Article 6 - Lotte (TRYKKEKLAR) 163 “No, Gender doesn’t make a Difference…?” Studying Negotiations and Gender in Organizations Lotte Bloksgaard Abstract. This article focuses on methodological challenges and strategies with regard to studying gender in organizations. Work organizations are often thought, theorized and talked about as gender-neutral arenas and, therefore, gender is often seen as irrelevant and of no importance in modern work-life. There is also often a discrepancy in relation to gender between discourse and practice, which makes it difficult to capture the significance of gender in organizations by the interview method alone. On the basis of two empirical studies, focusing on the significance of gender in negotiations of wage and parental leave in the work place, the article explores some of the challenges and complexities involved when researching gender as a social category of difference, which produces inequalities in organizations. Furthermore, the article demonstrates and discusses how applying and combining different methods and research strategies – for example following specific negotiations to a conclusion - provide insight into the production of gender in discourse and practice at the work place level. Thus, the article argues that reflexivity and methodological plurality are important when studying gender in work organizations. Keywords: inequality, difference, gender in organizations, work place negotiations, research relations, methodological strategies, mixed methods. Please cite this article as: Bloksgaard, L. (2012). “No, Gender doesn’t make a Difference…?” Studying Negotiations and Gender in Organizations. Qualitative Studies, 3(2): 163-180. Introduction This article focuses on how to study gender as difference and as a basis for the production of inequalities in modern work-life. A number of researchers point out and document that modern work-life and work organizations in the Western world are, to a large extent, perceived as gender- neutral (e.g. Acker, 1990, 1991, 2012; Korvarjärvi, 2003; Eriksson and Eriksson, 2003; Holgersson et al., 2004; Smithson and Stokoe, 2005; Højgaard, 2008; Bloksgaard, 2009). The term ‘gender-neutral’ refers to the notion that gender is of no significance within work organizations. Work-life is built on apparently gender-neutral ideas and routines; researchers argue that Human Resource Management policies (including competence assessments, wage determination processes and family friendly arrangements), for example, are assumed to be gender-neutral (Acker, 1990, 1991; Wodall, 1998; Rees and Garnsey, 2003; Andersen and Bloksgaard, 2004, 2006; Bloksgaard, 2009). Furthermore, in the Nordic countries - and especially in Denmark - there is a dominant discourse that “equality is already achieved” (Borchorst, 2009). As a consequence of these notions, gender is often seen as having no impact on – and, therefore, not being relevant to - work-life. At the same time, however, statistics show that gender has a high impact on modern work-life; labour markets and work organizations are, to a large extent, structured by gender - women and men work in different sectors, jobs and positions, they receive different pay and are to a different extent absent from the labour market due to family obligations, just to mention some differences. Still, these gendered patterns may often be ’invisible’ for us in our everyday lives in work organizations – 164 gender exists, so to say, as an implicit phenomenon in work-life1. As a consequence, research that focuses on gender in modern work organizations will be “looking for gender, where gender is not explicit” (Korvarjärvi, 2003), and where, in addition, the significance of gender is often denied. This means that researchers face special challenges when researching gender in work organizations. An additional reason for considering the analytical perspectives and methodological approaches carefully when planning studies on gender in organizations, follows from the theoretical framework developed within the research field of ‘gender in organizations’. In the well-known work about ‘gendered organizations’ by Joan Acker, a central point is that gender is being created in processes at several different levels within organizations (Acker, 1990, 2012). Following Acker, Kvande develops a perspective suggesting that there may be an asymmetric relationship between the gendered processes at the different levels (Kvande, 2007); for example there will often be a discrepancy between discourse and practice – that is, what we say we do, and what we actually do. This point is important to take into consideration when planning research projects on gender in organizations and choosing research strategies and methods. This article takes its point of departure from methodological issues and experiences arising from two Danish research projects focusing on the significance of gender in negotiations for wage and parental leave within a number of specific work organizations. It explores some of the methodological complexities and challenges involved when researching gender as a social category of difference, which produces inequalities in organizations understanding themselves as gender- neutral. The article demonstrates and discusses how ‘following specific work place negotiations to a conclusion’ and combining different sociological methods and research strategies are ways of gaining insights into how gender is produced at the work place level, both discursively and in people’s practices. Finally, it discusses how discursive practices produced in interviews may be analyzed as performances of gender, contributing an understanding of some discrepancies between discourse and practice. Researching gender as discourse and practice in organizations – how? Many studies on the significance of gender in work-life are based on interviews. Informants are asked about what consequences gender has or might have in work-life, or gendered meanings are analyzed from informants’ accounts about work-life in general. Several analytical and methodological strategies to capture ’gendered meanings’ in or deconstruct gender in interview material have been developed (e.g. Haavind, 2000; Søndergaard, 1996, 2000; Gunnarsson et al., 2003; Martin, 1992, 2003). At the same time, a central discussion in the methodological literature on how to study gender is the adequacy – or rather the inadequacy – of the interview method in this research field. In the article ‘Is it possible to gain knowledge about gender by interviewing?’ (2010) the Danish researcher Lis Højgaard discusses the danger of producing gender stereotypes when asking about gender in interviews, and how to (try to) avoid this when using interviews as a method in research projects on gender – an issue which Haavind, among others, has also reflected on (Haavind 2000). 1 Ellingsæter and Solheim point out, that gendered structures of meaning and differences are ”taken for granted and therefore remains unreflective, even if they may be visible enough. Paradoxically, they are so obvious that we don’t even notice them” (Ellingsæter and Solheim, 2002: 62). 165 Concurrently, a number of researchers on gender (in organizations) emphasize the importance of studying gender as a social practice – cf. also the development of concepts like ‘doing gender’ and ’gendered practices’ (West and Zimmerman, 1987; Acker, 1990, 1997; Rantalaiho and Heiskanen, 1997; Martin, 1992, 2003; Connell, 1987, 1995; Morgan, 2002; Kvande, 2003; Gunnarsson et al., 2003; Korvajärva, 2003). As mentioned above, central to Acker’s theory about ‘gendered organizations’ is the point that analytically, the construction of gender can be seen as occurring in several different and interacting organizational processes (Acker, 1990, 2012). Following Acker, Kvande argues that ”Gender in organizations is constructed and must be analyzed at different levels and taking into account the fact that gender can have different meanings at these different levels” (Kvande, 2007: 101). For example, there is often a discrepancy between the discursive and practical levels in gender constructions. Hochschild refers to this distinction by using the terms ‘gender ideology’ and ‘gender strategy’. Gender ideologies are women’s and men’s cultural ideas of how to act and prioritize, for example regarding family and work life, whereas gender strategies refer to how women and men really act (Hochschild, 1989: 190-203). The different terms illustrate the difference between what we think women and men do or should do, and what they are actually doing. In continuation hereof, I will argue that, from a methodological perspective, it is important for researchers to get not only the accounts of the people who act in organizations, but also to gain insights into the actual organizational processes and the interactions in which gender is also produced - for example by observing or participating in these (Goffman, 1989; West and Zimmerman, 1987; Hammersley and Adkinson, 1995; Kristiansen and Krogstrup, 1999). As Goffman states: “I don’t give hardly any weight to what people say, but I try to triangulate what they’re saying with events” (Goffman 1989: 131). In the rest of the article I will unfold and discuss different strategies and methods for capturing gender, as both discourse and practice, in organizations - based on experiences from two specific research projects. Two Danish research projects: Gender in wage and parental leave negotiations The methodological reflections and discussions in the article are based on two empirical studies: ‘Når køn forhandler løn’ [Gendered wage negotiations], (Andersen & Bloksgaard, 2004) and Arbejdsliv, forældreskab og køn. Forhandlinger af løn og barsel i tre moderne virksomheder [Work-life, parenthood and gender. Negotiations of wage and parental leave in three modern work places] (Bloksgaard, 2009). Both studies focus on the significance of gender in negotiation processes in modern work-life, and both are based on field work in a number of large, modern work organizations from the private labour market in Denmark. The first study2 focuses on the processes of wage formations in four large Danish companies and the significance of gender therein, especially in individual wage negotiations. The second study includes and expands on the former, by adding a case on negotiations of parental leave in three of the four participating companies. This study also contains transverse analyses of assessment of work participation, absenteeism due to parental tasks and gender in these three work place contexts. The participating companies are characterized by all having implemented individual wage systems and to a varying degree family friendly policies and incentives (for example parental leave incentives). 2 This study I conducted together with a colleague, Pernille Tanggaard Andersen. However, in the interests of consistency of this article, I use ’I’ when discussing the methodological challenges and considerations in relation to both of the studies, which the article draws on. 166 In general, negotiations of wage and parental leave are both understood as gender-neutral and objective processes within work organizations (Andersen and Bloksgaard, 2004; Bloksgaard, 2009). In the Danish labour, market employee work place agreements increasingly take place in individual negotiations between individual employees and their manager (Due and Madsen 2004, 2006). Furthermore, the two different types of negotiations are chosen on the basis of an assumption that gender may be a disadvantage to women with regard to negotiations of wage3, whereas gender- specific expectations and gender-stereotypes may constitute barriers for men with regard to prioritizing family life, in this particular case taking parental leave4. Thus, the two types of negotiations make it possible to gain broad insights into how gendered differences and inequalities are created at the work place level. One overriding aim in the two studies has been to analyze the individual negotiations, with the researcher conducting the research from as close to the negotiating table as possible - to gain insight into ‘the direct meeting’ and the social interaction there. The intention was to gain a broader knowledge than existing interview studies have provided about the complex dynamics which are at play at the work place level, and which contribute to the production of gender inequality in relation to wage and parental leave. To gain insight into how gender is created both discursively and through people’s practices in the work place, a number of specific women’s and men’s negotiations are followed to a conclusion. This research design, I argue, is a methodological strength, taking the point about asymmetry between discourse and practice into account. The fact that a number of specific wage and parental leave negotiations are followed to their conclusion means that the meaning of gender is settled5 in this specific situation, just as the consequences of gender with regard to wage and parental leave are revealed. So, the studies explore women’s and men’s actual choices and actions regarding wage determination and parental leave, and not just speculations on what individuals might have done in an imaginary situation. This research strategy opens a space for analyzing how gender is given meaning during the negotiations, as they happen. The meanings of gender must necessarily be captured and analyzed ‘in the specific context in which gender is given meaning’ (Gherardi, 1994). Accordingly, the studies analyze how gender and gendered inequalities are created in negotiation processes in a number of specific work place contexts. In the process of following specific negotiations to a conclusion, the studies combine and apply several different methods and research strategies: field work, observations, interviews, document analysis, statistics and longitudinal research strategies6. The rest of the article will demonstrate and 3 In Denmark there is a gender pay gap of 18 % (Larsen, 2010). 4 Danish men only take 7,7 % of the complete leave period (Nordic Statistic Yearbook, 2011). 5 At least the meaning which is ascribed to gender is determined in the situation, even if employee and manager not necessarily agree on the offered positions (see also Davies and Harré, 1990). The two studies are based on a social constructivist approach according to which one must study both how gender is constructed or ascribed meaning in relation to ‘ideas of appropriate gender behavior’ and how these constructions acquire social consequences, as they influence women’s and men’s practices in interaction (West and Zimmerman, 1987; Burr, 1995; Berger and Luckmann, 1999). 6 The empirical material consists of observations of 30 individual wage negotiations and 84 interviews with employees, managers and trade union and management representatives, documents regarding wage and parental leave, statistics on the use of parental leave and notes from the field work and from informal observations and conversations in the work place contexts. 167 discuss how these different methods and strategies have provided insights into the production of gender as discourse and practice at work place level. First, however, I will reflect on some of the challenges one might meet and strategies one may use, when as a researcher, you wish to access gender as a social category of difference which creates inequalities in work organizations. Accessing gender in work-life: Gender as an explicit and non-explicit theme As described above, there is a widespread notion in modern work-life and work organizations that gender is not relevant. I have often been met with surprise or even resistance, when I raise gender as a subject in Danish work organizations. Acker describes the resistance in this way: ”People who work with equality problems in employment seem to be faced with an often unspoken resistance that could be articulated as ’What more do women want? They have gotten everything they asked for. It’s boring to talk about this anymore’” (Acker, 1994: 43). These types of attitudes mean that getting access to companies as research fields or cases may imply special challenges when gender is the research subject. In the two studies, access to relevant companies was characterized by a number of particular barriers and challenges. The first hurdle was a more practical one: to find private companies, which had actually implemented a decentralized wage system and individual wage negotiations. The next challenge was to convince the few relevant companies to consent to participating in the study. When I approached the Boards of Directors and gatekeepers in the companies with a request to participate in a study, first on wage negotiations and later parental leave, gender was mentioned explicitly as a theme in the research. In most companies it was necessary to negotiate about the relevance of this as a research topic. To a large extent, I was met by the immediate rejection of the idea that gender should be a cause of differences in their organization – “no, gender doesn’t make a difference…?” was the typical reaction, people being almost surprised that I suggested this. There was also a lack of interest in gender issues or equal pay, or an attitude that “there are no inequalities here”. It was soon obvious that wage is one of the ‘toughest bones of contention’ in terms of employment issues, and it is, therefore, difficult to get access to do research about this. The attempts to gain access were further complicated by the fact that I asked for permission to observe the actual negotiations. This request caused some controversy within some of the Boards of Directors of the companies approached. In the end, however, I managed to get access to the number of companies I wished to conduct studies in7. Having a focus on the significance of gender in work-life does not mean that the researcher should necessarily address gender explicitly with informants. In accessing informants in the two studies, I used a strategy of treating gender as a non-explicit theme. This was partly due to the widespread view within companies that gender is irrelevant in work-life. Besides, if you present yourself as a gender researcher to informants in specific work organizations and address gender directly, you make gender relevant and ascribe certain meanings to gender yourself (Højgaard, 2010). For that reason, I intentionally sought to avoid making gender relevant in the informants’ consciousness, trying to prevent them from seeking to ’do’ gender according to the cultural ideas of “gender appropriate behavior” (West and Zimmerman, 1987) for example in the wage negotiations. I applied a similar strategy in the interviews. Thus, as a rule I did not ask explicitly about gender 7 For further information on and considerations about the type of participating companies in the studies and the possibility for generalization see Andersen and Bloksgaard, 2004; Bloksgaard, 2009. 168 (until the end of the interviews8, see Haavind, 2000; Højgaard, 2010) but about processes of wage and parental leave, ‘the good employee’, ‘the parent’, and other general employment-related topics. I used similar interview guides for men and women, and formulated the questions in the same way, regardless of the gender of the informants, and irrespective of whether I asked about the practices of women or men. This strategy provided me with valuable knowledge about the production and impact of gender in the presumed gender neutral processes of wage and parental leave. The strategy caused no trouble with regard to the questions about wage, whereas it was more difficult in the questions about parental leave. As a researcher, you often automatically feel the need to formulate the questions differently when asking about men’s parental leave practices, than when asking about women’s – for example you may typically ask men if they have taken parental leave, whereas you will most often ask women for how long they took parental leave. As I wanted to examine how presumably gender-neutral categories are ascribed gendered meanings, I used gender-neutral interview guides to ask both women and men if they had taken leave – a formulation which felt very unnatural to the interviewed women and which often resulted in answers beginning with ”Yes, off course I have taken leave…..?” (Danish women are legally bound to taking two weeks of maternity leave after the birth, and as mentioned above, they generally take the main part of the complete leave, as well as the sharable parental leave). This finding before and during the interviews gave me a good understanding of how strongly a connotation of femininity is attached to parental leave practices, which was precisely what I intended to examine by using gender neutral formulations of questions. Thus, already at this stage in the study I found that gender does make a difference in work-life. Wage negotiations: Combining field work, observation and interviews9 In the study on wage negotiations, the analysis is based primarily on field work in the participating companies, observations and interviews with people from ‘both sides’ in the negotiations (employees and managers) as well as trade union and management representatives. Field work is applied as a method to achieve a broad knowledge about the social fields that the companies constitute in the study, and to gain insight into the wage negotiation process and the production of meaning in the different work place contexts (Maaløe, 1996; Wadel, 2001; Järvinen and Mik-Meyer, 2005a). Observation method gives researchers opportunity to study direct interactions and negotiations between people, thereby providing access to studying not only people’s discourses but also their practices (Järvinen and Mik-Meyer, 2005b). Larsen and Haldrup argue that “compared with interviews, observations better capture the […] quality of practices because they focus on immediate physical doings and interactions rather than retrospective and reflexive talk about how and why such performances took place, and what they meant” (2010: 38-39). Therefore, observation is well-suited as supplement to interviews, as “observation studies bring us one step closer to the analyzed situation, than the interview is able to” (Dyrberg et al., 2000: 331, my translation). In the 8 The researcher, Haavind points out, should not ”bring up the theme [gender] by asking about the person’s perceptions about the significance of gender. […] When it is seen as a methodological advantage not to ask about what gender means it is because we look for the multiple in gendered meanings […] The informants’ thoughts about what gender might mean as a stereotype, as an ideal or as a reference for his own self may be added later, and preferably in the end of the interview” (Haavind, 2000: 20). 9 The combination of different sociological methods in the study of wage negotiations and the knowledge this research design provided is more thoroughly discussed in Andersen et al. 2009. 169 study, both ‘informal’ participant observation – spending longer time periods in the companies, participating in and observing ‘every day organizational life’ – and ‘formal’ observation of the individual wage negotiations were conducted. Observing the wage negotiation situations, I sat at the back of the room, usually behind the employee so that he or she could not see me, in an effort to try to minimize the effect of my presence10. Interviews were conducted with employees and managers after the wage negotiations, and were related to the assessment and wage determination which had just taken place. Thus, the main purpose with interviewing was to get the actors’ own explanations and interpretations linked to the observations, and to gain insight into how gendered meanings are constructed discursively in interactions (see below). The research strategy of following a number of specific wage negotiations in the work place and applying different methods has been crucial for the results and the production of new knowledge about the formation of individual wage in the Danish labour market (and the production of unequal pay). Instead of interviewing managers about how they would in principle distribute wage among their employees and the criteria they use to reward employees, it was possible to observe and ask questions about the individual managers’ actual distribution of an actual payroll among a group of specific employees in an ongoing wage process, and to hear the managers’ explanations thereof. Thus, it was possible to confront the explanations used by the managers in the negotiation situation towards the employee, and the explanations the managers revealed in the subsequent interviews. The combination of field work, interviews and observation made it possible to study both ’the sayings’ and ’the doings’ of the managers (and the employees) – discourse and practice. The combination of methods documented clearly that these two levels are not always consistent – a result which emphasizes the importance of using several methods to gain a broad insight into the significance of gender in wage formations and assessment of employees at work place level. Thus, combining field work, observations and interviews provided a broader insight into the criteria used by managers for distributing the payroll than could have been achieved by interviews alone. For example, the field work and the analyses of wage documents show that family needs and family friendly incentives are not thought of as a part of the reward systems and the reward criteria. However, the observations show that family needs and gendered ideas of parenthood do sometimes have consequences for pay. The following sequence is from a wage negotiation between Birgit and her somewhat younger male manager: Manager: “What do you expect to get” Birgit: “10 DKR [1,35 EUR] extra per hour” Manager: “Is that what you expect? What are your arguments for that?“ Birgit: “I think I cover a large area and I’m responsible…Are you surprised?” Manager: “Yes I am. ....and to be honest, 10 DKR is way too much, it is not possible for me to give you that much. And don’t misunderstand me, but in your job function 14 DKR more per hour than minimum wage is too much.…I’ll also have to say that because of your working hours I simply cannot justify giving you such a large pay raise, so that’s also a part of assessing the wage….” 10 However, an elimination of the ‘researcher effect’ (Kristiansen and Krogstrup, 1999) is never possible in social research; it is, therefore, important to always reflect on the consequences hereof on the research results (as I discuss below). 170 On the basis of the observation of the wage negotiation, it was possible for me to ask the manager about his arguments in the follow-up interview. He explained that in the day-to-day work in the department, Birgit did not live up to his expectations of ’a good employee’, because, in his opinion, she did not prioritize the work place high enough compared to her family needs (she had two children) – a fact that he also refers to in the negotiation situation by saying that he cannot justify a higher pay “because of her working hours” (she works 37 hours per week, which is the general norm, but the working hours are ‘family friendly’, from 7.00-15.00). The combination of interviews and observation shows that family needs can be a factor, which may have negative consequences for wage. The research design - following specific negotiations to a conclusion - and the combination of methods showed that wage formation, which is assumed to take place on the basis of a number of objective and gender-neutral criteria, is happening as something relational, where ideas of gender, competence and parenthood may have an impact. So, the applied design and methods provided new knowledge of how wage formation takes place at the work place level, and how gendered inequalities are produced in this process. Men’s parental leave negotiations: A longitudinal research strategy In the study on parental leave negotiations, the analysis is based on field work in the participating companies, interviews with people from ‘both sides’ in the negotiations (employees and managers) and trade union and management representatives, as well as documents regarding parental leave. Additionally, statistics on men’s and women’s use of parental leave in the participating companies were analyzed. Statistics were used to gain insight into practice, as the field work revealed that negotiations of parental leave between employee and manager in the work place are very informal, and often take place in an unplanned fashion11 –therefore, these are not easily studied by observation methods. The combination of statistics and interviews identified a discrepancy between the ideals of men and their actual leave practice (see Bloksgaard, 2009, 2014), which other research has also identified in the research field of ‘men and parental leave’ (e.g. Bekkengen, 2002; Brandth and Kvande, 2003). In a Swedish study, Haas et al. conclude that men’s use of parental leave is significantly affected by the organizational culture of their work place (Haas et al., 2002), and they suggest that further research on determinants of father’s use of parental leave should focus on the relations between the employed parent, his/her employers and work groups, and ideally follow him/her longitudinally to get a better understanding of how negotiations in the work place affect fathers’ leave use. In the following, I will demonstrate and discuss how applying a longitudinal research strategy, and thus, following the actual sequences of the negotiation process between male employees and managers in the study provides an understanding of the discrepancy between men’s ideals and leave practices, and of how the unequal distribution of parental leave between women and men is being reproduced in the work place. I will illustrate this using the empirical case of ‘Ulrik’: Ulrik is a male manager in a retail company whose parental leave negotiations I followed longitudinally. I interviewed Ulrik three times over one year. In the first interview, approximately one month before Ulrik was going to become a father, he explained to me that he planned to take the two weeks of paternity leave, and also one or 11 The field work also uncovered that in Denmark typically only men will have to negotiate parental leave in the work place, whereas women get leave ‘automatically’ (Bloksgaard, 2009). 171 maybe two months of the parental leave12 at the end of the leave period, so that his wife could get back to her job. Ulrik is the first man in his work place who wants to take more leave than the two weeks of paternity leave, that is, to take some of the parental leave, and Ulrik tells me that “HR said: ‘We’ll have to figure out how we do that…’ but I think they’ll accept it (…) 80 percent of the employees being women here, I think they’ll be more generous about parental leave”. The second time I interviewed Ulrik, he had become a father to a little girl one month earlier, and he had taken his two weeks of paternity leave. In this interview he still expressed an ideal about ’the present father’ (Brandth and Kvande, 2003), as one who takes leave: ”No matter if there is fire and brimstone [at the work place], I would take leave, I had decided that for sure, it was so important to me. Cause I will not be that father of whom you say ’this guy, he’s always working’ – damn it, I want to be there. It might sound like an old romantic speech that will not actually happen, but hell, I mean it!” Ulrik was promoted at work around the same time he became a father. Later during this second interview, I asked him about the one or two months of parental leave that he had indicated he would take in the first interview. He answered: ”I’m totally new in this position, so now we’ll have to see about that … There’s a new managing director who I report directly to ….and I haven’t discussed it with him yet, there have been so many other things that have been more important …. But of course I’ll suggest it and hear about the possibilities and then try to …to get that period. […] But the last month of the leave period that’s in December, and that’s the busiest time of the year being in retail, so I cannot take time off there, that’s for sure!” Ulrik indicated that he would still like to take a month of parental leave at the end of the paternity leave period, but that it would be problematic for him to be absent from the work place in December (when I asked female managers in the same company if they tried to adjust their leave period to busy periods at work, they totally rejected this idea, Bloksgaard, 2009). The third time I interviewed Ulrik, his daughter was almost one year old. In this interview he explained that for personal and family reasons, and because of his new position at work, he ended up not taking any parental leave. Ulrik’s actual sequence of negotiation and the interviews during this process with Ulrik, his manager and the HR-director in the company, show that Ulrik ended up not converting his original fatherhood ideals into parental leave practice. The longitudinal research strategy provides insights into how men are adapting to gendered cultures in the work place (and society), which make it unlikely for men to take parental leave. It explains how, at the same time, there might be goodwill among men to take parental leave at the 12 Under Danish legislation, two weeks of paternity leave is reserved for the father right after the birth of the child. There is 32 weeks of parental leave for each of the parents with a right to unemployment benefit for 32 weeks in total – in practice this means that 32 weeks is the possible leave length that the parents will have to share. However, Danish mothers take the majority of the parental leave (for more details about the Danish legislation on leave see Bloksgaard, 2009, 2014). 172 discursive level, while very few men take parental leave in practice. Following specific negotiations to a concrete conclusion using a longitudinal research strategy provides insight into why this is so - showing how men’s use of parental leave is influenced by several concurrent factors, and continuous and complex negotiations both in the work place and in family life (Bloksgaard 2009, see also Bloksgaard 2011, 2014). Thus, it helps to explain the low number of Danish men taking parental leave (Nordic Statistic Yearbook 2011). The categorizations which take place in relation to gender in the work place mean that men may possibly be sanctioned if they behave in ‘not gender- appropriate ways’ in the work place, for example taking parental leave. The example of Ulrik indicates, however, that men may perceive it as a choice of their own, when they end up not taking any of the parental leave. Seemingly, tendencies towards, and discourses of, individualization and ’freedom of choice’ in Denmark blur the structural aspects in the production of gendered differences and inequalities in organizations (for an elaboration of these points, see Bloksgaard 2009, 2014). Analyzing performances of masculinity in the interview situation As earlier mentioned, gender may be understood as something men and women ‘do’ or practice in interaction - but in relation to cultural ideals of what masculinity and femininity ‘is’ (West and Zimmerman, 1987; Connell, 1995; Morgan, 2002; Kvande, 1995). Accordingly, men and women develop their individual masculine or feminine identities under influence of dominant gender ideals in society and at work place level. Thus, men’s fatherhood and leave practices and masculine identities may be seen as the outcome of negotiation processes, for example in the work place (Brandth and Kvande, 2003). According to Järvinen (2005), the interview situation is an example of interaction where people’s presentations of self (Goffman, 1974) or performances of (gendered) identities (Butler, 1990) become particularly visible. Interviewees seek to present themselves as reasonable and sense-making human beings – ‘a competent manager’, or ’a good father’, for example. Thus, the articulations of an interviewee can be seen as attempts to negotiate as meaningful a social identity as possible in the interview situation. These performances of identity should not be seen as a problem, but rather as important data, as they offer insight into how identity is constituted in interaction (Hydén, 2001; Järvinen, 2005; Søndergaard and Staunæs, 2005; see also Jensen this issue). In this last section, I will discuss how the interviews with the male employees and managers in the study on parental leave can be analyzed as performances of masculinity, and thereby provide insight into how gender (masculinity) is produced in interaction; here how men negotiate masculine identities and parental leave practices in parental leave negotiations in the work place. Following the understanding presented above, the interview situation may be understood as one forum for men’s presentation of self, one arena (out of several) for ‘identity work’, and it thereby presents “an opportunity to study masculinities in the making” (Allen, 2005). Taking the same perspective as Järvinen, Schwalbe and Wolkomir, I concur with the argument that “…it would be a loss, we think, to treat the identity work men do in interviews as noise that must be filtered out to get the real data. Instead, we suggest that this identity work be treated as data – the first step being to see it, in the form of how and when men try to signify masculinity in interviews” (Schwalbe and Wolkomir, 2001: 92). So being aware of and analyzing interview material from the male interviewees - not only the content but also with an eye to the identity and positioning work going on within the interview - may provide further insight into the production of gender in interactions. 173 By default, an interview situation must be seen as a relation of power – where the interviewer is in command by setting the agenda for the interview13. The interview situation is an opportunity for men to signify masculinity, but at the same time men especially may experience the interview situation as threatening: “This threat may be heightened if the interviewer is interested in gender since this increases the salience of the participant’s identity as a man. […] Reactions to such threats are more likely to be matters of self-presentational habit than conscious strategy” (Schwalbe and Wolkomir 2001: 91-92). Applying these thoughts to my interviews with the male employees and managers about parental leave, forms the basis for further reflections on the discrepancy between men’s ideals and leave practices. The fatherhood ideal as one who takes leave expressed by Ulrik and several other men I interviewed, may be interpreted as them delivering performances of ‘the present father’ to ‘make sense’ to me, the researcher, in the interview situation. Given that the interviewees’ leave practices as men was the primary focus of the interviews, the male interviewees may have had a heightened awareness of how they were performing gender in the interview situation. In Denmark, and the Nordic countries in general, there is a strong equality ideal, including the ideal of ‘the present father’, which has gained strong legitimacy during the last decade (Brandth and Kvande, 2003; Bekkengen, 2002). Therefore, it may be difficult for male interviewees to express opinions about not taking parental leave, and risk being seen as ‘not a modern and politically correct man’. This also means that in other situations, contexts and relations, and towards people representing other positions than the ones the male interviewees positioned me in, (some of) the men may express other fatherhood ideals. However, as mentioned above, the idea that people are performing different identities in different social relations is central to the understandings of gender, on which this article is based (Goffman, 1974; West and Zimmerman, 1987; Connell, 1987, 1995; Kvande, 1995). The recognition of the impact of the relation and the interaction between researcher and the researched emphasizes the need to also be reflective about the impact of the social categories that both researcher and researched bring into the research relations, for example the interview situation (Staunæs and Søndergaard, 2005). As stated by Williams and Heikes: “Interview, like any other interaction, always take place in a gendered context – the context of either gender similarity or gender difference (…) gender is constantly ‘there’. The question is therefore not if gender makes a difference but, rather, how gender matters?” (Williams and Heikes, 1993: 282). Whereas class, to some extent, may be ‘neutralized’ or down-played by the researcher in the interview situation (see Faber, this issue), gender is a social category which cannot be treated in this way. When two people meet, one of the first things they notice about each other is the sex of the other. This categorization is accompanied by certain expectations about how the other will behave in the interview situation, just as the categorization will also imply that the individual expects a certain behavior from him/herself in the relation (Berger and Luckmann, 1999; West and Zimmerman, 1987). So, gender is a category of difference in the research process which becomes a premise for producing empirical material. Accordingly, it is relevant to reflect on how my female gender may have influenced the interviewed men’s performances of masculinity in the interview situation and, thereby, also the research results. Several studies have found that male interviewees present 13 The interviewer is sometimes also better positioned with regard to education level, job position, etc. See Jensen as well as Faber in this special issue. 174 themselves differently to researchers of a different gender (McKee and O’Brien, 1983; Williams and Heikes, 1993; Sallee and Harris, 2011). Sallee and Harris document that American men interviewed by female researchers present themselves as more thoughtful and rejecting of stereotypical gender roles than when interviewed by male researchers (Sallee and Harris, 2011). On the basis of a Swedish study, Nordberg emphasizes that when women interview men, aspects of equality are more often actualized. She finds that men very actively relate to and use the ideal of equality when presenting themselves to a female interviewer – “in this way he may present himself as a socially desirable and equal man to the female interviewer” (Nordberg, 1999: 68, my translation). Applied to my study on men and parental leave, this suggests that the fatherhood ideals of taking leave expressed by several of the male interviewees may be part of performing a ‘socially desirable (male) identity’ towards me, the female researcher. At the time of the interviews I was not only a young (33 year old) woman, but I was also pregnant with my second child – a fact that enhanced my position as a female subject. This may have made it difficult for the men to express a negative attitude to taking parental leave in the interviews. Such an analysis of the gendered performances in the research relation and of the impact on the research results (Haraway, 1988) may explain some of the difference in discourse and practice in the research field of ‘men and parental leave’. It is important to reflect on how performances of social categories are produced in the research relations and take these into account when analyzing our empirical data – these being valuable data of the production of these categories of difference, here gender14. Conclusion This article has addressed some challenges and complexities when researching gender in work organizations. In modern work-life, organizations will often be seen as, and understand themselves to be, gender-neutral entities. Accordingly, research projects on gender in work-life will be studying gender as ‘a non-explicit phenomenon’ in organizations, and be up against notions of gender as not being relevant in organizations and the things going on therein. On a theoretical and analytical level, the construction of gender in organizations can be understood as taking place in different and interacting processes at different levels: for example in discourses, and in men’s and women’s practices and interactions. These conditions constitute the premise for doing research, which need to be considered and dealt with methodologically when planning and conducting studies on gender in organizations – for example, these make it difficult to capture the construction of gendered differences and inequalities in organizations using interviews as the sole research method, which is the case in many studies of gender in organizations. 14 When reflecting on the meaning of gender in the research relation, though, it is important not to forget the intersection with other social categories at play (Valentine, 2002). Taking an intersectional perspective it is important to reflect on how not only gender, but also other social categories – class, race, ethnicity, generation, sexuality or disability – are woven together and gain significance in different ways, sometimes downplaying the significance of gender. Such a perspective has not been applied in this article but see, for example, Jensen in this special issue and Bloksgaard, 2014. In the field of organizational studies, still very few studies acknowledge the importance of intersectionality and reflect on the methodological challenges of studying intersecting social categories in organizations (exceptions are e.g. Acker, 2006, 2012; Holvino, 2010; Healy et al. 2011; Mulnari and Selberg, 2011). 175 The article has argued that ‘following men’s and women’s specific negotiations to a conclusion’ in organizations, and combining different sociological methods and research strategies are ways of gaining valuable insights into the production of gender as discourse and practice in the work place. The applied research strategies and methods have provided insight into both the ascribed meanings of gender and the social consequences of gender with regard to wage and parental leave – uncovering examples of how ‘gender is continuously said and done’ in the work place. Thus, gender-neutrality, as something that is taken for granted in organizational contexts, does not mean that gender is not an important factor in work-life – not even in Denmark, one of the Scandinavian countries which are often considered to be ‘world champions of equality’. This points to the importance of understanding and studying gender differences and inequalities as something which is produced in everyday interactions in the work place, also in modern work-life. Finally, looking closer at the discursive practices in the interviews with my male informants, the article has demonstrated how the interview situation can be understood and analyzed as a central context for performances and negotiations of identity, such as gender. The article has illustrated that reflecting on not only what is said in interview relations but also by whom, to whom and why it is said – taking the perspective of positionality into consideration - may provide further insight into the ways gender is produced in interactions in the work place, and, thereby, help explain research results, such as the identified discrepancies between gender as discourse and practice. Thus, the article argues that reflexivity and methodological plurality are important when studying gender in organizations. References Acker, J. (1990). Hierarchies, Jobs, Bodies: A Theory of Gendered Organizations. Gender and Society, 4, 2, 139-158. Acker, J. (1991). Thinking about Wages: the Gendered Wage Gap in Swedish Banks. Gender and Society, 5, 3, 390-407. Acker, J. (1994). Woman, Families and Public Policy in Sweden. In E. Chow and C.W. Berheide (eds.): Women, the Family and Policy. A Global Perspective. New York: State University of New York Press. Acker, J. (1997). Foreword. In L. Rantalaiho and T. Heiskanan (eds.): Gendered Practices in Working Life. London: MacMillan Press. Acker, J. (2006). Inequality regimes: gender, class and race in organizations. Gender and Society, 220, 4, 441-464. Acker, J. (2012). Gendered organizations and intersectionality: problems and possibilities. Equality, Diversity and Inclusion: An International Journal, 31, 3, 214-224. Allen, L. (2005). Managing masculinity: Young men’s identity work in focus groups. Qualitative Research, 5, 1, 35-57. 176 Andersen, P.T. and L. Bloksgaard (2004). Når køn forhandler løn. LO-dokumentation, nr. 1/2004. København: LO. Andersen, P.T., L. Bloksgaard and A.D. Christensen (2009). Magten ved forhandlingsbordet – et mikrosociologisk blik på arbejdsliv. Tidsskrift for Arbejdsliv, nr. 4/2009, 69-83. Berger, P. and P. Luckmann [1966] (1999). Den samfundsskabte virkelighed. København: Lindhardt and Ringhof. Bloksgaard, L. (2009). Arbejdsliv, forældreskab og køn. Forhandlinger af løn og barsel i tre moderne virksomheder. Unpublished PhD dissertation. Aalborg University. Bloksgaard, L. (2011). Mænd, barselsrettigheder og brug af barsel. En interviewundersøgelse blandt mandlige medlemmer i 3F. København: 3F. Bloksgaard, L. (2014). Fathers negotiating parental leave practices and fatherhood ideals in three Danish work places. In G.B. Eydal and T. Rostgaard (eds.): Nordic fatherhoods – Policies, Practices and Discourses of Contemporary Fatherhood. Polity Press (forthcoming). Borchorst, A. (2009). Scandinavian gender equality: Competing discourses and paradoxes. FREIA Research Series, 69, 2009. Bekkengen, L. (2002). Man får välja – om föräldraskap och föraldraledighet i arbetsliv och familjeliv. Malmö: Liber AB. Brandth, B. and E. Kvande (2003). Fleksible fedre. Maskulinitet, arbeid, velferdsstat. Oslo: Universitetsforlaget. Burr, V. (1995). An Introduction to Social Constructionism. London: Routledge. Butler, J. (1990). Gender Trouble. Feminism and the Subversion of Identity. London: Routledge. Connell, R.W. (1987). Gender and Power. Cambridge: Polity Press. Connell, R.W. (1995). Masculinities. Berkeley: University California Press. Davies, B. and R. Harré (1990). Positioning: The Discursive Production of Selves. Journal for the Theory of Social Behaviour, 20, 1, 43-63. Due, J. and J.S. Madsen (2004). Politisk indblanding truer den danske model. From www.lo.dk. Due, J. and J.S. Madsen (2006). Fra storkonflikt til barselsfond. Den danske model under afvikling eller fornyelse. København: Jurist- og Økonomiforbundet. Dyrberg, T.B., A. D. Hansen and J. Torfing (2000). Diskursteorien på arbejde. Roskilde: Roskilde Universitetsforlag. 177 Edley, N. and M. Wetherell (1999). Imagined futures: Young men’s talk about fatherhood and domestic life. British Journal of Social Psychology, 38, 181-194. Ellingsæter, A.L. and J. Solheim (2002)(ed.). Den usynlige hånd. Kjønnsmagt og moderne arbejdsliv. Oslo: Gyldendals Norsk Forlag. Eriksson, K. and M. Eriksson (2003). “Gender doesn’t matter but a gender-mix is good” – Doing Gender in Two Professional Contexts. In E. Gunnarsson et al. (eds.): Where have all the structures gone? Doing Gender in Organisations, Examples from Finland, Norway and Sweden. Stockholm: Center for Woman’s Studies, Stockholm University. Gherardi, S. (1994). The gender we think, the gender we do in our everyday organizational lives. Human Relations, 47, 6, pp. 591-610. Goffman, E. (1974). The Presentation of Self in Everyday Life. London: Penguin Books. Goffman, E. (1989). On Fieldwork. Journal of Contemporary Ethnography, 18, 123-132. Gunnarsson, E. et al. (2003b) (eds.). Where have all the Structures gone? Doing Gender in Organisations, Examples from Finland, Norway and Sweden. Stockholm: Center for Woman’s Studies, Stockholm University. Haldrup, M. and J. Larsen (2010). Tourism, Performance and the Everyday. London: Routledge. Hammersley, M. and P. Adkinson (1995). Ethnography: Principles in Practice. London: Routledge. Healey, G., H. Bradley and C. Forson (2011). Intersectional Sensibilities in Analysing Inequality Regimes in Public Sector Organizations. Gender, Work and Organization, 18, 5, 467-487. Haraway, D.J. (1988). Situated knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies, 14, 575-599. Hochschild, A. (1989). The Second Shift. New York: HarperCollins Publishers. Holgersson, C. et al. (2004). Det ordner sig. Teorier om organisation og køn. Lund: Samfundslitteratur. Holvino, E. (2011). Intersections: The Simultaneity of Race, Gender and Class in Organization Studies. Gender, Work and Organization, 17, 3, 248-277. Hydén, M. (2001). Forskningsintervjun som relationell praktik. In H. Haavind (ed.) Kjønn og fortolkende metode. Oslo: Gyldendal Norsk Forlag AS. Højgaard, L. (2008). Køn er ikke relevant her. Den vertikale kønssegregering på det danske arbejdsmarked. In R. Emerek and H. Holt (eds.)(2008): Lige muligheder - frie valg? Om det kønsopdelte arbejdsmarked gennem et årti. København: SFI - Det Nationale Forskningscenter for Velfærd. 178 Højgaard, L. (2010). Kan man interviewe sig til viden – om køn?. Dansk Sociologi, 1/21, 2010, pp. 9- 26. Haas, L., K. Allard and P. Hwang (2002). The impact of organizational culture on men’s use of parental leave in Sweden. Community, Work and Family, 5, 3, 319-342. Haavind, H. (2000)(ed.). Kjønn og fortolkende metode. Oslo: Gyldendal Norsk Forlag AS. Järvinen, M. (2005). Interview i en interaktionistisk begrebsramme. In M. Järvinen and N. Mik- Meyer (red.): Kvalitative metoder i et interaktionistisk perspektiv. København: Hans Reitzels Forlag. Järvinen, M. and N. Mik-Meyer (2005a). Indledning: Kvalitative metoder i et interaktionistisk perspektiv. In M. Järvinen and N. Mik-Meyer (eds.): Kvalitative metoder i et interaktionistisk perspektiv. København: Hans Reitzels Forlag. Järvinen, M. and Mik-Meyer, N. (2005b). Observationer i en interaktionistisk begrebsramme. In M. Järvinen and N. Mik-Meyer (eds.): Kvalitative metoder i et interaktionistisk perspektiv. København: Hans Reitzels Forlag. Korvajärvi, P. (2003). ‘Doing gender’ – Theoretical and Methodological Considerations”. In E. Gunnarsson et al. (eds.): Where have all the structures gone? Doing Gender in Organisations, Examples from Finland, Norway and Sweden. Stockholm: Center for Woman’s Studies, Stockholm University. Kristiansen, S. and H.K. Krogstrup (1999). Deltagende observation. Hans Reitzels Forlag. Kvande, E. (1995). Forståelser av kjønn og organisasjon. Sociologisk Tidsskrift, 4, 285-300. Kvande, E. (2003). Doing Gender in Organizations – Theoretical Possibilities and Limitations. In E. Gunnarsson et al. (eds.): Where have all the Structures gone? Doing Gender in Organisations, Examples from Finland, Norway and Sweden. Stockholm: Center for Woman’s Studies, Stockholm University. Kvande, E. (2007). Doing Gender in Flexible Organizations. Bergen: Fakbokforlaget. Larsen, M. (2010). Lønforskelle for kvinder og mænd i 2007. Analyser for Lønkommissionen, København: SFI. Martin, Y. (1992). Gender, interaction and inequality in organizations. In C.L. Ridgeway (ed.) Gender, interaction and inequality in organizations. New York: Springer-Verlag. Martin, Y. (2003). ’Said and done’ versus ‘saying and doing’. Gendering Practices, Practicing Gender at Work. Gender and Society, 17, 3, 342-366. McKee, L. and M. O’Brien (1983). Interviewing men: “Taking Gender Seriously”. In E. Gamarnikow, D. Morgan, J. Parvis and D. Taylorson (eds.): The Public and the Private. London: Heineman, 147-61. 179 Morgan, D. (2002). Epilogue. In B. Hobsen (ed.) Making Men into Fathers. Cambridge: University Press. Mulinari, P. and Rebecca S. (eds.) (2011). Arbete: intersektionella perspektiv. Stockholm: Gleerups. Nordberg, M. (1999). Synd att du inte kan hänga med in i duschen… - om könstillhörighetens betydelse vid intervjuer och fältstudier. Nord Nytt 75/76. Maaløe, E. (1996). Case-studier af og om mennesker i organisationer. København: Akademisk Forlag A/S. Nordic Statistic Yearbook 2011. Rantalaiho, L. and T. Heiskanen (eds.) (1997). Gendered Practices in Working Life. London: MACMILLAN Press. Rees, B. and E. Garnsey (2003). Analysing Competence: Gender and Identity at Work. Gender, Work and Organization, 10, 5, 551-578. Sallee, M.W. and F. Harris (2011). Gender performance in qualitative studies of masculinities. Qualitative Research, 11, 4, 409-429. Schwalbe, M. and M. Wolkomir (2001). The masculine self as problem and resource in interview studies of men. Men and Masculinities 4 (1): 90-103. Smithson, J. and E.H. Stokoe (2005). Discourses of Work-Life Balance: Negotiating ’Genderblind’ Terms in Organizations. Gender, Work and Organization, 12, 2, 147-168. Staunæs, D. and D.M. Søndergaard (2005). Interview i en tangotid. In M. Järvinen and N. Mik- Meyer (eds.): Kvalitative metoder i et interaktionistisk perspektiv. København: Hans Reitzels Forlag. Søndergaard, D.M. (1996). Tegnet på kroppen. Københavns Universitet: Museum Tusculanums Forlag. Søndergaard, D.M. (2000). Destabiliserende diskursanalyse: veje ind i poststrukturalistisk inspireret forskning. In H. Haavind (ed.): Kjønn og fortolkende metode. Oslo: Gyldendal Norsk Forlag AS. Valentine, G. (2002). People Like Us: Negotiating Sameness and Difference in the Research Process. In P. Moss (ed.) Feminist Geography in Practice: Research and Methods. Oxford: Blackwell Publishers. Wadel, C. (1991). Feltarbeid i egen kultur. Flekkefjord: Seek A/S. West, C. and D.H. Zimmerman (1987). Doing Gender. Gender and Society, 1, 2, 125-151. 180 Williams, C.L. and Joel H. (1993). The Importance of Researcher’s Gender in in-depth Interview: Evidence from Two Case Studies of Male Nurses. Gender and Society, 7, 2, 280-291. Wodall, J. (1998). Human Resource Management and Women: the Vision of the Gender-blind? In B. Towers (ed.): The Handbook of Resource Management. Oxford: Blackwell Publichers Inc. Author Lotte Bloksgaard is Assistant Professor in the Dept. of Culture and Global Studies, Aalborg University, Denmark. Email: bloksgaard@cgs.aau.dk