005 can be interpreted with a change in role expectations. A few methodological problems might have influenced the results. First, the sample size was rather small, with 12 episodes analyzed. This was largely a result of the choice of the TV series “Tatort”. Recall that this series was chosen because men and women act the same role in the same movie genre. Unfortunately, from the first broadcasting of “Tatort” until the end of the 80ies just two women were acting as chiefs. To be able to enlarge the sample size, two episodes with the first chief were analyzed. It is possible 806 that this necessary choice produced distortions in the results. But on the other hand, there are only 5% women in leading positions in the German police nowadays (GdP 2002), and only one woman is reported that directed a homicide division (Koelbl 2004), whereas in “Tatort” there were five female chief inspectors in 15 different teams in 2005. Second, in addition to restrictions with regard to sample size, the amount of data collected from the “Tatort”episodes was small. Maybe the plot of “Tatort”-episodes provoked few strong emotions. A more likely reason, however, seems to be the (necessary) scene selection: The coding of facial expressions was often accomplished in conversational scenes. In general, conversations were held in a rather calm fashion by the chief inspectors. It seems as if, in that case, the specific role as a chief could overlay the more general gender role to some extent. Possibly, the role as a chief required the inhibition of certain emotions, such as strong anger, in order to appear trustworthy to a potential witness or suspect or to show a “professional coolness”. Additionally, the “Tatort”-series, produced by the German public, non-commercial TV station ARD is intended to show events close to everyday life, be authentic and not full of pathos or flashy effects (Wacker 2000). For this reason, it may be avoided to emote or dramatize. Third, further restraints resulted from the algorithm used to interpret the data, the Emotional Dictionary (Wagner 1986). Only about half of the coded facial events could be interpreted (men: 46%; women: 56%). This is largely a result of the rather conservative interpretation of the data by the Emotional Dictionary that can also be found in the analysis of everyday facial behavior (Merten 2001). Only AU combinations are interpreted that were empirically verified as depicting a certain emotion. Further study would be needed to enlarge the amount of unambiguously interpretable AU combinations. While taking these restraints into account, the drench hypothesis on the other hand stresses that popular media examples can be more influential than the sheer frequency of TV characters or behaviours. Thus, the presented results found in a very popular TV program may have important impact on viewers’ perceptions even if the sample is limited in representativeness. Nevertheless, in further studies the results should be tested for generalization including more (popular) series and genres or expanding the analysis to verbal expressions of emotions. In addition coding not only police chiefs’ facial expressions, but those of other characters would allow for interaction testing. Furthermore, the question arises whether the change in the display of gender-stereotypical emotions in the media affects the viewers’ perception and stereotypes. While Eagly (1987) assumes an influence of the media for the perpetuation of traditional gender role attitudes, the results we found point to a more positive role of the media, at least in the case of Sex Roles (2010) 62:798–809 the studied German crime series: it does seem to take into account societal changes — or even go ahead of these, by showing more modern role images, or having not only one female chief inspector (as in ‘real German life’), but five of them. The effect of this rather progressive portrayal of the sexes in “Tatort” on viewers’ perceptions or stereotypes would also require further study. Conclusions In sum, the results show tendencies to depictions of gender stereotypical portrayals in the popular German TV series “Tatort”. Evidence for gender specific emotional behavior of women and men acting in the same professional role could be found. Furthermore, a tendency for these differences to diminish over time could be seen. Recall that Eagly (1987) pointed out that the cycle of role distribution could not be stopped until women would not have access to other than the traditionally female jobs in healthcare etc. She furthermore stressed the important role of the media for the perpetuation of traditional gender role attitudes by showing mainly traditional role images. Since the late 70ies, an increasing number of women has gained access to previously male domains of work (e.g., UNECE Statistical Database 2009). The picture observed in “Tatort” gives reason to assume that a change in the display of genderstereotypical emotions in the media is parallel to the change in society. Taking into account the role of the media in the perpetuation or formation of beliefs (Gerbner et al. 1994), the media, in this case TV, may also play a part in the change of the cycle of role division by “modeling” new, more modern roles for men and women in their broadcasts. Buck (1988) assumes that these changes are in progress “If we as a society decided that males are insufficiently expressive, and females insufficiently aggressive, we could choose to change the sorts of media models presented to our children in ways consistent with these images. To some extent, this has actually happened, in the increased depiction of assertive women and empathic men. The implications of these changes for the emotional lives of the audience members, however, have not been fully appreciated.” (Buck 1988, p. 56). It is likely that media and society mutually influence this development. All in all, it can be pointed out that the media play an important part when it comes to the formation or reinforcement of beliefs about gender. The results of the current study show that this role is not necessarily a negative one: The portrayal of men and women nowadays (in the TV series “Tatort”) seems to take into account the changed standards in society and does not stick to depictions of traditional/conservative role images. “Tatort” may even be a trailblazer for these changes in the studied vocational field. Since intercultural research (Williams and Best 1990b) has shown that the most modern Sex Roles (2010) 62:798–809 beliefs were found in European countries (including Germany), these results found for a prominent German television series may have implications for other countries, too. There is not only an increasing number of women playing the leading part in “Tatort” (Süss 1993), but the way they are portrayed has changed as well — at least when it comes to the display of emotions. Acknowledgements The authors would like to thank the editors and the anonymous reviewers for their very helpful comments. References Aldrich, N. J., & Tenenbaum, H. R. (2006). Sadness, anger, and frustration: Gendered patterns in early adolescents’ and their parents’ emotion talk. Sex Roles, 55, 775–785. Alexander, M. G., & Wood, W. (2000). Women, men, and positive emotions: A social role interpretation. In A. H. Fischer (Ed.), Gender and emotion: Social psychological perspectives (pp. 189–211). Cambridge: University Press. Alfermann, D. (1996). Geschlechterrollen und geschlechtstypisches Verhalten. [Gender roles and gender typical behavior]. Stuttgart: Kohlhammer. Ashmore, R. D., & Del Boca, F. K. (1979). Sex stereotypes and implicit personality theory: Toward a cognitive-social psychological conceptualization. Sex Roles, 5, 219–248. Averill, J. R. (1983). Studies on anger and aggression: Implications for theories of emotion. American Psychologist, 38, 1145–1160. Bakan, D. (1966). The duality of human existence. An essay on psychology and religion. Chicago: Mc Nally. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs: Prentice Hall. Bartsch, R. A., Burnett, T., Diller, T. R., & Rankin-Williams, E. (2000). Gender representation in television commercials: Updating an update. Sex Roles, 43, 735–743. Best, D. L., & Williams, J. E. (2001). Gender and culture. In D. Matsumoto (Ed.), The handbook of culture and psychology (pp. 195–219). New York: Oxford University Press. Brody, L. R. (1985). Gender differences in emotional development: A review of theories and research. Journal of Personality, 53, 102–131. Brody, L. (1999). Gender, emotion, and the family. Cambridge: University Press. Bronstein, P., Briones, M., Brooks, T., & Cowan, B. (1996). Gender and family factors as predictors of late adolescent emotional expressiveness and adjustment: A longitudinal study. Sex Roles, 34, 739–765. Brovermann, I. K., Brovermann, S. R., Clarkson, F. E., Rosenkrantz, P. S., & Vogel, S. R. (1972). Sex-role stereotypes: A current appraisal. Journal of Social Issues, 28, 59–78. Buck, R. (1988). Emotional education and mass media. A new view of the global village. In R. P. Hawkins, J. M. Weimann, & S. Pingee (Eds.), Advancing communication science: Merging mass and interpersonal perspectives (pp. 44–76). Beverly Hills: Sage. Bundesministerium für Familie, Senioren, Frauen und Jugend (2005). Gender-Datenreport. 1. Datenreport zur Gleichstellung von Frauen und Männern in der Bundesrepublik. [Gender Data Report. 1st report of data concerning the equilization of women and men in the Federal Republic of Germany]. Retrieved from http://www.bmfsfj.de/Publikationen/genderreport/01-Redaktion/ PDF-Anlagen/gesamtdokument,property=pdf.pdf. Calvert, S. L., & Huston, A. C. (1987). Television and children’s gender schemata. In L. S. Liben & M. L. Signorella (Eds.), New 807 directions in child development Vol. 38. Children’s gender schemata: Origins and implications (pp. 75–88). San Francisco: Jossey Bass. Calvert, S. L., Kotler, J. A., Zehnder, S. M., & Shockey, E. M. (2003). Gender stereotyping in children’s reports about educational and informational television programs. Media Psychology, 5, 139– 162. CEWS Statistikportal (2009). Frauenanteile an Habilitationen, NeuBerufungen, Professuren und C4/W3-Professuren, 1980 – 2008 [Women’s share of state doctorates, new appointments, professorships, 1980–2008] Retrieved from http://www.cews.org/statistik/ hochschulen.php?aid=50&cid=18. Coltrane, S., & Messineo, M. (2000). The perpetuation of subtle prejudice: Race and gender imagery in 1990s Television advertising. Sex Roles, 42, 363–389. Chory-Assad, R. M., & Cicchirillo, V. (2005). Empathy and affective orientation as predictors of identification with television characters. Communication Research Reports, 22, 151–156. Cosmides, L., & Tooby, J. (2000). Evolutionary psychology and the emotions. In M. Lewis & J. M. Haviland-Jones (Eds.), Handbook of emotions (2nd ed., pp. 91–115). New York: Guilford. Covert, J. J., & Dixon, T. L. (2008). A changing view: Representation and effects of the portrayal of women of color in mainstream women’s magazines. Communication Research, 35, 232–256. Diekman, A. B., & Eagly, A. H. (2000). Stereotypes as dynamic constructs: Women and men of the past, present, and future. Personality and Social Psychology Bulletin, 26, 1171–1188. Diekman, A. B., Goodfried, W., & Goodwin, S. (2004). Dynamic stereotypes of power. Perceived change and stability in gender hierarchies. Sex Roles, 50, 201–215. Dietz, T. L. (1998). An examination of violence and gender role portrayals in video games: Implications for gender socialization and aggressive behavior. Sex Roles, 38, 425–442. Duchenne, G. B. (1990). The mechanism of human facial expression. [Mécanisme de la physionomie humaine] In R. A. Cuthbertson (Ed.), (Trans.). Cambridge, NY: Cambridge University Press. (Original work published 1862). Eagly, A. H. (1987). Sex differences in social behaviour: A social role interpretation. Hillsdale: Erlbaum. Eagly, A. H. (2003). The rise of female leaders. Der Aufstieg von Frauen in hohe Führungspositionen. Zeitschrift für Sozialpsychologie, 34, 123–132. Eagly, A. H., Johannsen-Schmidt, M. C., & van Engen, M. L. (2003). Transformational, transactional, and laissez-faire leadership styles: A meta-analysis comparing women and men. Psychological Bulletin, 129, 569–591. Ekman, P., & Friesen, W. V. (1975). Unmasking the face. A guide to recognizing emotions from facial clues. Englewood Cliffs: Prentice Hall. Ekman, P., & Friesen, W. V. (1978). Facial action coding system. A technique for the measurement of facial action. Palo Alto: Consulting Psychologists Press. Ekman, P., & Friesen, W. (1982). False, felt, and miserable smiles. Journal of Nonverbal Behavior, 6, 238–252. Europäische Kommission/Referat Chancengleichheit für Frauen und Männer (1999). Das Bild der Frau in den Medien. Ein Bericht über bestehende Untersuchungen in der Europäischen Union. [The portrayal of women in the media. Report about studies in the European Union] Luxembourg: Amt für Amtliche Veröffentlichungen der Europäischen Gemeinschaften. Feingold, A. (1994). Gender differences in personality. Psychological Bulletin, 116, 429–453. Fischer, A. H. (Ed.) (2000). Gender and emotion: Social psychological perspectives. Cambridge, UK: Cambridge University Press. Fischer, A. H., & Manstead, A. S. R. (2000). The relation between gender and emotion in different cultures. In A. H. Fischer (Ed.), 808 Gender and emotion. Social psychological perspectives (pp. 71–94). Cambridge: University Press. Fischer, A. H., Rodriguez, P. M., van Vianen, A. E. M., & Manstead, A. S. R. (2004). Gender and culture differences in emotion. Emotion, 4, 87–94. Frank, M. G., & Ekman, P. (1997). The ability to detect deceit generalizes across different types of high-stake lies. Journal of Personality and Social Psychology, 72, 1429–1429. Frank, M. G., Ekman, P., & Friesen, W. V. (1993). Behavioral markers and recognizability of the smile of enjoyment. Journal of Personality and Social Psychology, 64, 83–93. Friesen, W. V., & Ekman, P. (1983). EMFACS: Emotional Facial Action Coding System. Unpublished manuscript. University of California, San Francisco. Ganahl, D. J., Prinsen, T. J., & Netzley, S. B. (2003). A content analysis of prime time commercials: A contextual framework of gender representation. Sex Roles, 49, 545–551. Gerbner, G. (1997). Gender and age in prime-time television. In D. A. Krischner & A. Kirschner (Eds.), Perspectives on psychology and the media (pp. 69–95). Washington, DC: American Psychological Association. Gerbner, G. (1998). Casting the American scene: A look at the characters on prime time and daytime television from 1994–1997. A Cultural Indicators Project Report [Internet]. Retrieved from http://www. media-awareness.ca/english/resources/research_documents/reports/ diversity/upload/Casting-the-American-Scene-Report-pdf.pdf. Gerbner, G., & Gross, L. (1976). Living with television: The violence profile. Journal of Communication, 26, 172–199. Gerbner, G., Gross, L., Jackson-Beeck, M., Jeffries-Fox, S., & Signorielli, N. (1978). Cultural indicators: Violence profile no. 9. Journal of Communication, 28, 176–207. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1994). Growing up with television. The cultivation perspective. In J. Bryant & D. Zillmann (Eds.), Media effects. Advances in theory and research (pp. 17–41). Hillsdale: Erlbaum. GdP — Gewerkschaft der Polizei (2002). Deutsche Polizei, 2/2002. [Police Union, German Police]. Glascock, J. (2001). Gender roles on prime-time network television: Demographics and behaviors. Journal of Broadcasting & Electronic Media, 45, 656–669. Grabe, M. E., & Drew, D. G. (2007). Crime cultivation: Comparisons across media genres and channels. Journal of Broadcasting & Electronic Media, 51, 147–171. Greenberg, B. (1988). Some uncommon television images and the drench hypothesis. In S. Oskamp (Ed.), Television as a social issue: Applied social psychology (Annual 8) (pp. 88–102). Newberry Park: Sage. Grossman, M., & Wood, W. (1993). Sex differences in intensity of emotional experience: A social role interpretation. Journal of Personality and Social Psychology, 65, 1010–1022. Hall, J. A. (1984). Nonverbal sex differences. Communication accuracy and expressive style. Baltimore: John Hopkins University Press. Hall, J. A., Carter, J. D., & Horgan, T. G. (2000). Gender differences in nonverbal communication of emotion. In A. H. Fischer (Ed.), Gender and emotion. Social psychological perspectives (pp. 97– 117). Cambridge: University Press. Harrison, K. (2003). Television viewers’ ideal body proportions: The case of the curvaceously thin woman. Sex Roles, 48, 255– 264. Hecht, M. A., & LaFrance, M. (1998). License or obligation to smile: The effect of power and gender on amount and type of smiling. Personality and Social Psychology Bulletin, 18, 1–26. Hess, U., Adams, R. B., Jr., & Kleck, R. E. (2005). Who may frown and who should smile? Dominance, affiliation, and the display of happiness and anger. Cognition & Emotion, 19, 515–536. Sex Roles (2010) 62:798–809 Hess, U., Banse, R., & Kappas, A. (1995). The intensity of facial expression is determined by underlying affective state and social situation. Journal of Personality and Social Psychology, 69, 280–288. Hess, U., Beaupré, M., & Cheung, N. (2003). Who to whom and why: Cultural differences and similarities in the function of smiles. In M. H. Abel (Ed.), An empirical reflection on the smile (pp. 187– 216). New York: Edwin Mellen. Hilton, J. L., & von Hippel, W. (1996). Stereotypes. Annual review of psychology, 47, 237–271. Hochschild, A. R. (2003). The managed heart. The commercialisation of human feeling. (20th anniversary edition). Berkeley: University of California Press. Hoffner, C. (1996). Children’s wishful identification and parasocial interaction with favorite television characters. Journal of Broadcasting and Electronic Media, 40, 389–402. Hoffner, C., & Buchanan, M. (2005). Young adults’ wishful identification with television characters: The role of perceived similarity and character attributes. Media Psychology, 7, 325–351. Hundhammer, T. (2008). Female=Submisse and Male=Assertive. Sexuality-priming leads to gender-based self-perception. Retrieved from http://kups.ub.uni-koeln.de/volltexte/2008/2464/ pdf/TanjaHundhammerDissertation.pdf. Jansz, J. (2000). Masculine identity and restrictive masculinity. In A. H. Fischer (Ed.), Gender and emotion. Social psychological perspectives (pp. 166–187). Cambridge: University Press. Koelbl, H. (2004). Die Kommissarinnen. [The female chief inspectors]. Berlin: Nicolaische Verlagsbuchhandlung. LaFrance, M. (2003). Facework and emotion work: The role of positive facial expression in constituting the social self. In J. P. Forgas & K. D. Williams (Eds.), The social self. Cognitive, interpersonal and intergroup perspectives (pp. 219–232). New York: Psychology Press. LaFrance, M., & Hecht, M. A. (2000). Gender and smiling: A metaanalysis. In A. H. Fischer (Ed.), Gender and emotion: Social psychological perspectives (pp. 118–143). Cambridge, UK: Cambridge University Press. LaFrance, M., Hecht, M. A., & Paluck, E. L. (2003). The contingent smile: A meta-analysis of sex differences in smiling. Psychological Bulletin, 129, 305–334. Lee, V., & Wagner, H. (2002). The effect of social presence on the facial and verbal expression of emotion and the interrelationships among emotion components. Journal of Nonverbal Behavior, 26, 3–25. Mealey, L. (2000). Sex differences. Developmental and evolutionary strategies. London: Academic. Merten, J. (2001). Beziehungsregulation in Psychotherapien. Maladaptive Beziehungsmuster und der therapeutische Prozess [Regulation of relationships in psychotherapy]. Stuttgart: Kohlhammer. Müller, U. (2000). Asymmetrische Geschlechterkultur in Organisationen und Frauenförderung als Prozess — mit Beispielen aus Betrieben und der Universität. In I. Lenz, H. Nickel, M. Hildegard, & B. Riegraf (Eds.), Forum Frauenforschung: Vol. 12. Geschlecht Arbeit - Zukunft. [Series women’s studies: Vol. 12. Gender - Work Future] (pp. 126–149). Münster: Westfälisches Dampfboot. Nabi, R. L., & Riddle, K. (2008). Personality traits, television viewing, and the cultivation effect. Journal of Broadcasting & Electronic Media, 52, 327–348. Osterloh, M., & Littmann-Wernli, S. (2000). Die “gläserne Decke”. Realität und Widersprüche. [The “glass ceiling”. Reality and contradictions]. Retrieved from http://www.iou.unizh.ch/orga/ downloads/publikationen/C79glaesernedecke.pdf. Pease, A., & Pease, B. (2001). Warum Männer nicht zuhören und Frauen schlecht einparken (12. Auflage). [Why men don’t listen and women can’t read maps]. Berlin: Ullstein. Potter, W., & Chang, I. (1990). Television exposure measures and the cultivation hypothesis. Journal of Broadcasting & Electronic Media, 34, 313–333. Sex Roles (2010) 62:798–809 Rosenberg, E. (2005). The study of spontaneous facial expressions in psychology. In P. Ekman & E. Rosenberg (Eds.), What the face reveals (2nd ed., pp. 3–18). New York: Oxford University Press. Rosenberg, E., & Ekman, P. (Eds.). (1998). What facial expression reveals about emotion, development, psychopathology, and health. New York: Cambridge University Press. Rivadeneyra, R., & Lebo, M. L. (2008). The association between television viewing behavior and adolescents dating role attitudes and behaviors. Journal of Adolescence, 31, 291–305. Scherer, K. R., & Wallbott, H. G. (1994). Evidence for universality and cultural variation of differential emotion response patterning. Journal of Personality and Social Psychology, 66, 310–328. Segrin, C., & Nabi, R. L. (2002). Does television viewing cultivate unrealistic expectations about marriage? Journal of Communication, 52, 247–263. Shanahan, J., & Morgan, M. (1999). Television and its viewers. Cultivation theory and research. Cambridge: Cambridge University Press. Shields, S. A. (2002). Speaking from the heart: Gender and the social meaning of emotion. Cambridge: University Press. Signorielli, N., & Bacue, A. (1999). Recognition and respect: A content analysis of prime-time television characters across three decades. Sex Roles, 40, 527–544. Smith, S. (2007). Perps, pimps, and provaocative clothing: Examining negative content patterns in video games. In P. Vorderer & J. Bryant (Eds.), Playing video games. Motives, responses, and consequences (pp. 57–76). Mahwah: Lawrence Erlbaum Ass. Süss, D. (1993). Der Fernsehkrimi, sein Autor und die jugendlichen Zuschauer. Medienkommunikation aus drei Perspektiven am Beispiel des “Tatort”-Krimis “Kameraden”. [The crime scene, the author and the juvenile viewers. Media communication in three perspectives exemplified at the Tatort-movie “The buddies”]. Bern: Huber. Thibault, P., Gosselin, P., Brunel, M.-L., & Hess, U. (2009). Children’s and adolescents’ perception of the authenticity of smiles. Journal of Experimental Child Psychology, 102, 360–367. Thompson, T. L., & Zerbinos, E. (1995). Gender roles in animated cartoons: Has the picture changed in 20 years? Sex Roles, 32, 651–673. 809 Tomada, G., & Schneider, B. H. (1997). Relational aggression, gender, and peer acceptance: Invariance across culture, stability over time, and concordance among informants. Developmental Psychology, 33, 601–609. UNECE Statistical Database (2009). Labour force by Age, Sex, Measurement, Country and Year. Retrieved from http://w3. unece.org/pxweb/DATABASE/STAT/30-GE/03-WorkAndeco nomy/03-WorkAndeconomy.asp. Von Salisch, M., Kristen, A., & Oppl, C. (2007). Computerspiele mit und ohne Gewalt. Auswahl und Wirkung bei Kindern. [Computer games with and without violence. Selection by children and effects on children]. Stuttgart: Kohlhammer. Van Engen, M. L., & Willemsen, T. M. (2004). Sex and leadership styles: A meta-analysis of research published in the 1990s. Psychological Reports, 94, 3–18. Wacker, H. (2000). Das große Tatort-Buch. Filme, Fakten und Figuren. [The big Tatort-book. Episodes, facts and actors]. Berlin: Henschel. Wagner, G. (1986). Entwicklung eines automatisierten Verfahrens zur Affektinterpretation. [Development of an automatized procedure for the interpretation of emotions] (unpublished manuscript). Fachrichtung Psychologie, Universität des Saarlandes. Wallbott, H. (1988). Big girls don’t frown, big boys don’t cry. Gender differences of professional actors in communicating emotion via facial expression. Journal of Nonverbal Behavior, 12, 98–106. Wilson, B. J., & Drogos, K. L. (2007). Preschoolers’ attraction to media characters. Paper presented at the annual meeting of the NCA 93rd Annual Convention, TBA, Chicago, IL Retrieved from http://www.allacademic.com/meta/p192599_index.html. Williams, J. E., & Best, D. L. (1982). Measuring sex stereotypes: A multination study. Newbury Park, CA: Sage. Williams, J. E., & Best, D. L. (1990a). Measuring sex stereotypes: A multi-nation study (Revised edition). Newbury Park: Sage. Williams, J. E., & Best, D. L. (1990b). Sex and psyche: Gender and self viewed cross-culturally. Newbury Park: Sage. Wittenbrink, B., Hilton, J. L., & Gist, P. L. (1998). In search of similarity: Stereotypes as naive theories in social categorization. Social Cognition, 16, 31–55.