“What about the child issue?” Group negotiations of gender and parenthood contracts in recruitment situations

ABSTRACT Men and women becoming parents are supposed to have equal opportunities in working life. However, inequality and discrimination are not easily avoided. The aim of this study was to investigate how argumentation related to parenthood and careers takes form in group discussions of a fictional recruitment situation, and how stereotypes such as competence and warmth are manifested in such discussions. Thirty-five ad-hoc groups of university students were asked to make a choice between three candidates for a consultancy position. The first two candidates were a man and a woman, while the third alternately was described either as a man or a woman described as having a newborn child. Parenthood was sometimes seen as reducing competence, but it was more often viewed as adding to competence. Parenthood was also considered to add warmth to the organization. Interestingly, all groups avoided relating the parenthood issue to gender. Three conversation patterns were found, differing in the amount of elaboration of the topic of parenthood and work. It was concluded that the most elaborated kind seems to foster a situation where implicit norms are made visible more easily.


Introduction
It is a truism that women's careers risk being challenged by several obstacles. Many of them originate in the asymmetrical roles and responsibilities that men and women traditionally have had in child rearing, where men to a greater extent have had the role of breadwinner and women the role of caregiver (Gaunt, 2013). Many women tend to slow down their career path (Brown, 2010), taking leave for family reasons to a much greater extent than fathers do. In Sweden, mothers still use the major part of parental leave (74% compared with 26% among fathers; SCB, 2016). This may have a negative impact on performance ratings at work and salary increases (Judiesch & Lyness, 1999). On the other hand, women may choose to postpone raising a family so as not to jeopardize their career opportunities (Ranson, 1998). For example, in Sweden, the mean age of first-time mothers is 29 years (SCB, 2013). However, this certainly does not mean that women are less qualified than men, rather that women in roles traditionally predestined for men meet with prejudice and discrimination (Campbell, 2012;Teigen, 1999), especially in combination with parenthood (Cuddy, Fiske, & Glick, 2004;Guillaume & Pochic, 2007;Judiesch & Lyness, 1999). As fathers are traditionally seen as breadwinners, this combination is expected, while mothers, supposed to be caregivers in the first instance, are expected to perform worse at work. Such views are part of a broader framework of sexist ideologies.

Gender ideology and parenthood
Sexism does not necessarily constitute solely negative attitudes towards women. As men and women are mutually interdependent in society, it is impossible to have completely negative attitudes towards the other group (Glick & Fiske, 1996). Moreover, despite the fact that men are perceived as having higher status and greater power, many traditional female traits, such as nurturance, caring, and so on, are evaluated positively (Eagly & Mladinic, 1994). So, in addition to hostile sexism (HS), which is directed at women transgressing traditional gender roles, there is a benevolent form, paternalizing and belittling of women, rewarding women staying within traditional roles (Glick & Fiske, 1996), thus forming an ambivalent gender ideology that supports the status quo for women (Fiske, 2013). This ambivalent dynamic has been found to be quite universal (Glick et al., 2000) and reflects the distribution of societal power and interdependence between men and women in society (Rudman & Glick, 2008). The more uneven the distribution of power between men and women, and the more interdependence between men and women, the higher the levels of hostile and benevolent sexism among both men and women. On the other hand, the more even the distribution of power between men and women and the less interdependence between men and women, the lower the levels of sexist attitudes that are expected. In particular, benevolent sexism (BS) is supposed to be low under such circumstances, as "perhaps people who are highly egalitarian are more likely to recognize BS as a form of sexism and reject it along with HS" (Glick et al., 2000, p. 772), as a core element of measures taken to enhance gender equality is for women to be less dependent on men, for example by societal measures for childcare, subsidized parental leave, and so on (Melhuish & Petrogiannis, 2006;Thévenon, 2011). A study in Sweden, one of the most gender-equal countries in the world (World Economic Forum, 2016), supports this notion, as lower levels of BS compared with HS were found (Zakrisson, Anderzén, Lenell, & Sandelin, 2012).
Ideologies such as sexism are seldom expressed explicitly. Most often, they are subtle and applied quite automatically (Fiske, 2013). They appear when we effortlessly make judgements or make decisions about what we think is something completely different, unaware that we are applying stereotypes. According to the stereotype content model (Cuddy, Fiske, & Glick, 2008), people are generally perceived and evaluated along two main dimensionswarmth and competencebased on two questions that we want answered when we meet new people: (1) whether their intentions good (i.e. whether they are warm), and (2) whether they can realize their intentions (i.e. whether they are competent). According to the model, perception of the competition for societal resources renders stereotypes, such that allies and members of our in-group are viewed as warm, friendly, and sincere, while out-group members are viewed as cold, unpleasant, and untrustworthy. Perceived status in others predicts competencerelated stereotypes. Hence, we see rich people as competent, skilled, and self-confident, and we see poor people as incompetent, ignorant, and insecure (Fiske & Taylor, 2013). These two dimensions are evidently at work regarding how men and women, fathers and mothers, are perceived. Breadwinners need to be competent in order to be able to support their family, while caregivers are supposed to express warmth towards their offspring. Cuddy et al. (2004) tested this proposition in an experiment where participants rated fictitious candidates in a recruitment scenario on characteristics related to competence and warmth. They rated three candidates, of which the third one was alternatively described as either a man or a woman. In half of the cases, this candidate was also described as being a parent, all other things being equal between the candidates. The results revealed that while a man and a woman without children, together with a man having a child, were seen as equally competent, the woman described as a mother was regarded as less competent. Furthermore, the man described as a father gained in warmth. These findings were later replicated by Heilman and Okimoto (2008). In addition, they found that parents (regardless of gender) were anticipated to have lower job commitment and achievement strivings and to express less dependability. Furthermore, such stereotyping has also been found to be related to ambivalent sexist ideologies, with negative correlations between hostile sexism and competence and warmth for a female transgressor, and positive correlations between benevolent sexism and competence and warmth for a female non-transgressor (Gaunt, 2013), indicating that such stereotyping is part of a sexism ideology.
Thus, it seems that parenthood influences how people are perceived. Men tend to win, and women tend to lose when it comes to warmth and competence. The combination of parenthood with gender challenges the traditional gender ideology where men and women are supposed to hold different roles in the reproduction cycle. Traits stereotypically associated with men and women become problematic, especially regarding women seeking a career (competence) at the same time as being a mother (warmth), for a person having to make job-related decisions (recruitment, career opportunities, salaries). How is it possible to be both, especially as it is assumed that she must take the responsibility for the family at the same time? For a male in the same position, it is not problematic. Becoming a father adds warmth to an already competent person. However, mothers' and fathers' roles in the child-rearing practices are subsequently altered, as demands for gender equality are increasing. One strategy is to reduce obstacles for women to gain access to spheres of society traditionally dominated by men (occupations, power positions within business and politics, etc.), reducing the asymmetrical power between men and women in society. A qualitatively different route is to reduce the constraints of childbirth and child rearing, thus reducing the interdependence between men and women. Such measures involve subsidized childcare and paid parental leave, and efforts to encourage men to take part in child rearing, for example by allocating some of the subsidized parental leave to the father (Melhuish & Petrogiannis, 2006;Thévenon, 2011). This would suggest that another ideological view might emerge, where child rearing is not seen as a problem and where it is not necessary to regard women as less competent at work, as both men and women are supposed to take part in child-rearing activities. It is even possible that parenthood might be seen as an asset in its own right at work. However, as evidenced by the research of Heilman and Okimoto (2008), parenthood might evoke problems at work in comparison to people not having children, regardless of gender, as they risk being seen as less motivated and focused on their work. So instead of causing an ambivalence about how to perceive women in career positions, as a traditional ideology would suggest, it might cause a dilemma between this more "modern" ideology, where child rearing should not cause a problem at work, and the traditional view, where taking care of children would be seen as an obstacle, although now it is generalized to both men and women. Thus, prejudice formerly directed at women at work might now be directed at parents at work.

Group decision making and gender discourses
How prejudice influences judgements and decisions is mainly studied on an individual level as perceptions, judgements, and decisions made by individuals. However, much decision making takes place in groups. At work, although a single person is entitled to the final decision, issues concerning hiring people, job assignments, salaries, career planning, and so on are at least discussed and prepared, and recommendations are made in groups. Thus, decisions are influenced by the combination of the participants' views, and the dynamics within the group. How are different gender ideologies negotiated in such a situation? Ideologies are upheld and mediated through different discourses by sharing beliefs and values with others through language. Discourses serve the function of reproducing social institutions and also constructing people's ways of thinking (Potter & Edwards, 2001). Through different discursive positions, individuals understand and give meaning to different aspects of life (Elvin-Nowak & Thomsson, 2001), for example how to view the combination of parenting and work positions that may be expressed through the language used in groups.
There are some studies focusing on discourses regarding parenthood in general and parenthood in relation to work, and mainly regarding motherhood. Wetherell, Stiven, and Potter (1987) found in a study among final-year university students that the participants experienced a dilemma between their endorsement of equal opportunities for men and women and the practical constraints that might limit these opportunities. In a study among mothers in Sweden (Elvin-Nowak & Thomsson, 2001), three different discourses were found. One focused on the mother-child relation, implying the importance of the child's access to the mother for its well-being. The other discourse focused on the child-mother-woman relationship, which also puts the child in the centre but where the mother as a happy woman (with her own interests) is important for the child to be happy. The last discourse focused on motherhood as separate from employment. In this position, gender equality was important, which made negotiations necessary between the two spheres, with their spouses or with their own conscience. A study from New Zealand (Baker, 2010), based on interviews with different types of parents, evidenced that the integration of work and family was a more complex matter for mothers than it was for fathers, and centred around perceptions of support, work requirements, and ideas about "good mothering". A study in Norway, among women with high-commitment careers (Halrynjo & Lyng, 2009), identified three steps taken by these mothers, from "careerists by choice", over "maximizers" trying to fulfil commitment to both work and family, to end up as "satisfiers", finding less demanding and more family-friendly positions, with a new discourse that career withdrawal is the moral thing to do. An American study (Bass, 2015) among couples, not yet parents, found different ways of thinking among men and women, where the women already had a lot of thoughts about the dilemma between parenthood and career, while the men, on the other hand, did not think as much about a future role as a father but leaned more towards a breadwinning position.
Studies that focus solely on fathers seem to be less frequent. A British study among first-time fathers found that although they thought that there were opportunities to challenge gender norms, they seemed to retreat into traditional roles (Miller, 2011). Despite the fact that these findings span over almost two decades and originate in countries differing in how equality opportunities are implemented, they all show similar patterns. Discourses seem to centre on the practical constraints that parenting is perceived to entail, but also on the implied "free choice" of mothers to refrain from careers. These discourses to a certain extent exemplify an ideology where men and women have complementary roles, which is in line with the Stereotype Content Model and Ambivalent Sexism Theory. But they also evidence a dilemma between such norms, and a wish both among men and women, mothers, and fathers to overcome the obstacles raised by such norms.

Aim of the study
As shown above, discursive studies about parenthood and work have mainly departed from the perspective of the parents and, most of all, the implications for mothers. However, such discourses are most certainly also present among people making career decisions about people being parents. However, those kinds of judgements have mainly been studied at the individual level. To the authors' knowledge, studies about how groups negotiate gender and parenthood along these lines are absent. The aim of this study was to study how gender, in relation to parenthood and career decisions, was negotiated within a group setting. Does parenthood constitute a problem when selecting a candidate for a position? Is it more problematic if the parent is a woman? In this pursuit, the study attempted to apply the stereotype content model (Cuddy et al., 2008) in order to see if the discursive content centred on the dimensions of warmth and competence. Furthermore, a conversational analysis we was applied (Canary & Seibold, 2010;Meyers & Brashers, 2010) in order to investigate how different discourses are negotiated within groups. This model penetrates the various components that build up an argumentative conversation. The analysis was based on ad hoc groups consisting of university students in an experimental setting.

Method
Participants A total of 130 students (85 women) participated in a group experiment. The mean age of the participants was 25.9 years (SD = 5.82 years). They were recruited from various study programs, mainly in social and behavioural sciences.

Procedure
When the participants arrived at the laboratory, they were assigned into groups of four people (sometimes three or five for practical reasons) in one of three conditions: all male, all female, or mixed gender. They were placed in a room around a round table. At the beginning of the session, they completed a questionnaire individually. This questionnaire was modelled after Cuddy et al. (2004) and Heilman and Okimoto (2008). The questionnaire contained descriptions of three candidates for a position as a consultant. The first candidate was a man (Jörgen), the second a woman (Lena), and the third was in half of the cases a man (Daniel) and in half of the cases a woman (Katarina). Information was given on each candidate's age, education, work-life experience, preferred work style, and hobbies. For the last candidate, it was also added that he/she had a newborn child. The two versions of the questionnaire were randomized across the groups. Each candidate was evaluated on 18 items related to competence and warmth in order for the participants to form their own individual impressions of the candidates.
When this was done, the participants were instructed to make a joint decision about these three candidates. They were given a protocol in which they were asked to rate the candidates on four items (on a response scale ranging from 1 to 5): to what extent they were considered to work efficiently, to what extent they would create a good atmosphere, to what extent they would recommend them for promotion, and to what extent they were regarded to be eligible for further training. Finally, they were asked to rank order the candidates for the position as a consultant. The group discussion was constrained to 15 minutes and was filmed. There were 35 groups in total, 20 female groups (10 each for the two conditions), seven male groups (four with the female condition and three for the male condition), and eight mixed groups (three for the female condition and five for the male condition). As the true aim of the study was not revealed to the participants when the experiment began, they were fully debriefed immediately after the session. They were given the opportunity to ask questions and to withdraw from the investigation. Finally, they signed a written agreement form allowing the collected data to be used for research purposes.

Data analysis
The filmed material was transcribed and analysed in three different steps. The first step was to identify which kind of information was given about the candidates that were used in the discussions. The second step was to analyse what said about the third candidate (the parent) and to apply the stereotype content model (Cuddy et al., 2004) by categorizing extracts into dichotomies of the dimensions of warmth and competence. The third step was to analyse how parenthood, gender, warmth, and competence were negotiated in the groups. Here, a brief version of the conversational argument coding scheme was applied (Canary & Seibold, 2010;Meyers & Brashers, 2010). This scheme has previously been used successfully in a similar context (Löfstrand, 2015). According to the scheme, argumentation can be divided into three major components. The first component contains two categories: one that constitutes the input into the discussion in the forms of facts and opinions (arguables), and the other that consists of elaborations of these facts and opinions (reasoning activities), for example how these are justified or extended. The second component is labelled convergence markers and may be viewed as supporting activities such as "I agree on that", "that's a good point", and so on. The third component is labelled prompters and could be viewed as disagreement activities such as challenges and objections such as "how do you mean?" and "I don't agree on that".

Results
The information that was given about the candidates was their age, education, work-life experience, preferred work style, and hobbies. For the third candidate, the information also included that he or she had a newborn child. The candidates' age was discussed in every fifth group. In some of the groups, it was mentioned that one of the candidates was older than the others, but almost immediately another participant answered that all candidates were of a similar age. Education was mentioned in a quarter of the groups, for example "everyone has a good education and has worked a couple of years" or "she has studied communication, but does that make her more efficient?" Work-life experience was discussed in slightly less than half of the groups. It was seen both as an advantage -"He has worked many years in the company so he will be efficient"and as a disadvantage -"He has worked in many different companies; he will probably not stay for long". Hobbies were discussed in slightly more than half of the groups: "she likes to play tennis and has probably a competitive approach" or "she likes to take a walk in the parkthat doesn´t sound good when it comes to creating a good atmosphere". Preferred working style was the second most discussed topic. It came up to discussion in almost three out of four groups. It was mentioned both as an advantage -"she likes to work one-to-onethat means that she is a great communicator"and as a disadvantage -"as she works one-to-one, it probably takes her longer time to be done with the work". Parenthoodinformation that referred only to the last candidatewas discussed in every group. So, how did they talk about parenthood?
All the group conversations were analysed on the basis of how they used the stereotype contents of warmth and competence. Characteristics of high competence mentioned were experienced, responsible, organized, efficient, and strong. Characteristics of low competence were such as inefficient, absent, unfocused, and tired. Examples of high warmth were loyal, trustful, humble, cuddly, and emotional. Among low warmth attributes were grumpy and stressed.
As can be seen in Table 1, although there were some attributes that were common for both the male and the female target, the main picture reveals different patterns of statements according to the target person's gender. For example, for high competence, the only common attribute was efficient. Furthermore, there were also indications of different patterns depending on context. The discussions differed according to the gender composition in the group. How the competence and warmth attributes were used is presented separately for the female and male targets in the next sections and ordered according to the gender composition of the groups.

Female target
Groups made up only of women often used the attribute "efficiency" when they talked about the female target. She was supposed to be efficient because of her parenthood status. "She is in a new phase of her life, and that may have impact independent of gender. She may make her 40 hours a week, but in a situation with reorganization, more may be needed, but I think she is very efficient", or "One thing is for sure, when you get children, as me for example, or as here, recently has got a child, then you are damn efficient to get it all together", "She may be more efficient while at work", "Katarina is more efficient, as she has a child". The female target was also seen as organized: "I thought like this, if you have children you are probably also organized", "Strong", "I believe that women with children are strong persons". Finally, in some of the groups, the female target was described as a career woman, high in competence, such as "focused": "I believe that Katarina is a career woman who is focused but not so interested in social activities when at work, as she has a child". Groups made up only of men talked less in terms of high competence about the female target than about the male target. When they used highcompetence attributes about the female target, they often simultaneously moderated such statements with views that motherhood could be a problem: "I believe that she is very efficient, although she will be absent a lot", "I believe she is experienced but the child is a negative". The only example of high competence, found in groups with both men and women and related to high competence for the female candidate, was "efficient", also used with a slight reservation: "I believe that she will be efficient while at work".
Groups made up only of women used attributes relating to low competence in two different ways. The first was to describe personal factors, for example that she may be tired -"Katarina is probably really tired, as she doesn´t get any sleep during the nights because of her kid"or that she may not have the best conditions -"Katarina's conditions may not be the best because she recently got a child". The second way was related to structural factors, as it could cause problems for the company: "If you consider what is the best for the company's economy, Katarina is not a choice. It is best for the company, even if it is really, really awful, and I will work against it as long as possible". In the male groups, the reasoning relating to structural factors implied that it is impossible for a mother to have the assets that are important for the company: "It is her, not the company, it is she who wants more time with her family", "It would be macabre to pick Katarina, as she recently has become a mother". The same kind of reasoning was found in the mixed groups: unfocused -"She who has got a child may be unfocused", "It may not be time for promotion for her as she recently has got a child"or absent -"She will take the day off very often".

Male target
The female groups also talked about the male parent in terms of efficiency: "I believe that Daniel could be efficient, he has got a child and has come far". The male target was also presented as experienced: "I should say that the person who is the most experienced will do the job best, that person is Daniel who has become a dad", "Daniel is experienced and will be efficient, as he has recently become a father". In some of the female groups, the male target was regarded as responsible and motivated, as he would prefer to be at work rather than at home because of the child: motivated -"If Daniel recently has got a child he will perhaps be more motivated to escape from home"responsible -"As he has got a child I think that he is responsible". Male groups talked more in terms of high competence about the male parent than they did about the female parent and also without reservations. Instead, they used parenthood as something that gave strength to the high-competence attributes. First, Daniel was mentioned as efficient: "You may be efficient when you get children because you learn to think in a new way". Second, he was described as experienced: "I believe that Daniel is experienced and he will be efficient, as he has become a father". The male parent was also described as alert -"He has got a child and is probably alert", disciplined -"He has got children and stuff, he will be disciplined", and motivated -"If Daniel has got a child he will probably be at work rather than at home". In the mixedgender groups, there were few competence attributes for the male parent. One example was "Those who have small children may be attracted to Daniel", indicating that as a father, he has competences of use to other people. The reasoning relating to low-competence attributes in the female groups were of two kinds. The first one was similar to the one presented for the female target, namely personal factors, such as inefficient -"He has recently got a child, how efficient may he be?", tired -"Daniel who recently got his first child may not sleep so much and will probably be very tired", or poor conditions -"Daniel has recently got a child and may not want promotion". The other kind is similar to the way male groups discussed the female target, namely the inevitability of parenthood being incompatible with the company needs such as being absent -"Daniel will probably be at home to take care of his child almost always" or "He will be absent often because of childcare". The male groups displayed reasoning similar to the female groups. The personal factors mentioned were unfocused -"If Daniel has got is first child it will be much talk about that child", absent -"He may be absent because of sick children and such", or has a limited time for work -"Daniel has recently got his first child, it may take lots of time and such". The mixed groups used the attribute lack of focus: "Yes, it may be too much for him, it is pictures to be shown and such".
To summarize, parenthood was discussed in most of the groups as an asset for the organization. They thought that it adds to the competence, as the candidate was seen as more focused, organized, experienced, and efficient. This was especially evident in the female groups discussing a female target and in male groups discussing a male target. Male groups discussing a female target shared this view, albeit with some reservations: "I believe she is experienced, but the child is a negative". For the male target, there were also some examples from both female and male groups of competence attributes from a traditional breadwinner perspective such as motivation. The few examples from the mixed group resembled most the statements in the male groups. Regarding lack of competence, there were more similarities than there were differences between the groups. They all presented statements of personal factors, such as being tired, and structural factors, for example absence and causing a problem for the company. While most of the groups saw absence as an inevitable consequence of parenthood, the female groups with a female target did not discuss absence at all. In fact, one group explicitly formulated a dilemma regarding parenthood as between an ideal situation and structural obstacles in reality.

Female target
Statements related to warmth were less frequent in the group discussions. Female groups talked of warmth in relation to the female target in two ways. The first category involved statements of warmth as an asset for the company such as trustfulness -"I think she, who has got a child has to be trustworthy" and loyalty -"as she has got a child she will maybe feel loyalty" and "Katarina is probably the most loyal among them". The second category contained reasoning about more general traits, for example humble -"I have a picture of parents as humble", or emotional -"Yes, but women who have children may be emotional". Male groups only displayed statements in the second category, and the female target was described as emotional and understanding: "She who has got her first child may be more emotional and understanding. I don't know, what do you think, does it have any sense, I think so". In these groups, there were also examples of warmth solely directed towards the target's role as a parent: felicitous -"She is of course felicitous". There were no examples of warmth found in the mixed groups. There were only a few statements indicating low warmth, all of them in the female groups. The female target was described as unstable: "She may have lots of hormones and such, hahaha".

Male target
For the male target, there were only warmth statements in the female groups. Most of them related to the target's role as a parent: felicitous -"Daniel may be really felicitous as he recently has become father", cuddly -"And then we have Daniel, the cuddly dad", and satisfied -"I have a feeling that he is satisfied with his situation". There was also an example of general traits such as friendly: "He has children; he is friendly and warm". Also, for the male target, there were few examples of low warmth. He was described as stressed -"I believe that he can be easily stressed" and bored -"Or perhaps he is damned bored".
To summarize, warmth attributes were almost only discussed within the female groups. Regarding the female target, both female and male groups exemplified warmth as general traits such as friendly and humble. Warmth as an asset from a company point of view (trustful, loyal) was evidenced only in the female groups and only for the female target.

Group negotiations of parenthood
In the previous section, statements from the conversation were presented exemplifying different gender ideologies. One of them was that parenthood in itself is an asset to the organization. Motherhood adds both competence and warmth to the company, while fatherhood adds competence, sometimes more of a traditional breadwinner character. How are these ideologies negotiated within a situation where a decision has to be made regarding the selection of one candidate over the others? Three discussion patterns were found in the present material, differing in terms of the complexity of the various argumentation categories used, such as facts and opinions presented (arguables), how these facts and opinions were elaborated upon and motivated (reasoning activities), in what ways they were supported (convergence markers), and how they were challenged and objected to (prompters).
The first pattern was characterized by a set of opinions that were not negotiated at all. They were not elaborated upon, and they were met with neither supporting activities nor challenges or objections. The statements and corresponding beliefs seem to have been taken for granted. The first example is taken from a group of four women with a female target when, at the end of the discussion, they were to rank the order of the three candidates. In this example, they seem to be of the same opinion. At least, they are not challenging each other's opinions.
Woman 1 : "She may be less good than the others, as she has recently got a child". Woman 2 : "Yes, is that what you think of." Woman 1 : "OK, then we put her here" Woman 3 : "It is because of the child; that is what you think".
The second example comes from a group of four men with the female target. The conversation took place after a few minutes of discussion when they were to rate the candidates' likeliness for promotion, on a scale from 1 to 5. Although it seems that Man 2 asks for another person's opinion, the conversation ends there and then takes another direction.
Man 1 : "Experienced, good, but the child gives her a lower score". Man 2 : "What do you think?" [asks Man 3] Man 3 : "She has worked with things that sound relevant; the question is if she is ready".
The third example is taken from a mixed group with the male target. The discussion took place when they were to rate the three candidates' likeliness for promotion. They seemed to make suggestions, which seldom were reasoned about and which never met with either support or challenges.
Man 1 : "Daniel has got his first child, lots of talk about that child". Woman 1 : "Yes, it can be too much". Woman 2 : "Then we put Jörgen on top". Woman 1 : "I think Lena is good. Those who have children may be attracted to Daniel". Man 1 : "Recommend for promotion". Women 2 : "That will be tough".
Man 1 : "I thought Jörgen and Lena, as Daniel has got his first child".
All of these examples involve statements that are never meet with counter-arguments, not even when it seems that argumentation is asked for (i.e. "Is she ready?"). Furthermore, they are also all examples of when parenthood implicitly is seen as a problem. Examples of such consensual discussions where parenthood was considered an asset were not found.
The second type of conversation involved statements that were backed up with argumentation, predominantly in the form of reasoning activities motivating the opinions made. These arguments were mostly used to promote the parent above the other two candidates. The first example is taken from a group of women with the male target. It took place when they were asked to rate the target person's efficiency. Woman 1: "I thought, he has a child then you will become responsible". In this case, the woman clearly finds the person efficient and backs this opinion up with an argument. However, nobody responded, and the discussion continued with another topic.
Another example was found in a group of women with the female parent. The discussion started with the conversation below. In this example, an opinion is justified by adding further information. This is responded to by the others with other arguments (strong, emotional), and finally a further elaboration is made in the form of an explanation of why parenthood is an advantage.
Woman 1 : "I like Katarina, a woman with speed". Woman 2 : "That is what I also think". Woman 1 : "I believe that women with children are strong persons". Women 3 : "they should in any case be more emotional". Women 1 : "I believe that you can see things more broadly then. It doesn't matter if it is a man or a woman who has a child, but I don't think that there is a disadvantage to have children".
The third example comes from a male group with the female parent. The argumentation occurred when they were set to rate the candidates' efficiency. The example starts with a question, which is met by a convergence marker in the form of support and at the same time a further elaboration of the topic.
Man 1 : "What do you think about her efficiency?" Man 2 : "I think she may be more efficient, but at the same time she has got her first child, which may mean that she will be absent".
Man 1 : "Experienced yes, but the child takes her down a bit".
The difference between this pattern and the first one is that here the standpoints are backed up by argumentation, even when the members of the group seemingly have the same opinion. The arguments mainly have the form of reasoning activities elaborating on the topic at hand by providing more information or points of view. The argumentation is focused on why or why not parenthood is or ought to be seen as an asset in working life. As this is a more untraditional view, it seems that the participants felt that they needed to back it up with arguments, even when they seemed to agree.
The third and most complex pattern, which was the most frequent type of discussion, involved not only argumentation but also situations where opinions changed as an effect of the argumentation. Not only were opinions made, but they were also elaborated upon through reasoning activities and often met by prompters such as challenges and objections.
The first example was found in a group consisting of four women with the female parent. The argumentation occurred at the end of the discussion when the group was to rank the order of the three candidates. This conversation starts with a general opinion that motherhood is a problem. However, this is then forcefully rejected ("I will work against it"), which opens up for new arguments, which ends with a further elaboration of the argumentation where motherhood is seen as an asset.
Woman 1 : "Katarina's conditions are not the best, as she recently has become a mother." Woman 2 : "Should we reverse it? We can put Katarina in a lower position". Woman 4 : "Yes, we put Katarina here and the others above her". Woman 3 : "I thought that this question is really hard, as it is so many facts to consider". Woman 4 : "But if we should consider the best for the company's economy, Katarina should not be chosen. It is really awful, and I will work against it, but it is how it is in work life". Woman 3 : "Yes, but as she has recently become a mother, she will be loyal". Woman 4 : "I have heard that it is positive for the company when employees get children, as they can separate work and private life, and this leads to more efficiency".
The second example is taken from a group of men with the female parent. The extract occurred when discussing the target's likeliness for promotion. It is an example where reasoning activities in the form of elaborations are met with challenges leading to more elaboration, which then is met with even more challenges.
Man 1 : "She has recently become a mother, then she could be more emotional and understanding. I don´t know, what do you think? Does it have any impact that she has got a child? I think so". Man 2 : "She is of course happy". Man 3 : "Yes, exactly, she is happy as you say. She has got a child and life is easy. I should put her on top. According to promotion, I think that if you have a newborn, then you want to take it easy and put effort to family life, but it doesn't need to be that way?" Man 4 : "I don't want to think like that. She is the best of them". Man 3 : "For promotion, how do you motivate that?" Man 4 : "Even if she has a child?" The third example comes from a mixed-gendered group with the female parent when they were to rate the three candidates' potential for efficiency. This is an example of when different opinions are put forward (is she efficient or not), where one argument is met with a counter-argument (objection). The discussion ends with some kind of compromise.
Woman 1 : "I believe Katarina is most efficient, Lena least, and Jörgen in between". Woman 2 : "But Katarina has got her first child".
Man 1 : "That doesn't mean that she is less efficient". Woman 1 : "No, but it may mean that". Woman 2 : "She may be efficient while at work".
The fourth example was found in a group of four men with the male parent at the end of their discussion where they were set to rank the order of the three candidates. This situation seems to start with consensus that Daniel (the parent) is the best candidate. An opinion is backed up by support from the other group members. However, this consensus is then challenged by one of the group members ("the child issue"). Then this issue is elaborated upon.
Man 1 : "I should say that the person with the most experience would work best". Man 2 : "That was Daniel". Man 3 : "Yes, who recently had become a father. I should put Daniel high". "Man 2 : "But the child issue?" Man 1 : "If they are sick and such it may be a problem".
These examples display a more complex conversation pattern where different conversation components are used: arguables, reasoning activities, convergence markers, and prompters. Furthermore, the alternate use of these components also seems to drive the discussion forward so that more information is provided, more opinions are added, broadening the scope of the discussion, and finally sometimes the way they think about the decision to be made is changed. Argumentation is related not just to competence in general but also to specific competences originating in the experience of being a parent. Also warmth arguments are used as an asset for the organization. At the same time, "the child issue" causes a dilemma in the discussions. The same discussion can contain parenthood as both an asset and a drawback. Sometimes this dilemma seems to emanate from perceptions of personal factors (wanting to put family first) or from perceptions of structural factors ("it is how it works in work life" and "I don't want to think like that").

Discussion
This paper explored how parenthood was discussed in a fictive recruitment situation. The methodological approach was a qualitative analysis of the conversation in small groups assigned a decision task. The main findings were that parenthood was the dominating issue discussed. Parenthood was discussed more in relation to competence than it was in relation to warmth. Parenthood was seen as influencing competence both negatively and positively and thus both as an advantage and a disadvantage for the organization. Finally, three conversational patterns were discovered, differing in the amount of and how conversational arguments were used.
All groups discussed parenthood, which is quite remarkable, given the fact that more relevant information such as work experience and preferred working style was not. It is of course possible, as the parenthood information was only given about one of the candidates, that it made this candidate stand out from the others. However, it could be argued that it should not have been discussed at all, since it was not a work-related issue that should be considered in a recruitment situation. Hobbies, for example, another non-work related issue, were only discussed in about half of the groups. Besides, as it is a strong norm in Sweden to make it possible for parents to take an equal part in working life, it would be assumed politically incorrect to give it so much weight in the discussion. Interestingly, they most often were very careful to avoid relating the parenthood issue to gender.
Parenthood was discussed most often in terms of competence and less often in terms of warmth. It seems that parenthood was considered an asset for the organization, adding competence, as the candidate was regarded as more organized, experienced, and efficient. This was especially true in groups with women discussing motherhood and in groups with men discussing fatherhood. The male groups discussing the female target as competent sometimes did this with reservation. That female groups so strongly advocated the enhanced competence that motherhood brings without any reservation is perhaps not very surprising, as the dilemma between work and family is closer to home for women, even without children of their own. In the study by Bass (2015), it was found that young women already think about this dilemma before becoming mothers to a much higher degree than young men not yet fathers. It also mirrors one of the discourses among mothers found by Elvin-Nowak and Thomsson (2001), which focused on the child-mother-women relationship where the mother as a happy woman with her own interests is regarded a separate entity and not just an extension of the child. In the male groups, there was some reservation regarding the female parent's ability. They acknowledged that there was a dilemma between motherhood and work. They made no such reservations when discussing the father. For the male target, statements regarding breadwinning were found in both the female and the male groups. Men, regardless of their family situation, fit into the standard role of a person in the workforce, while mothers are considered as a contrast to this. Very few previous studies have focused on the modern father's role, but even though roles that challenge gender roles are imagined and welcomed, it seems that it is difficult to persevere in this position (Miller, 2011).
When parenthood was seen as lowering competence, the groups viewed the parent as being tired but more often as being absent a lot. It is interesting to note that the groups took absence for granted as a structural aspect of having children. As far as we can see, no group challenged this statement. It is taken for a fact. Yet, nobody can say that this specific candidate, Katarina or Daniel, would be absent a lot. As such, it could implicitly mean that they regard parents as less motivated for work. On the other hand, as parents in Sweden have a legal right to stay at home with children when they are ill, there is a fact that sometimes they are absent. However, as a considerable proportion of the workforce are parents, this would be regarded as normal. It is also a fact that women use these rights more often than men (SCB, 2015), which might indicate that parenthood in terms of motherhood is seen as something out of the ordinary in an organization. Given that men, regardless of their family situation or where the family situation does not affect working life, are seen as a "default group" (Cuddy et al., 2008), parenthood, when it does affect the work situation, is seen as odd. So, in order to get parenthood to be regarded as a normal part of working life, more effort has to be directed at having fathers take parental leave as often as mothers do. This is a considerable challenge for employers, but it has been found that companies with a father-friendly culture, with support from the managers, results in men using more of their parental leave (Haas, Allard, & Hwang, 2002).
Three different discourses about how parenthood was negotiated were discovered. The first involved opinions and statements that were seldom met with responses from the other group members. The second discourse also involved opinions and statements, but these were followed by reasoning activities backing up those statements. Finally, the third and most common discourse involved activities from both of the first two, and in addition contained challenges, objections, and disagreement with the opinions presented. The first discourse in this material seemed to be most frequent when parenthood was discussed as a disadvantage. The statements presented took the form of "facts"as common knowledge that did not need to be discussed. This is how norms are often expressed, and not as the opinions they are, originating in a certain ideology. From the perspective of ambivalent sexism theory (Glick et al., 2000), a sexist ideology cannot see that traditional male roles (career) can be combined with traditional female roles (childcare). Thus, a solution to the dilemma, as presented for the participants in this experiment, was that one kind of roles had to losein this case parenthood. In this way, when sexist attitudes are never challenged, the negative views of traditional female roles are easily preserved.
The second pattern seemed to be more common when the groups discussed parenthood as an asset for working life. When they stated opinions, they usually backed them up with more informationjustifications to motivate their point of view and so on. Often, other group members added to this kind of reasoning activity and also showed support. As such views are contradictory to a traditionally sexist ideology, perhaps they thought that they had to give these opinions more ammunition, hence the more elaboration. In spite of the fact that much information was provided in the discussions, it did not facilitate challenges or objections. To talk about parenthood as adding competence to the organization was most prevalent in female groups discussing the female target. If this lack of questioning argumentation was due to them all sharing the same opinion or that the supporting group members just wanted to be politically correct is an open question. Anyhow, this conversation pattern suffers the same risk as the first pattern, that is, not to challenge a certain set of norms, although the norms evidenced in this group is of a more progressive kindthat parenthood should be a normal part of working life. The problem, though, is that this discourse overlooks the complexity of the question of work/family integration. The studies presented in the introduction based on interviews with parents, not yet parents, first-time parents, and so on clearly demonstrated the struggle that these people have over these issues (Baker, 2010;Bass, 2015;Elvin-Nowak & Thomsson, 2001;Halrynjo & Lyng, 2009;Miller, 2011). Further, as Halrynjo and Lyng (2009) conclude, demands and expectations of both work and family need to be exposed and challenged in order for real change to happen.
It seems that the final pattern is the one with the best preconditions for making such norms visible. In this line of conversation, not only were opinions stated, further elaborated, and supported, they were also met with disagreement and other forms of challenges. As a consequence, questions were penetrated more deeply, and the course of the argument sometimes took a different route and even changed the initial proposal into a more anchored final decision. It has previously been found that such elaborated argumentation not only leads to more satisfaction (Löfstrand, 2015), but also to higher decision quality (De Dreu & West, 2001;Rijnbout & McKimmie, 2012).

Methodological considerations
This study was based on a student sample randomly assigned to ad hoc groups, thus differing from natural decision groups. They were not used to work together, they did not have a joint history, and they would probably not have anything to do with each other after the experiment ended, while a natural decision group is systematically composed, have worked together before, and will do so in the future. Still, we found implicit use of stereotypes such as competence and warmth, although the groups avoided relating them to gender but instead focused on the child issue. Maybe a natural decisionmaking group would not be that careful, given that they perhaps already know each other's viewpoints and share a belief system based on their common experiences, especially in single-gendered groups.
Three different conversational patterns of which the third was the most elaborated were also found where gender and parenthood were clearly negotiated by weighing pros and cons against each other. Such a pattern would be assumed to happen when participants feel comfortable in the group, that is, which is more probable when they know each other. However, in a natural group which has worked together before, it is possible that they feel that they have gone beyond such a stage where they always need to present their standpoints and argue about it. This might lead to an illusory consensus, risking not challenging implicit norms, such as in the first two conversational patterns. Furthermore, natural decision-making groups are shaped by their shared history of norms and practices.

Conclusion
It is clear that parenthood evokes a dilemma between work and family life among people, even though they are not parents themselves. All the groups in the present study discussed parenthood, even though it is not a work-related issue and should not be considered in a recruitment situation. As this study was based on a student sample, further research should focus on conversation patterns in real decision-making groups. The present study elaborated aspects of gender and parenthood. It would be interesting to broaden the scope to other issues that might risk discrimination. Further research could thus study how the discussion would proceed if, for example, gender was contrasted with ethnicity or sexual orientation. Three conversation patterns were found, differing in the amount of elaboration of the topic of parenthood and work. It seems that the most elaborated kind tends to foster a situation where implicit norms are more easily made visible. Thus, another line of research could be to educate groups into more elaborated conversation and to investigate how this influences the discussion and the actual decisions.
The bottom line of this paper is that a conversation pattern characterized by divergent opinions challenging each other seems to be a fruitful path to follow in order to avoid too many discriminating decisions. If this is true in a more general sense, it would be possible to set up a few conversational rules in decision-making groups in order to minimize the risk of faulty decisions and discrimination. First, make sure that everyone has the opportunity to state their opinions. Second, make sure that all opinions are elaborated upon, that is, by providing more information, by justifying the standpoints, and so on. Third, and most importantly, make sure that counterarguments are provided, regardless of whether they are endorsed by anyone. In this way, implicit norms are more easily made visible and thus easier to challenge.

Disclosure statement
No potential conflict of interest was reported by the authors.