Abstract
Research into journalism and gender to date has found somewhat contradictory evidence as to the ways in which women and men practice journalism. Some scholars claim that women have inherently different concepts and practices of journalism and that this has led to a feminization of journalism, others have found little evidence to suggest that men and women differ significantly in terms of their role conceptions. While numerous studies have been conducted into this issue around the world, few have taken a truly comparative approach. This article presents results from a large-scale comparative survey into gender differences in journalists’ professional views in 18 countries around the world. Results suggest that women and men do not differ in any meaningful ways in their role conceptions on either the individual level or in newsrooms dominated by women, or in sociocultural contexts where women have achieved a certain level of empowerment.
Over the past few decades, the number of women in journalism has grown remarkably around the world, with the global percentage of women reporting the news having risen from 28% in 1995 to 37% in 2009 (Gallagher, 2010). In Europe, women’s average proportion of journalists has even been put at 47% across the continent (Witt-Barthel, 2006). Nonetheless, female journalists have not yet reached a critical mass in ‘serious’ news beats, and they remain a minority in the higher echelons of news organizations where a glass ceiling effect continues to limit women’s promotion to key decision-making positions (Chambers et al., 2004; Fröhlich and Lafky, 2008; Peters, 2001). Thus, when it comes to symbolic power in the field, gender balance is still far from a reality (Robinson, 2008), with newsroom culture still organized around a ‘man-as-norm and woman-as-interloper structure’ (Ross, 2001: 535).
In the late twentieth century, there was hope that the news would change into directions more relevant to women if only there were more female journalists, or if women reached a ‘critical mass’ in the newsroom (Rush et al., 1982; van Zoonen, 1998). Chambers et al. (2004: 122) contend that news agendas and priorities have clearly changed with the increasing number of women journalists, and that men and women seem to adopt distinctly different approaches to reporting. Key characteristics include an increased emphasis on personalization, foregrounding of personal standpoints over professional ones, as well as an increased use of women as sources. However, the gender determination hypothesis, as we would call it, is problematic for at least two major reasons, as van Zoonen (1998) rightly points out. First, it assumes that journalists have sufficient autonomy to perform in a uniquely individual manner. Second, it rests on the belief that women journalists are distinguished more by their femininity than by any other dimension of identity, like professionalism or ethnicity.
There is no consensus over the extent to which gender matters in the process of news production. Evidence is ambiguous and often dependent on whether results are based on journalists’ self-declarations or actual content. While most surveys of journalists have not been able to present substantial differences, content analyses do point to gender-related patterns. The question, however, is if these differences really constitute a ‘feminization’ of news that is driven by the growing number of women journalists, or if they simply reflect general changes in the media that are largely unrelated to gender effects (Chambers et al., 2004; van Zoonen, 1998). The causal relationship between journalism and gender might actually work in the opposite direction: the ongoing transformation of news-making may open up journalism as a profession for women (van Zoonen, 1998).
Holding gender responsible for distinctive journalistic orientations and practices may also be overly simplistic and obscure more complex and fine-grained realities on the ground. Ross (2001) argues that gender alone will not make a difference in changing the culture of newsrooms or in the type of news produced. Gender roles are more complex than simply saying men always favor one approach and women another, with some men holding more feminine values than typically masculine values, and vice versa (Sakr, 2004). It is this construction of different versions of ‘femininity’ and ‘masculinity’ that is often referred to as ‘doing gender’ or ‘gendering’ (de Bruin, 2004: 4).
The purpose of this article, therefore, is to explore the relationship between journalists’ professional values and gender across the individual, organizational and societal levels of analysis. The following basic research questions have motivated this article: Does gender make a difference in terms of journalists’ professional predispositions? Does this relationship change when women journalists equal or even outnumber men in the newsroom? Does the larger cultural context have any substantial effect on gender differences in journalists’ professional views? Results are based on 1800 interviews with professional journalists in 18 countries around the world. Thus, this article explores new ground as actually very little research on gender and journalism is comparative (Steiner, 2009).
A brief clarification of terminology is in order, however. The term ‘sex’ is usually applied to refer to the physiological differences between the female and male. The concept of ‘gender’, on the other hand, is used to acknowledge that what constitutes ‘women’ and ‘men’ may be as much, or even more, a product of socialization as of biology (Best and Williams, 1997). In the context of news work, ‘sex’ often translates into gendered stereotypes of news practice that are socially and culturally constructed. We will therefore use the term ‘gender’ throughout the article since we are interested in the implications of cultural constructions of gender on professional values.
Gender differences in professional views
Van Zoonen (1998: 36) defines four major areas in what she calls ‘the gendered nature of journalism’. These include the selection of topics (men report hard news while women focus on soft news); story angles (men focus on facts and sensation, while women are concerned with backgrounds and effects, as well as compassion); the use of sources (men predominantly report about other men, women focus on women); as well as ethics (the masculine nature of journalism is detached, while the feminine nature is concerned with audience needs). Of these areas, most research has been conducted on journalists’ professional views. We are therefore dealing with journalists’ subjective perceptions of their professional roles and epistemological perspectives. These two domains mark an important part of the contested terrain of journalism’s professional culture. As sets of culturally negotiated professional values, these conventions operate mostly behind the backs of individual journalists. Partly following a model proposed by Hanitzsch (2007), we argue that differences in journalists’ professional views are articulated with respect to two fundamental domains:
The area of institutional roles refers to the normative and performed functions of journalism in society that are mostly referred to as role perceptions, consisting of three dimensions: Interventionism reflects the extent to which journalists pursue a particular mission and promote certain values. The distinction tracks along a divide between two ideal-types of journalist – one involved and socially committed; the other detached and uninvolved. Power distance refers to a journalist’s position toward loci of power in society. It distinguishes between an approach to journalism as ‘Fourth Estate’ and ‘watchdog’ on one hand, and an opportunist, loyal or collaborative mode toward power centers on the other. Market orientation accounts for journalists’ perspectives on audiences as either citizens or consumers. In the latter approach, journalism cultures strongly submit to the market logic, while the former gives priority to the public interest and the creation of an informed citizenry.
The domain of journalistic epistemologies relates to the accessibility of reality and the nature of acceptable evidence. Here, objectivism marks the distinction between two fundamental beliefs: one claims the existence of an objective truth ‘out there’ that can be reported ‘as it is’, the other concedes that news is an inevitably subjective representation of the world. Empiricism, on the other hand, refers to the relative weight given to an empirical justification of truth, emphasizing observation, measurement, evidence and experience, and an analytical justification by accentuating reason, ideas, values, opinions and analysis.
These dimensions were found to be meaningful ‘areas of disagreement’ toward journalism’s professional identity in a recent comparative study of journalists (Hanitzsch et al., 2011). Here, it is important to distinguish professional identity – carried by an imagined community of journalists – from an organizational identity that is bound to the ‘cultural and spatial territory’ of a given newsroom (de Bruin, 2000: 229).
If male and female journalists differ in their professional identities, this variation should also be apparent with respect to the above-mentioned five dimensions. Indeed, interviews with journalists in Israel as well as the United States have indicated that women journalists are more attuned to audience needs than men (Lavie and Lehman-Wilzig, 2005; Weaver, 1997). Furthermore, women were found, or believed, to favor a more involved and less detached approach (Melin-Higgins, 2004; van Zoonen, 1998). We therefore advance the following two hypotheses:
H1: Women journalists tend to be more attuned to audience needs than men.
H2: Women journalists tend to value detachment less than their male colleagues.
In addition, a number of studies have found female journalists to encourage positive news reporting, or at least focus on the positive aspects of stories (Craft and Wanta, 2004; Kim and Kim, 2005; Rodgers and Thorson, 2003). If that is the case, it seems plausible they are less likely to use a confrontational approach such as being a ‘watchdog’ of political and economic elites, leading us to the following hypothesis:
H3: Women journalists tend to see themselves less in a watchdog role than men.
With respect to journalistic epistemology, empirical evidence is scarce, but van Zoonen (1998: 45) argues that women tend to favor ‘a more human and involved approach that is seemingly at odds with professional values of objectivity and detachment’ that dominate among their male colleagues. The foregrounding of a ‘personal standpoint over a professional one’ is also assumed to be characteristic of news in which women journalists are heavily involved (Chambers et al., 2004: 122). Lumby (1994: 50) believes that ‘facts, objectivity and the public sphere belong to the men’, implying that female journalists favor a more analytical and subjective approach to reporting, a view echoed by Mahtani (2005: 301), who maintains that ‘masculinized newsroom practices’ tend to encourage rationality and objectivity. We therefore advance our fourth hypothesis:
H4: Women journalists tend to value the importance of subjectivity, analysis and opinion more than men.
Surveys of journalists have generally not found any substantial differences between female and male journalists in terms of their professional orientations. Weaver et al. (2007) found that gender was not a significant predictor of US journalists’ perceptions of their professional roles. Similar results were produced by studies in Indonesia (Hanitzsch, 2006), Israel (Lavie and Lehman-Wilzig, 2005) and Tanzania (Ramaprasad, 2001). Weaver (1997) concluded that newsroom and community environments may exhibit stronger influences on journalists’ professional values than gender, a conclusion echoed by Keuneke et al. (1997), who argue that gender roles in journalism seem to be superimposed by professional role models.
Most studies examine the (potentially) gendered practices and orientations of journalists at the individual level of analysis, assuming that gender is some sort of essentialist force that drives female and male journalists to think and act in different ways. Professional views and behaviors, however, are always deeply embedded in organizational and societal contexts. Furthermore, the gendered nature of news work extends well beyond the level of the individual journalist. It seems plausible that female journalists may be more likely to embrace different – perhaps ‘feminine’ – professional values in newsrooms with higher percentages of women. In their study of issue agendas and story focus at websites of US newspapers, Craft and Wanta (2004) found that newspapers with a high percentage of women in managerial positions did tend to cover news in a more positive light. Further, female and male reporters tended to cover a similar agenda of issues only when they worked for newspapers with a high percentage of women in managerial positions. The latter finding particularly seems somewhat counterintuitive. If it is true that women have different approaches to news making, it makes sense to expect female journalists to apply ‘feminine’ standards, especially in news organizations where they constitute a significantly large group. As a result, newsrooms as a whole may approach the news in a different way, or the differences between women journalists and their male counterparts grow larger. We therefore propose the following hypothesis at the organizational level:
H5: Women and men journalists are more likely to disagree in their professional views in news organizations where female journalists equal or even outnumber men.
The sociocultural environment in which journalists work is another important context of their values and practices. One well-known attempt to measure a society’s general tendency in terms of the distribution of gender roles is Hofstede’s Masculinity Index (MAS). According to Hofstede’s (2001: 297) definition, Masculinity quantifies the extent to which social gender roles in a given society are clearly distinct: ‘Men are supposed to be assertive, tough, and focused on material success; women are supposed to be more modest, tender, and concerned with the quality of life.’ The opposite of Masculinity, Femininity, denotes a society in which social gender roles overlap. A higher Masculinity Index value therefore indicates a higher polarization between the values of men and women, and thus implies strong gender differentiation in that society. Although Hofstede’s work has been criticized for conceptual and methodological shortcomings (McSweeney, 2002), it is widely applied in cross-cultural psychology, sociology and management research. In the area of news production, we would expect to find more gender-related differences between female and male journalists in societies that score higher on Hofstede’s Masculinity Index as gender roles are more likely to exhibit stronger polarization:
H6: There are larger gender-related differences between female and male journalists in societies that score higher on Masculinity.
Another composite measure related to gender imbalance at the societal level is the Gender Empowerment Measure (GEM) used by the United Nations Development Program for its annually published Human Development Reports. Unlike Hofstede’s index, which was calculated from survey responses, the GEM is based on aggregate data collected for three dimensions: political representation, indicated by female representation in national parliaments; representation in the senior positions in the economy, proxied by female representation as legislators, senior officials and managers, and representation as professional and technical workers; and power over economic resources, indicated by earned incomes of males and females (Klasen, 2006; UNDP, 2007). The GEM was originally intended to measure ‘whether women and men are able to actively participate in economic and political life and take part in decision making’ (UNDP, 1995: 73). Similar to the Masculinity Index, the GEM has some conceptual, methodological and empirical problems that are mostly related to an elite bias in two of its components and the availability of data (Klasen, 2006). Although the measure has never been used in journalism and mass communication research, we suspect differences between women and men journalists to be more outspoken in contexts that score lower in terms of GEM and therefore advance the following hypothesis:
H7: There are larger gender-related differences between women and men journalists in societies that score lower in terms of Gender Empowerment.
Methodology
To test our hypotheses, this study examines data from simultaneously conducted surveys in 18 countries around the world, including Australia, Austria, Brazil, Bulgaria, Chile, China, Egypt, Germany, Indonesia, Israel, Mexico, Romania, Russia, Spain, Switzerland, Turkey, Uganda and the United States. These countries, drawn from all six inhabited continents and in line with the idea of a most different design (Przeworski and Teune, 1970), constitute a varied mix of political and cultural contexts, as well as a variety of levels of Masculinity and Gender Empowerment. Further, the availability of committed researchers in the respective countries was an important criterion for inclusion.
A total of 100 journalists were interviewed in each of the 18 countries, following a predetermined quota scheme in order to sample a variety of news organizations. The interviewed journalists covered a broad array of beats and subject areas, stretching from political reporting to celebrity news. Thus, we were able to account for a large diversity of professional identities in journalism, as well as for the differential gender distribution in the various areas of coverage, with women journalists often assigned to lifestyle, background and ‘soft’ beats (Ross, 2001). Basic sample parameters are provided in Table 1.
Basic sample parameters.
Respondents had to have at least some ‘editorial responsibility’, in line with similar studies of journalists’ role conceptions (Weaver and Wilhoit, 1986: 168). Generally, five journalists were chosen from each news organization, representing three levels of newsroom hierarchy (senior managers, such as chief editors and their deputies; middle-level managers, such as senior editors and desk heads; as well as the non-management level, such as general reporters). One journalist each was randomly chosen from the top two levels, while three were sampled from the non-management level in each organization.
News organizations were chosen according to a common quota scheme in order to include a mix of national and regional or local media. We further differentiated between print media that were citizen-oriented (so-called ‘quality’ publications) as well as consumer-oriented (or popular) organizations, and broadcast media that were publicly or privately owned (see Table 2). Popular media were chosen based on audience size, while quality publications were selected according to their agenda-setting power. We excluded online-only media because the degree of their institutionalization still varied considerably across countries during the time of fieldwork.
Sample.
A total of 356 newsrooms were approached for this study, with 22 refusing to participate. These were subsequently replaced. We also had to substitute 236 respondents out of a total of 1800 journalists due to refusal. Fully standardized surveys were conducted mostly by telephone, but because telephone interviews were impractical or viewed with distrust by journalists in some countries, surveys were conducted face-to-face in Bulgaria, Egypt and Indonesia, and also partly in China and Mexico. In Turkey, journalists completed questionnaires on their own with a researcher present. Fieldwork was carried out between October 2007 and June 2009.
To examine journalists’ professional views, we compiled two lists of statements. The first list of 12 items was designed to measure the relative importance of institutional roles and consisted of a list that described ‘some of the things the news media do or try to do’. Respondents were given five answer options: ‘extremely important’, ‘very important’, ‘somewhat important’, ‘little important’ and ‘not important at all’. The second list of eight items was intended to capture journalists’ epistemological beliefs and was introduced thus: ‘The following statements describe different approaches to news coverage.’ Response options were ‘strongly agree’, ‘somewhat agree’, ‘neither agree nor disagree’, ‘somewhat disagree’ and ‘strongly disagree’. A comprehensive list of the indicators is documented in Table 3.
Gender differences in journalism cultures.
p < .05; **p < .01; ***p < .001.
Results
Table 3 reports group differences between women and men journalists with regard to their professional views. A simple inspection of these differences and their test of significance shows that Hypothesis 1, which assumed that women journalists tend to be more attuned to audience needs than men, was not supported. There was no disagreement between female and male journalists regarding the relevance of providing information that is interesting to the audience. While women and men disagreed on the importance of attracting a large audience, it was the male journalists who cared significantly more about large audiences than their female colleagues (t = 2.81; d.f. = 1747; p < .001). The size of the effect (Cohen’s d), however, was not particularly large (d = .13).
Hypothesis 2 was supported. Women journalists do tend to appreciate the value of detachment significantly less than their male counterparts (t = 3.19; d.f. = 1765; p < .001; d = .15). The lower regard for detachment, however, does not imply that female journalists favor a more involved approach. There were no significant gender differences in the journalists’ responses to the statements 1, 2 and 3 in Table 3.
Hypothesis 3 was also supported. Male journalists did see themselves more than women in the role of a watchdog of the government (t = 2.32; d.f. = 1774; p < .05; d = .11) and a watchdog of business elites (t = 2.59; d.f. = 1761; p < .01; d = .12). These effects, however, diminish if country is entered into the equation. In a two-way ANOVA including gender and country as factors, the effects of gender on the two watchdog roles were rendered non-significant.
Hypothesis 4 was clearly not supported. Male journalists did not value the importance of objectivity, facts and evidence more than their female counterparts, nor are women more attuned to subjectivity, analysis and opinion than men (statements 13–20 in Table 3). For all other aspects of journalists’ professional views, respondents of both genders tended to agree with each other, with only one exception: men found it more important to motivate people to participate in civic activity (t = 2.76; d.f. = 1766; p < .01; d = .14). Again, this difference diminished when we controlled for country variation in the outcome variables. Most of the remaining differences were minor, with effect sizes of .06 and less. This means that for these traits, gender explains less than 1% of the variance in the outcome. Importantly, none of the effects of gender on aspects of journalists’ professional orientations even crossed the conventional threshold for small effects (d ≥ .20). Based on this evidence, it is fairly safe to say that gender does not make a difference with respect to most aspects of journalists’ professional views.
Since a pooled analysis does not provide a fine-grained picture of nation-specific results, we also inspected the country-level breakdown of our findings. Because the number of cases is considerably smaller in a country-level analysis, we doubled the significance threshold to p < .10 (see Table 4; differences that are significant only at the p < .10 level are marked with an asterisk). In order to facilitate intuitive interpretation of the results, Table 4 only reports information about significant differences and whether women or men scored higher on the investigated aspects of journalists’ professional orientations. Two aspects were omitted from the table – ‘Influence public opinion’ and ‘Remain strictly impartial’ – as country-level analysis revealed no significant differences.
Gender differences in journalism cultures across countries
Only significant relationships are reported (p < .05): ♀ = women rate this aspect higher than men; ♂ = men rate this value higher than women; * significant only at p < .10.
Table 4 largely confirms the pattern found in the pooled analysis. Women and men agreed on most aspects of their professional views. Results from the country-level analysis slightly deviate from the general pattern in only a few cases: women exhibited higher support for three items – ‘Set the political agenda’, ‘Do not allow beliefs and convictions to influence reporting’ and ‘Always make clear which side has the better position’ – although pooled analysis did not indicate any significant differences.
The most inconsistent results were found for ‘Advocate for social change’, ‘Support official policies’, ‘Convey positive image of political and business leadership’ and ‘Provide analysis of events and issues’. Women valued these aspects of their work significantly more in some countries, while in other countries, it was their male colleagues who scored higher on these dimensions. A pooled analysis across all countries may have only limited power to make these differences visible as they tend to neutralize each other. This is also true for a few other aspects of the journalists’ professional views, including ‘Provide interesting information’, ‘Provide citizens with political information’ and ‘Facts speak for themselves’.
Hypothesis 5, which assumed that women and men were more likely to disagree in their professional views in news organizations with higher percentages of women, was not supported. To test it, we divided the newsrooms in our sample into two groups according to gender distribution: newsrooms with fewer than 50% female journalists and newsrooms with 50% or more women. Gender differences in the first group were then compared to those in the second group. A t-test did not reveal any significant results, meaning that across the board, female and male journalists did not disagree more in newsrooms with high percentages of women. A two-way ANOVA did not indicate any significant interaction effects between gender and gender distribution of newsrooms. Significant main effects of gender distribution were found for only six variables. In newsrooms where women equal or outnumber men, journalists were significantly more likely to provide analysis (see Table 5). In those dominated by men, journalists scored significantly higher on the following aspects: ‘Act as watchdog of government’, ‘Act as watchdog of business elites’, ‘Convey positive image’, ‘Provide political information’, as well as ‘Stay away from unverified information’. These effects, however, were very minor, as their partial η values indicate.
Gender differences and organization (gender distribution in the newsroom).
Only significant main effects are reported: *p < .05; **p < .01; ***p < .001.
At the national level of analysis we hypothesized that there would be larger gender-related differences between female and male journalists in societies that score higher on Hofstede’s Masculinity Index. A visual inspection of marginal means indeed pointed to such a tendency for nine of the 26 investigated aspects of journalism culture. However, only for ‘Support official policies to bring about prosperity’ did the two-way ANOVA indicate a significant interaction effect (F = 4.894; d.f. = 1; p < .05). As the difference in the mean scores of women and men journalists widened only slightly from .08 to .21, with differences being larger in countries that scored higher on Masculinity, this effect was marginal. Overall, we have to reject Hypothesis 6. In addition, ANOVAs also revealed significant main effects of Masculinity on journalists’ professional views for 14 indicators (see Table 6).
Gender differences and national context (Hofstede’s Masculinity and Gender Empowerment).
Only significant main effects are reported: ***p < .001; **p < .01; *p < .05.
The situation is similar for Hypothesis 7. A visual inspection of marginal means did suggest a slight increase in gender-related differences for three aspects of journalists’ professional orientations in societies with lower Gender Empowerment scores: ‘Motivate people to participate in civic activity and political discussion’, ‘I do not allow my own beliefs and convictions to influence my reporting’ and ‘I always make clear which side in a dispute has the better position’. However, none of these differences was large enough to produce a significant interaction between gender on the individual level and Gender Empowerment in a two-way ANOVA. Moreover, the results clearly contradicted our hypothesis for five indicators, as differences between women and men journalists tended to be even larger in societies with higher levels of Gender Empowerment. We therefore have to also reject Hypothesis 7. As with Masculinity, ANOVAs detected several significant main effects of Gender Empowerment on 14 aspects of journalists’ professional orientations (see Table 6).
Discussion
The findings of this study reinforce much of the existing survey-based evidence on the role of gender in journalists’ professional views. For most of the indicators we used, we were unable to find any substantial support for the assumption that women and men differ significantly in the way they approach their jobs. This is the case on all levels of analysis, including individual predispositions, the newsroom and the sociocultural environment.
On the individual level, even for the five role perceptions which showed statistically significant differences (out of 20 statements in total), none reached the critical threshold for small effects. This may not be very surprising given that past surveys were similarly unable to find meaningful gender differences in journalists’ role perceptions (Hanitzsch, 2006; Lavie and Lehman-Wilzig, 2005; Ramaprasad, 2001; Weaver et al., 2007). We thus find ourselves agreeing with Steiner’s (2009: 121) remark, discussed earlier, that women ‘largely adopt journalism’s structures as part of the profession and choose to embrace its reward system’.
However, it was somewhat surprising to find so little meaningful variation in areas that had traditionally been marked for such differences. While there exists some evidence in the literature to suggest that women are more oriented toward audience needs than men, our results actually attest to the opposite. On the whole, men are more likely to concentrate on news that will attract the widest possible audience. In terms of a second related role conception – providing information that is interesting to readers – we could find no differences in the pooled analysis. Our findings therefore clearly do not support the assumption that women are more attuned to audience needs. Similarly, the assumption that objectivity, facts and evidence are more valued by male journalists was not supported either.
Only a few hypothesized gender effects proved significant. Our results did indicate that women journalists tend to value detachment less than their male colleagues, and that they find the watchdog role less relevant than men. However, the differences were not particularly pronounced, with effect sizes not even small judged by conventional standards of social science research. With such a relatively large sample of journalists, minor differences can turn out significant, even if they are not meaningful in practice. The examination of individual country differences largely supported this picture, with often inconsistent gender differences across countries. Thus, the findings on the individual level demonstrate that men and women generally do not differ sufficiently on any aspect of their professional views to claim that gender has any substantial impact.
On the level of the newsroom, our hypothesis – partly based on evidence provided by Craft and Wanta (2004) – that women and men differed more significantly in newsrooms where women outnumbered men was also not confirmed. This casts some doubt on the proposition that journalistic culture will become more ‘feminized’ once women make up a substantial proportion of newsroom staff. In fact, it seems that professional socialization processes override any feminine values that may work against traditional masculine values.
There may also be alternative explanations. Cultural change usually takes time, and the same may be true for the transition from male-dominated newsroom cultures to a more ‘feminine’ approach. Moreover, better representation of women in newsrooms alone may not be a sufficient condition for such a change, at least as long as men are still holding key positions in the news management and decision-making process. Furthermore, newsroom cultures may actually change as a result of women increasingly entering the profession, but men increasingly embrace the new values of their changed working environment, rather than uphold ‘masculine’ values. This would explain why we did not find larger differences between female and male journalists in non-male-dominated newsrooms.
At the national level, we also found no substantial evidence to support the argument that women and men differ more significantly in societies where women have fewer rights and gender roles may be more polarized. There may be several explanations for this. One might be that existing cross-national differences in the levels of Masculinity and Gender Empowerment may not be relevant enough to processes of news production to engender greater or smaller differences between women and men journalists. Gender gaps in journalists’ professional views may be largely resistant to the contextualizing effects of Masculinity and Gender Empowerment.
Another explanation pertains to the somewhat problematic nature of the two aggregated measures and the fact that gender and power are widely contested concepts. Hofstede’s work on culture has been heavily criticized for conceptual and methodological reasons. McSweeney (2002: 102), for instance, argues that Hofstede’s ‘evidence’ for national cultures is merely a self-fulfilling prophecy as any other classification than country would also have produced response differences. Similarly, the GEM has been criticized for its inability to capture more opaque workings of gender, an elite bias and limited data availability (Charmes and Wieringa, 2003; Klasen, 2006).
However, while we could not confirm our hypothesis in relation to more pronounced differences between men and women related to MAS or GEM scores, our analyses did show some significant main effects for both indices. This means that journalists as a group in high MAS and high GEM countries did differ significantly from journalists in countries that score low on both indices, largely irrespective of individual gender. For example, in countries with higher levels of masculinity, journalists are more likely to act as watchdogs of business elites, while they are less keen to influence public opinion (see Table 5). In societies that score higher on GEM, they are more willing to set the political agenda, influence public opinion, advocate for social change, support official policies and provide analysis, whereas they are less eager to provide interesting information. This could mean that a ‘feminization’ of journalism may be a grand effect of societal developments.
The direction of these trends depends very strongly on the type of index one applies to the data. For example, journalists in low-MAS countries are more likely to embrace an involved approach to their work, while in countries which score high on GEM, journalists are less, rather than more likely to embrace an involved approach. The only aspect of professional culture where we can see some consistency is the importance of being an absolutely detached observer. Journalists are more likely to favor this approach in both low-MAS and high-GEM countries.
Again, a reason for such results may lie in significant differences between the indices. Many of the societies classified as highly ‘masculine’ according to Hofstede also score high on GEM. Intuitively, this does not make sense and may point to problems with the validity and timeliness of these contextual measures. Furthermore, the differences between journalists’ professional views are most likely related to a number of contextual factors, of which Masculinity and Gender Empowerment constitute only two elements. For example, Gender Empowerment is positively correlated with economic wealth (GDP), quality of women’s rights and democratic performance, and negatively related to income inequality (Slusser, 2009). Masculinity is not associated with macro-economic measures, but positively correlated with population growth and population size, and negatively related to geographic latitude (Hofstede, 2001).
Conclusion
This analysis has chosen a fairly ‘traditional’ approach, one which de Bruin and Ross (2004: vii) have referred to as ‘body count’ in their argument that a concentration on the proportion of women in newsrooms is too narrow a focus. While we agree on the need for a more thorough investigation of the complex relationship between gender, professional ideology and organizational culture, we also think that survey-based comparative research is a useful tool in order to establish a general pattern that goes beyond national or organizational particularities.
This study has presented strong evidence to reject claims that men and women differ significantly in their role conceptions and epistemological beliefs. Where differences exist, they tend to be small and often negligible, supporting numerous previous studies in this field. We were unable to provide substantial evidence to support assumptions of the existence of any particular ‘feminine’ view of journalism. Rather, we found that male and female journalists around the world mostly think about their work in relatively similar terms. One reason may be that journalists appropriate their country’s dominant journalistic ideology early in their careers. Given the dominance of masculine values in the newsroom, women may feel pressured to become ‘one of the boys’ (North, 2009: 208).
Neither of the three levels of analysis – individual, newsroom and sociocultural – showed meaningful or strong patterns of gender differences. Some findings contradicted past research, such as the fact that men were more strongly focused on providing news that would attract the widest audience. On the newsroom level, as well as the level of the sociocultural environment, tendencies for larger gender gaps were also negligible. Thus, a major conclusion of this study is that a journalist’s gender alone is not a significant determinant of journalists’ professional views.
The similarities between female and male journalists are perhaps less surprising if one considers that women journalists almost everywhere are judged by male standards and performance criteria (de Bruin, 2000; Gallagher, 2008), necessitating an adherence to journalistic values created by a dominant male proportion over centuries. Furthermore, differences between the sexes are often much more opaque. In Canada, Robinson (2005: 92) noted that women tend to act ‘as though they were equal partners in the heterosexual newsroom and use their gender strategically and merely under certain circumstances’ (original emphasis). Similarly, Steiner (2009: 121) concluded that ‘women recognize that many of their male colleagues are sexist, but they largely adopt journalism’s structures as part of the profession and choose to embrace its reward system’.
The often contradictory and inconclusive nature of the evidence regarding the impact of gender on journalistic values may well be related to the fact that female journalists seem to have a more fragmented professional identity (Melin-Higgins, 2004). At the same time, it may be, at least partly, a question of whether the data has been generated through content analysis or interviews with journalists (Chambers et al., 2004). Lavie and Lehman-Wilzig (2005) demonstrated that different methodologies do indeed produce inconsistent findings. While journalists’ survey responses regarding professional values did not indicate any significant gender-related differences, their news stories did. Future research needs to pay closer attention to this relationship between attitudes and practice.
At the same time, we discovered some notable findings about national journalism cultures when comparing countries with high and low GEM. It makes intuitive sense for journalistic values to shift in directions associated with ‘feminine values’ in countries where women are more empowered, especially when we consider the relatively weak explanatory power of the organizational context. It seems the key to a feminization of journalism does not necessarily lie in an increase in the number of women in journalism, but in wider societal developments that encourage a more active and empowered role for women in society. However, the results with respect to Masculinity and Gender Empowerment are somewhat contradictory and contingent on the type of index one applies. Consequently, societal influences on the gendered nature of journalism cultures deserve closer scrutiny.
Footnotes
Funding
The study was funded by the German Research Foundation, Swiss National Science Foundation, Rothschild-Caesarea School of Communication at Tel Aviv University and School of Journalism and Communication at the University of Queensland.