Abstract
Women physicians are confronted with incompatible gendered role prescriptions, whereby the role of the ‘ideal’ mother contrasts sharply with that of the ‘ideal’ physician. This study introduces four goal frames that reflect how women physicians internalize these conflicting role prescriptions and investigates the relationship between women’s goal frames and their career motivation. It also examines the relationship between gender-equality arrangements – inspired by the same underlying ideals – and women physicians’ career motivation, and whether these arrangements moderate the relationship between goal frames and career motivation. Cross-sectional data on 1070 Dutch women physicians collected in 2008 indicate that women physicians with switching goal frames (i.e. those who want to live up to both ideals) are no less career-motivated than women with one dominant goal frame. However, gender-equality arrangements mainly seem to support women physicians who prioritize one role over the other. No evidence was found that gender-equality arrangements support those who try to combine conflicting role expectations.
Keywords
In many European countries, more and more women have entered the medical profession during the past two decades. On average, 35 percent of all medical practitioners in countries such as France, Germany and the UK are women, and in Finland that proportion is about 50 percent (e.g. Van der Velden et al., 2008). The Netherlands, where the current study was conducted, would seem to represent an excellent location in which to examine the potential for gender inequality in the medical profession. This is because the Netherlands is a pioneer when it comes to part-time working in the medical profession, with about 70 percent of women physicians having a part-time position (Meijer and Heesen, 2005). At the same time, however, the medical profession is known for its demanding work ethos: working at least eight hours of overtime every week is considered the norm, even for those who work part-time (Pas et al., 2011a; Jagsi and Surender, 2004). Although Dutch legislation on part-time working is designed to facilitate those with caretaking responsibilities, 1 this is a task that typically still falls to women (Bagilhole, 2006; Hochschild, 1989). Moreover, such legislation has not been able to rectify gender inequality in Dutch healthcare institutions, which is demonstrated by two phenomena. Firstly, the Dutch medical profession continues to be segregated. Women are more likely to work in the less prestigious specialities, such as paediatrics (or community health centres), family practice or for health insurance companies, where part-time jobs are more common and considerably fewer women enter more prestigious areas, such as surgery (Heiligers and Hingstman, 2000). Secondly, women medical specialists are still underrepresented in senior academic positions and higher management in hospitals (Van Doorne-Huiskes and Van Beek, 2009). The medical sector in the Netherlands is thus subject to the same gender inequality that has been noted in other professions in other countries, such as dentistry (e.g. Adams, 2005), law (e.g. Bolton and Muzio, 2007) and veterinary medicine (e.g. Irvine and Vermilya, 2010).
Empirical research conducted over the years in various scientific fields has substantially improved our knowledge of women’s career paths. Studies have examined societal, organizational, household and individual characteristics as predictors of women’s labour force participation and career development. However, as there are different theoretical explanations for these empirical findings, the results have been the subject of much academic and public debate. Three key aspects of this debate are particularly relevant for our research. The first source of controversy in the debate about women’s employment is that priority is accorded to women’s role as caregivers, domestic activities and family life (e.g. Hakim, 2002; Walsh, 1999). The role of motherhood ideology, or what is believed to be good mothering, has been included in some studies as a critical aspect that affects not only women’s labour participation, but also their career motivation (Himmelweit and Sigala, 2004; Stone, 2007). The feminist critique on this type of research is that it emphasizes women’s roles as mothers and that it runs the risk of reproducing ideological stereotypes that motherhood is incompatible with career commitment and advancement, reinforcing the traditional sex-specific public−private divide (Benschop, 2006). Moreover, studies on the effect of motherhood and societal motherhood ideology on women’s career investment have neglected the inherently multifaceted nature of social identity and the complex way in which individuals combine different roles when constructing their identity (Bodenhausen, 2010). Highly educated women, in particular, invest a great deal of time and energy in their career. Working and aiming for professional growth and career development is an important aspect of a woman’s creation of her self-identity (Stone, 2007). Moreover, highly educated working women, such as the women physicians whose attitudes and beliefs are investigated in this study, face not only stereotypical beliefs regarding the ‘ideal’ mother, but also gendered beliefs about the ‘ideal’ worker (Acker, 1992; Bailyn, 2006). The characteristics of an ideal worker are fairly straightforward and, often indirectly, gendered (Acker, 1992). The ideal worker is willing to work overtime, to sacrifice private life for work, to be on call, and to relocate for work; he or she is highly appreciated by colleagues, is not ‘encumbered’ with caretaking responsibilities at home and is physically capable of working long hours. The ideal physician resembles this ideal worker in that he or she is expected to be highly dedicated and always available, which obviously implies working overtime as a matter of routine (e.g. Brennan and Zinner, 2003). A role conflict occurs because one cannot – by definition – be an ‘ideal’ mother and an ‘ideal physician’ simultaneously, although researchers have also noted the image of ‘superwomen’, the working mothers who do it all, or at least seek to (Thompson and Walker, 1989). Researchers broadly agree that nowadays ever more employees are struggling to reconcile the incompatible roles associated with work and family life (e.g. King, 2008; Nippert-Eng, 1996). Research focusing on the problematic interaction between women’s family and work roles has already identified many harmful effects. For example, Raskin (2006) found that women’s work−home conflict is a cause of significant psychological distress, while King (2008) notes that work−life conflicts may adversely affect the way superiors perceive women’s commitment and career advancement. However, these studies have not examined whether women do in fact internalize gendered norms on the ideal mother and the ideal worker; rather, they have taken this conflict for granted and investigated how women then go about separating, integrating or managing these domains (e.g. Nippert-Eng, 1996) and to what extent this ‘balancing act’ affects women’s career commitment or labour participation.
The second aspect of the debate around women’s participation in the labour market concerns the relationship between working hours, career motivation and women’s alleged inability to reach more senior positions. In both the public and scientific debates on gendered inequality in senior positions, it is often – implicitly, explicitly or for the sake of convenience – assumed that women’s working hours can be explained by their (alleged) lack of career motivation (e.g. Hakim, 1993; Taylor et al., 2009). However, such an explanation is far too simplistic. Women’s employment histories often comprise both full-time and part-time jobs and the level of their career motivation is not directly related to either full- or part-time working (Tomlinson, 2006). Moreover, in a profession in which long working hours are taken for granted, the number of hours worked is unlikely to indicate a significant difference in the motivation of distinct groups of medical physicians. This study does not therefore look at the number of hours worked, but focuses on career motivation, and specifically the interests, attitudes and personality variables that may be relevant to career development (London, 1983). The construct of career motivation in this study includes the facets of: (i) career centrality; (ii) career insight; and (iii) career ambition.
A third aspect of the debate about women’s employment concerns the effects of the human resource (HR) practices designed to overcome gender inequality, for example, in health care institutions. Organizations have developed and introduced ‘family-responsive’ or ‘work−life balance’ policies, which are mainly intended to help employees manage the balance between paid work and other activities in life (Den Dulk et al., 2010; Dikkers et al., 2007). The common argument for implementing these arrangements is that more women would get into top positions if employers could be persuaded to adopt family-friendly work arrangements (Hakim, 2006). But, as Kossek et al. (2010) point out, although these work−life initiatives may enable employees to combine their work and care-giving responsibilities, they may also perpetuate stereotypes of ideal workers. Based on this line of reasoning, we will look in turn at bundles of gender-equality arrangements. ‘Bundle’ refers to a set of specific HR arrangements that logically fits together, such as grouping together all types of leave arrangements, because they are expected to have a similar relationship with the dependent variable, for instance (see, for example, Guest et al., 2004, on ‘bundling’ of HR arrangements). In this study, we ‘bundled’ gender-equality arrangements depending on which ‘ideal’ these initiatives implicitly reinforced. Studies into the effects of HR arrangements have yielded contradictory results when it comes to how effective these arrangements are in mitigating psychological distress among workers (e.g. Dikkers et al., 2007), and commitment (e.g. Lewis, 2001). To our knowledge, the relationship between gender-equality arrangements and career motivation – or what Hakim (2006) would call ‘lack of interest’ – on a person-centred level has not yet been studied empirically.
Based on the debates surrounding women’s employment presented above, this study focuses on how gender role prescriptions affect women physicians’ attempts to remain motivated in their career, and also the effect of organizational efforts to support these women. More specifically, our core research question in this study is as follows: how do the internalized role prescriptions for the ‘ideal worker’ and the ‘ideal mother’ influence career motivation among Dutch women physicians, and how do the gender-equality arrangements made by employing organizations affect these women physicians’ career motivation and the relationship between role prescriptions and career motivation? The first part of our research concerns how internalized role prescriptions for the ‘ideal worker’ and the ‘ideal mother’ influence career motivation among Dutch women physicians. Using goal-framing theory (Lindenberg and Frey, 1993), we will introduce a typology of four goal frames (Lindenberg and Steg, 2007). A goal frame refers here to the cognitive process through which goals govern or frame how people evaluate various aspects of a situation and which alternatives they consider (see next section for a further explanation of goal frames). On the basis of these goal frames, women go through a cognitive process − though perhaps not consciously − and internalize the role prescriptions associated with the ‘ideal’ mother and the ‘ideal’ physician in order to gain approval from their social environment. To answer the first part of our research question, we will test the relationship between four goal frames and career motivation among women physicians.
The second part of our research question focuses on the relationship between gender-equality arrangements and career motivation among women physicians. More particularly, we will investigate whether specific bundles of gender-equality arrangements, based on how rooted these are (implicitly) in gendered role prescriptions, affect career motivation differently. The third and final part of our research question addresses the extent to which gender-equality arrangements can alter the relationship between the goal frames of Dutch women physicians and their career motivation. Organizations often focus significant attention and resources on developing and implementing gender-equality arrangements. However, researchers have identified a possible unintended consequence of family-friendly arrangements, such as resistance and jealousy on the part of childless women and fathers (e.g. Bagilhole, 2006). We are interested in whether these gender-equality arrangements might also ‘backfire’ when it comes to career motivation among women physicians. For example, when gender-equality arrangements that emphasize the role prescription of the ‘ideal’ mother (such as part-time working) are offered, does this weaken the relationship between having a career goal frame (in which fulfilling the role of the ‘ideal’ physician is prioritized over that of the ‘ideal’ mother) and women physicians’ career motivation?
To summarize, this study introduces four goal frames that reflect how women physicians internalize conflicting gendered role prescriptions and investigates the relationship between these goal frames and their career motivation. Additionally, it examines the relationship between gender-equality arrangements – which are inspired by the same underlying ideals – and career motivation among women physicians, and whether these arrangements moderate the relationship between goal frames and career motivation.
Social roles and goal frames
Several theories have been used in studies on how societal role expectations affect men and women’s preferences and behaviour when it comes to combining work and family life. For example, the gender role theory (Eagly and Steffen, 1984), the boundary theory (e.g. Nippert-Eng, 1996) and the preference theory (Hakim, 2002) have often been used as theoretical frameworks (e.g. Tomlinson, 2006). These theories are useful in helping us to understand how the gendered allocation of the roles of breadwinner and homemaker, as well as personal preferences, lead to gender differences in skills and attitudes, in expectations and stereotypes, and, ultimately, in behaviour (King, 2008). Nevertheless, these theories implicitly cast working women either in the role of subjects who are influenced by gendered role stereotypes without questioning them critically or rejecting them (gender role theory), or as agents that consciously disregard societal norms and choose a certain lifestyle which is either home-centred, work-centred or adaptive (preference theory). Although the latter theories do at least acknowledge that women can combine work life with family life – for example, referred to as the ‘adapters’ – most of them bracket all women who combine work and family life into one group in which they are ‘not totally committed to work career’ (Hakim, 2006: 288). Alternatively, they make a distinction between ‘career primacy’ women, who are assumed to have an equal chance to men of reaching a top position, and ‘career-and-family’ women, who could do well for the organization in middle management positions (Schwartz, 1989). Other theories, such as boundary theory (e.g. Nippert-Eng, 1996) and the work/family border theory (Clark, 2000), focus on the strategies, principles and practices that people use to create, maintain and modify cultural categories, such as ‘home’ and ‘work’ (Nippert-Eng, 1996: 564). Although the notion of boundary work includes societal ideals regarding motherhood and the ideal worker when considering women’s labour market participation, boundary work theory seems to focus mainly on people’s strategies in combining work with private life, without questioning the underlying societal ideals regarding motherhood and work. For example, if a woman physician ‘integrates’ work and private life (Nippert-Eng, 1996: 567) by using a single diary for both work and private life, and by answering her work-related emails from home, she may still believe she should live up to the cultural norms of both the ideal mother (she is at home) and the ideal physician (who is always available and thus constantly checking her mail). Although these boundaries are permeable, the culturally defined ideals remain unchanged. The goal-framing theory of Lindenberg and Frey (1993) offers a theoretical basis on which to explain exactly how these unchanged ideals can explain career motivation. Individuals generally act within a restricted framework, which includes their internalized ideals, and act only on what they perceive their options and choices to be (Tomlinson, 2006). This framework, in turn, results in all sorts of ‘boundary work’ when dealing with complex situations, but the deeper conflict lies in the interplay of conflicting ideals and trying to achieve the goals that stem from these ideals. The cognitive process of ‘framing’ refers to the process by which people reorient their thinking on an issue (e.g. Chong and Druckman, 2007). An individual’s perception of a situation is an important determiner of how a person will behave in that situation. Lindenberg’s goal-framing theory (Lindenberg and Frey, 1993) has provided insight into this process by identifying people’s goals as the main determiners of a person’s perception of a situation (Lindenberg and Steg, 2007). The notion at the core of the goals-framing theory is that goals govern or frame what people pay attention to, which knowledge and attitudes are the most accessible cognitively, how people evaluate various aspects of a situation and which alternatives are considered. Behaviour results mainly from a combination of several goals which may, or may not, be compatible (Keizer et al., 2008; Lindenberg and Steg, 2007). Lindenberg and Frey (1993) state that the need for social approval and physical well-being are universal, but that human beings also have normative goal frames that activate a range of instrumental (sub-)goals associated with appropriateness; these relate to what people think they ought to do in particular situations. What people ought to do, according to gendered sex role prescriptions, differs between men and women (Bem, 1974) and is inextricably bound up with situations in the work and home domain. Such gendered sex-role prescriptions can be seen as normative goal frames. In this study, these goals are represented by the cultural norms relating to the ‘ideal’ mother and the ‘ideal’ physician. Living up to these cultural norms, which underpin cultural classifications such as ‘work’ and ‘home’, is one way of gaining social approval (Baaijens et al., 2005). For women physicians, there are two ways to gain approval. One way is via the private domain by being a good mother, which in the Netherlands means staying at home or working a limited number of hours (Mills and Täht, 2010). The other way is via the public domain by being an ‘ideal’ physician who is always available, dedicated to work and physically strong enough to work long hours. These instrumental goals can be contradictory and therefore place competing demands on one’s time and energy. For instance, if one’s dominant goal (‘frame’) is to be an excellent physician, one will focus one’s attention and other cognitive processes on achieving this goal (framing), and other demands on one’s time will come to appear as distractions and of secondary importance. However, changing circumstances can alter the priority attached to these instrumental goals. For example, when a woman becomes a mother − a changing circumstance − and wishes to gain social approval by being a good mother according to traditional standards, this new goal may challenge her career goals. To analyse the framing processes that take place in women physicians, we introduce a typology which reflects four goal frames that are based on women physicians’ own internalized notions of the ideal physician and the ideal mother (see Figure 1).
Typology of goal frames.
The horizontal dimension represents the notion of a good mother among women physicians, and ranges from the traditional notion of the ‘good’ mother to a non-traditional view. What is considered to be ‘ideal’ or good parenting differs considerably between mothers and fathers in the Netherlands. For fathers with children under the age of four, a four-day working week is considered ‘good parenting’, yet if the children are older, a full-time working week is still considered the ideal (SCP, 2010). Despite the family-friendly policy that has long been adopted by the Dutch government when it comes to part-time working, gender inequality in organizations persists, stemming partly from the traditional motherhood ideology, according to which a woman’s main priority in life is her care-taking responsibilities for husband and children (Arendell, 2000; Hays, 1996). A Dutch woman would be perceived as having a traditional view of motherhood if she believes good mothering involves not working outside the home more than three days a week, and if she sees herself as the best caretaker for her children and household and takes primary responsibility for that (Bagilhole, 2006; Hays, 1996; SCP, 2010). Furthermore, while the number of paid hours worked in the labour market by Dutch women has increased over the past several years, public acceptance of mothers working full-time has not increased (SCP, 2010). One could say that women who score high on the traditional view of motherhood are those that are likely to be among Hakim’s home-centred or family-centred women, who prefer to prioritize private and family life and who tend to avoid paid work unless financially necessary (Hakim, 2006). In our study, having a non-traditional view of motherhood only means that these women do not agree with, or see themselves according to, the traditional view of women’s roles.
The vertical dimension in the typology represents women physicians’ notion of the ideal physician and ranges from a traditional role identity of a medical physician (e.g. ‘I believe I am a good physician when I am always available and routinely work overtime, fully devoted to medicine, willing to relocate in order to pursue my medical career and not distracted by other responsibilities in life’) to a non-traditional role-identity. Again, we are interested in the extent to which a woman physician has internalized societal norms regarding the ideal physician (worker). The two axes form four quadrants, which together represent a typology consisting of four different combinations of how women physicians can combine their notions of the ideal physician with those of the ideal mother. For instance, a Dutch woman physician with a traditional care goal frame believes that she has to be at home with her children at least four days a week and work only three days outside the home (SCP, 2008), but she also believes that she does not have to work overtime routinely in order to be a good medical physician. She has thus internalized gendered role prescriptions on motherhood, yet has not internalized traditional role prescriptions on being an ideal physician. In this example, applying goal-framing theory, the woman physician prioritizes her role as a mother over her role as a medical physician in order to receive social approval. A woman physician with a career goal frame does the exact opposite: she has internalized and prioritized role prescriptions of the traditional ‘ideal’ physician to obtain social approval for her role as a medical physician, but she disagrees with normative role prescriptions regarding motherhood. When it comes to ‘boundary work’, both women physicians with a care goal frame and those with a career goal frame are more likely to separate the two domains, as according to the traditional prescriptions only one domain can be prioritized.
Rather than prioritizing one role over the other, women physicians can also decide to internalize both role prescriptions, which results in switching goal frames, depending on the environment they are in, and the approval they expect to receive for these roles. Women physicians with switching goal frames find both the roles and the social approval associated with these roles equally important, and will switch between goal frames depending on time and place. Johnston and Swanson (2007) call this a coping strategy in which the contradiction is neutralized by satisfying both competing needs to some extent, but without fully realizing either. Women with switching goal frames integrate work and private life to the extent that is permissible in each of the two environments, in order to be able to live up to both ideals. Women physicians who internalize neither the traditional view of the ideal physician nor the traditional view of the ideal mother are defined as having non-traditional goal frames. Women physicians with non-traditional goal frames construct their own concepts of good motherhood and being a good physician in the way that best suits their abilities and their current private circumstances. For example, a woman physician may focus on her work as a medical specialist because she needs professional challenges or because she feels the need to educate herself, but she may reject the ideal worker norms completely. In that sense, one could argue that women with non-traditional goal frames represent the ‘transformed deal’ (Baruch, 2004: 66) that has also been noted in the post-modern stance of career theory (e.g. Collin, 2006). From this point of view, success refers to an inner feeling of achievement, rather than the more traditional interpretation, where success is represented by progress up the hierarchical ladder.
Career motivation
Many authors acknowledge that, in research, theory and practice alike, the concept of career is undergoing a transformation, with the narrowly defined concept of a career moving towards a broader definition. This broader definition takes into account the personal and social context of an individual’s professional career, and includes periods of activities undertaken ‘in between jobs’, or activities such as unpaid work and family life matters (e.g. Collin, 2006). The definition of what constitutes ‘career’ is inevitably related to definitions of career motivation. In behavioural psychology, there are numerous concepts associated with career motivation, such as ‘career aspiration’ (Rainy and Borders, 1997), ‘achievement ambition’ and ‘career commitment’ (Murphy and Alexander, 2000). However, many of the existing theoretical concepts are one-dimensional, dealing with only one or a few aspects of an individual’s motivation to succeed in their career, such as their need for success or achievement (Mitchell and Daniels, 2003). London (1983, 1997) conceptualized career motivation as a multidimensional construct internal to the individual, which is reflected in that individual’s attitudes, decisions and behaviour regarding his or her career. In this study, we build on London’s conceptualization and distinguish three dimensions of career motivation: career centrality, which is the importance of career in one’s life; career insight, which is the degree to which one makes strategic plans to obtain certain career goals; and career ambition, which is the will to achieve a higher position in one’s field. Whereas the first of these dimensions reflects the way one sees one’s career in a wider context and thus refers to a broader definition of career – for example, the importance of career compared to other aspects of life − the latter two dimensions rest on the assumption that career is a linear process that can be managed and that career ‘success’ relates to achieving a higher position, more power and, consequently, a higher salary. Regardless of whether this should be taken as an indicator of success in one’s career (Arthur et al., 2005), assessing levels of career insight and career ambition reflects the career ladder and career incentives that many women physicians are confronted with within their organizational and professional context (Van den Brink, 2011).
By combining goal-framing theory (Lindenberg and Frey, 1993) with the construct of career motivation, we seek to provide a new theoretical explanation for the interaction between, on the one hand, internalized gendered role expectations that stem from the need for social approval and, on the other hand, women physicians’ career motivation. As indicated, the four quadrants represent four different cognitive goal frames. We have reason to believe that each goal frame would have a different effect on career motivation, a supposition which we will now explain further.
We expect that women physicians with different goal frames will also have different levels of career motivation (career centrality, career insight and career ambition). We will focus particularly on the possible effect of switching goal frames on career motivation compared to the other goal frames. As tempting as this ‘coping strategy’ may appear − it seems to offer the possibility of living up to both sets of gendered role prescriptions at the same time − it runs the risk of frustrating women physicians through a lack of social approval in either domain (Stryker and Burke, 2000) or hastening the onset of a burn-out (Allen et al., 2000). It is often suggested that the feeling of falling short of expectations in both domains is what leads women to ‘opt out’ (e.g. Hewlett and Luce, 2005). As such, it may ultimately result in lower levels of career motivation. Our first hypothesis regarding the relationship between women physicians’ goal frames and their career motivation is therefore: Hypothesis 1: Women physicians with switching goal frames have higher levels of career motivation (career centrality, career insight and career ambition) than women physicians with a care goal frame, but lower levels of career motivation than women with a career goal frame.
Gender-equality arrangements
Nowadays, many hospitals and medical partnerships offer family-friendly gender-equality arrangements (which are sometimes enshrined in the terms of their collective labour agreements; Den Dulk et al., 2010), especially in more feminized specialities, such as paediatrics and family practices (Pas et al., 2011b). These arrangements are ostensibly gender-neutral and intended for men and women alike, but in practice they tend to focus implicitly on women, creating a so-called ‘mommy track’ with childcare provisions and opportunities for leave (e.g. Mescher et al., 2010). The real effect of gender-equality arrangements is disputed, however. The existence of family-friendly policies does not seem to make a major difference to gender equality on the labour market, and in fact they are often cited as exacerbating the problem of inequality (Hakim, 2006). Apparently, take-up rates for these arrangements are limited and gender-differentiated, owing to fear of career hindrance (e.g. Bagilhole, 2006; Thompson et al., 1999).
Several classifications of gender-equality arrangements have been described in feminist theory literature (Calás and Smircich, 2006; Ely and Meyerson, 2000). Our categorization differs from other research in two ways. First, based on Nippert-Eng’s (1996) boundary theory and research on ‘boundary work’ in organizations by means of work−life policies (Mescher, 2011), we distinguish between those arrangements that keep work and private life separate by maintaining strict boundaries, such as leave policies and part-time work (here, we refer to these as ‘ideal-worker/ideal-mother arrangements’), and those arrangements that allow for the integration of work and private life by blurring the boundaries between the two domains, such as flexible start and end times and teleworking (we refer to these as ‘revising work-culture arrangements’). Second, we divide the former category into ideal-worker arrangements and ideal-mother arrangements, depending on which normative ideals predominate. For example, an arrangement which offers leave over and above statutory levels will allow women physicians to live up to the motherhood ideology, whereby women with infants stay at home for as long as possible.
Ideal-mother arrangements are, for example, the opportunity to work part-time or job share, and additional leave arrangements. All these measures are basically meant to enable women to spend more time at home with their children, and thus conform to the norm of the ideal mother. By reducing structural barriers to being at home more often, the aim of these arrangements is to create equal opportunities. Although these arrangements could ease the tension between work and family in women’s lives, the work−family balance remains a ‘woman’s problem’ and the arrangements run the risk of being counter-productive (Ely and Meyerson, 2000). Our second category, ideal-worker arrangements, includes measures to make women better equipped to meet gendered ideal-worker norms, such as extra management development training, mentoring and coaching. Women’s networks and positive gender discrimination policies (i.e. bonuses for appointing women to higher positions) also fall into this category. Feminists have characterized these arrangements as either ‘fixing’ or ‘blaming’ women, as women are implicitly ‘blamed’ for lacking motivation or for lacking the required networking skills (Ely and Meyerson, 2000). Other types of ideal-worker arrangements, such as women’s leadership training programmes, are associated with a ‘value the feminine’ approach, whereby women’s skills are still identified as different from those of men, and are recognized and valued as such. Our third category, revising work-culture arrangements, includes arrangements that increase organizational flexibility, such as offering flexible working hours, teleworking and giving employees a say in scheduling. The main difference between ideal-mother arrangements and revising work-culture arrangements is that the latter are often not presented as a strategy to counter gender inequality, but are introduced to tackle commuting problems and reduce overhead costs or to allow employees to work in isolation in order to get more work done (Peters et al., 2004).
Several researchers have studied the effectiveness of family-friendly arrangements on work−family conflict, work−home interference, and phenomena such as turnover and absenteeism (Allen, 2001; Thompson et al., 1999). To our knowledge, apart from studies on the effect of part-time working on women’s careers (e.g. Roman et al., 2004), hardly any empirical research has been done into the effects of gender-equality arrangements on the career motivation of women physicians. Additionally, we are interested in whether a combination of several arrangements based on organizations’ strategies to deal with gender inequality will help us to understand the effect of these arrangements on career motivation. As ideal-mother arrangements implicitly refer to women physicians’ roles as (potential) mothers, we would expect these arrangements to correlate negatively with career motivation. Furthermore, based on goal-framing theory, one would expect women with a particular goal frame to be more susceptible to a particular type of arrangement. However, ideal-worker arrangements and, more especially, revising work-culture arrangements can be expected to correlate positively with the career motivation of women physicians, as by offering these types of arrangements, the organization signals that they value women and are willing to invest in them by offering training and mentoring. By doing this, they draw on women physicians’ own career goals, thereby presenting these goals as more feasible. Our second set of hypotheses is therefore: Hypothesis 2a: Ideal-mother arrangements correlate negatively with the career motivation of women physicians (career centrality, career insight and career ambition). Hypothesis 2b: Ideal-worker arrangements and, more especially, revising work-culture arrangements correlate positively with the career motivation of women physicians.
Moderating model
By offering gender-equality arrangements, organizations are seeking to challenge societal norms and beliefs about how a good mother and a good physician are supposed to behave. Regardless of the direct effect of these arrangements on the career motivation of women physicians, the question for organizations is whether the effects of goal frames on career motivation are undermined or reinforced by offering gender-equality arrangements. Changing environmental circumstances can alter the salience of roles and goal frames, leading to different behaviour (Lindenberg and Frey, 1993; Stryker and Burke, 2000).
Our third set of hypotheses concerns the moderating effect of gender-equality arrangements on the relationship between goal frames and career motivation. We expect to see gender-equality arrangements having both positive and counter-effective moderating effects on the relationship between goal frames and career motivation. Researchers have addressed the possible ‘backlash’ against family-friendly arrangements, such as the resistance and jealousy of childless women and fathers (e.g. Bagilhole, 2006). But in addition to this, we are interested in whether these arrangements can also ‘backfire’ at the individual level. For instance, offering ideal-mother arrangements may have the intended effect for a woman physician with a care goal, and encourage her to combine her career with caretaking responsibilities; however, for women physicians with a career goal frame or with switching goal frames, ideal-mother arrangements may have a different effect. As these women believe they have to live up to the norms of the ideal worker, signals from the organization that emphasize their role as a mother may serve to undermine their native career motivation. Ideal-worker arrangements also risk being counterproductive in a similar way. Although ideal-worker arrangements can be expected to strengthen the positive relationship between career goal frame and switching goal frames on career motivation, they may also discourage women physicians with a care goal frame because they signal that extra effort is required – such as participating in training programmes – to attain a higher position in the organization. As revising work-culture arrangements do not seem to be based on gendered societal ideals about the ideal mother or the ideal physician, we expect that these arrangements will reinforce the positive relationship between both a career goal frame or switching goal frames and career motivation, and that they will temper the negative relationship between a care goal frame and career motivation: Hypothesis 3a: Ideal-mother arrangements temper the negative relationship between a care goal frame and career motivation of women physicians, yet they also weaken the positive relationship between a career goal frame or switching goal frames and career motivation. Hypothesis 3b: Ideal-worker arrangements strengthen the positive relationship between a career goal frame or switching goal frames and the career motivation of women physicians, yet they also reinforce the negative relationship between a care goal frame and career motivation. Hypothesis 3c: Revising work-culture arrangements strengthen the positive relationship between a career goal frame or switching goal frames and the career motivation of women physicians, and they also temper the negative relationship between a care goal frame and career motivation.
Method
Participants and procedures
To test our hypotheses, we conducted a web survey in the summer of 2008 entitled ‘Physician and career’. The survey was sent to a representative sample of 3380 women physicians and assistant physicians aged 25 to 50. Year of birth and speciality were used as criteria to select physicians from the official database of medical registration commissions in the Netherlands (KNMG), in which all Dutch physicians are registered. Respondents were invited to participate by letter and then asked to visit a website to answer an online questionnaire. To do this, they used a personal log-in code provided in the letter. After three months and three reminders by mail and email, 1070 women physicians had completed the survey, giving an AAPOR (2008) response rate of 32 percent. This is a fairly good response rate, given that the average response rate for web surveys in the Netherlands lies between 25 percent and 45 percent (De Leeuw and De Heer, 2001). The sample that completed the survey was representative of the population of women physicians in the Netherlands as a whole, in terms of both age and speciality (e.g. Van der Velden et al., 2008).
Descriptive analyses were conducted to identify the career characteristics (i.e. medical speciality, years of work experience, career motivation) and personal/household characteristics (age, partner, children) of the women physicians. T-tests were used to examine whether women physicians with different goal frames varied in their perceptions of the work arrangements offered and career motivation, type of speciality (medical specialists, surgeons or generalists, including insurance employees), whether or not they live and work ‘in house’ (also referred to as residents (US) or Speciality Registrar (UK)), years of work experience, spouse and children. Hierarchical regression was used to test the direct relationships between goal frames and gender-equality arrangements and career motivation and to test the moderating effects of gender-equality arrangements on the relationship between goal frames and career motivation. The regression analyses were conducted after list-wise deletion of missing values. The number of cases included in the analyses was N = 715. This subsample was also representative of the target population.
Measures
Career motivation. The career motivation items were based on the career motivation inventory (London, 1997), the Emancipation monitor (SCP, 2008) and a proactive coping scale (Ouwehand, 2005). Using principal axis factoring (PAF, rotation oblimin), the three sub-dimensions of career centrality, career insight and career ambition were identified. Tests on the scales employed in this study have been reported previously (Pas et al., 2011). However, all dimensions of career motivation were retested for this study. Career centrality consisted of six items, with a Cronbach alpha of .85, and included items such as ‘I am very focused on my career’ and ‘I get the most satisfaction in life out of my career’. Career insight had three items (Cronbach’s alpha = .81), such as ‘I have formulated clear career goals for myself’ and ‘I have developed a strategy to achieve my career goals’. The third dimension, career ambition, consisted of four items (Cronbach’s alpha = .78). Examples are: ‘I want to obtain a top position in my field’ and ‘I want to obtain a position with a higher salary level’. The items were measured on a five-point Likert scale ranging from 1 for ‘totally disagree’ to 5 for ‘totally agree’.
Goal frames. As explained in the theoretical framework, we constructed four goal frames based on respondents’ scores on two scales: the ‘ideal mother scale’ and the ‘ideal physician scale’. The ideal mother scale was tested and used in another study (Pas et al., 2011), and consisted of eight items representing a traditional ideology of motherhood, for example, on how to allocate time between children and activities outside the home (e.g. paid work). Examples of these items are: ‘I want to share a lot of time with my child/children’ and ‘I find it important to share breakfast and dinner together with my children during the (work) week’. Internal reliability was assessed using Cronbach’s alpha (α = .77). Women physicians with no children were asked to respond as if they had children. They were presented with hypothetical items, such as ‘I believe I would want to share a lot of time with my child/children’.
The ideal physician scale was designed specifically for this study on the basis of qualitative research into the characteristics of the ideal worker (Kelly et al., 2010) and the ideal physician (Price et al., 1971). The scale used in this study was subjected to principal axis factor analysis. The results of this analysis and the reliability test (Cronbach’s alpha) are reported in Table 1. The ideal physician scale consisted of five items concerning the respondents’ views of being an ideal physician. Cronbach’s alpha for this scale was .78.
Ideal physician inventory (N = 1070).
On the basis of these scales, we constructed the following four goal frames: career goal frame, care goal frame, switching goal frames and non-traditional goal frames. To construct these goal frames, we derived the mean scores of the respondents on the two scales. For the ideal mother scale, the mean score was 3.19 and for the ideal physician scale the mean score was 2.92. Women who had a below-average score on the ideal mother scale (<3.19) and an above-average score on the ideal physician scale (>2.92) were categorized as having a ‘career goal’. Women with a below-average score on the ideal physician scale (<2.92) and an above-average score on the ideal mother scale (>3.19) were categorized as having a ‘care goal frame’. Women who had an above-average score on the ideal physician scale (>2.92) and an above-average score on the ideal mother scale (>3.19) were categorized as women with ‘switching goal frames’. Finally, women with a below-average score on the ideal physician scale (<2.92) and a below-average score on the ideal mother scale (<3.19) were categorized as women with ‘non-traditional goal frames’.
To control for the potential effects of response biases, we applied different response formats to measure the three dimensions of career motivation (career centrality, career insight and career ambition) (five-point Likert scale with a ‘don’t know’ option), the ideal mother scale (five-point Likert scale with numerical entry) and the ideal physician scale (five-point Likert scale with ‘not applicable’ option). Also, on the basis of qualitative research on the formulation of the items and a pilot study, we improved the scale items by reducing item ambiguity, social desirability and demand characteristics, and we deleted equivalent and irrelevant items. Because we used tested and widely used scales to measure most of the concepts, we were reluctant to alter the scale formats and scale values in order to preserve the original scale validities.
We also conducted the Harman one-factor test (Harman, 1967), as described by Podsakoff et al. (2003), to test for common method bias. As an unrotated principal component factor analysis of all relevant variables resulted in five components, with the first component accounting for only 22.3 percent of the variance, our findings were not much affected by the problem of common method variance.
Gender-equality arrangements offered. Respondents were asked whether the listed arrangements were offered, according to them (0 = not offered; 1 = offered; and 2 = don’t know [later coded as missing value]). We then divided gender-equality arrangements into three subcategories. The first subcategory, ideal-mother arrangements, included the following: job sharing, a mothers’ room to express milk, ‘mother contracts’ (the option of working only during school hours), the option of working part-time, doing a part-time training programme, extra maternity, paternity, adoption and care leave (i.e. above the statutory requirements), career breaks and leave-saving options. The category of ideal-mother arrangements was used as a continuous variable with a minimum of zero arrangements and a maximum of 11 arrangements. On average, 4.13 arrangements were offered, according to the women physicians respondents. The subcategory of ideal-worker arrangements was also a continuous variable, with a minimum of zero and a maximum of 10 arrangements. It included the following: on-site childcare, childcare arrangements in other care facilities, childcare at home (nanny/nurse), financial support for childcare, mentoring, coaching, work−life balance training, tenure-track/moving-up programmes designed especially for women, women’s networks and sabbatical leaves. The average number of ideal-worker arrangements offered, according to the respondents, was 1.85. The category of revising work-culture arrangements was constructed as a continuous variable with a minimum of zero and a maximum of three arrangements. The three arrangements were: flexible start- and end-times for the working day, the option of working from home (teleworking) and having a say in scheduling. The average number of arrangements offered was 1.02.
Interaction effects
To examine the interaction between goal frames and the arrangements provided, we included interaction variables that related to a particular goal frame and a particular arrangement. For example, the interaction effect of ideal-mother arrangements on the relationship between a care goal frame and career centrality was obtained by creating a variable named care goal frame × ideal-mother arrangements offered. To avoid potential collinearity, all independent variables, except for the dummies, were mean centred. Subsequently, 12 interaction variables were created (3 types of arrangements × 4 goal frames). In a first step, the hypothesized main relationships were included in the analysis. In a second step, the interaction variables were added to test for moderation.
Control variables
All the parameters estimated were controlled for the respondent’s years of work experience as a physician, 2 speciality (0 = general practitioners; 1 = non-surgical specialists; 2 = surgical speciality), having a partner (0 = no, 1 = yes), presence of a child/children (0 = no, 1 = yes), respondent’s age and whether or not they were a trainee (0 = no, 1 = yes).
Results
To gain a better view of the characteristics of the women physicians with different goal frames, we first conducted descriptive analyses for all categories (see Table 2). Perhaps the most striking feature of the variations among women physicians was that those with switching goal frames tended to be younger (average age 33.9 years) than those with a career goal frame (35.9), a care goal frame (34.2) or non-traditional goal frames (37.2). They also had less work experience (average 7.7 years) and 45 percent of them had children. Most women physicians had a partner. Women physicians with a career goal frame and switching goal frames had the highest levels of career motivation, although the differences between women physicians with non-traditional goal frames and women with a care goal frame were not large and not always significant. As for speciality and the goal frame categories, we found no significant differences, although those working in hospital settings (physicians and surgeons) seemed to be more likely to have either a career goal frame (43 percent physicians; 9 percent surgeons) or switching goal frames (46 percent physicians; 10 percent surgeons). When it came to gender-equality arrangements, we did not find any significant differences between the four goal frame categories.
Descriptives (mean and standard deviations) and T-tests of women physicians for four goal frames.
SD = Standard deviation; *** P<0.001; ** P<0.01; * P<0.05, n.s. = ‘not significant’.
Step 1 concerned the first part of our research question addressing the relationship between goal frames and career motivation. In our first hypothesis (H1), we predicted that women physicians with switching goal frames would have higher levels of career motivation than women physicians with a care goal frame, yet lower levels of career motivation than those with a career goal frame. The results presented in Table 3 (column Career centrality, Step 1) show that women physicians with switching goal frames had higher levels of career centrality than women physicians with a care goal frame (β = –.17, p < 0.001, one-tailed). Turning to career insight, the first step in the regression analysis revealed that women physicians with switching goal frames did not differ significantly in their level of career insight from women physicians who did not have switching goal frames (see Table 3, column Career insight, Step 1). The results also showed that women physicians with switching goal frames had a higher level of career ambition than women physicians with a care goal frame (β = –.14, p < 0.01, one-tailed) (see Table 3, Career ambition, Step 1/2), yet their career ambition did not differ significantly from women physicians with a career goal frame. Our first hypothesis (H1) was thus only partly supported by the data. Women physicians with switching goal frames did not differ in their level of career insight compared with others, nor did they have lower levels of career ambition than those with a career goal frame, as we hypothesized.
Hierarchical regressions for types of goal frames, offered gender-equality arrangements and control variables on career motivation.
p < 0.001; ** p < 0.01; * p < 0.05 (one-tailed); n.s. not significant at the 0.05 level.
Switching goal frames served as reference category.
For sake of clarity, only significant interaction effects are reported.
General practitioners served as reference category; for Career ambition, Step 2 to stepwise deletion of all non-significant interaction variables, eventually had similar results to Step 1.
We now come to the second part of our research question. In our second set of hypotheses, we conjectured that ideal-mother arrangements would correlate negatively with career motivation in women physicians (H2a), whereas ideal-worker arrangements and revising work-culture arrangements would correlate positively (H2b). We found no significant relationship between ideal-mother arrangements and the career motivation of women physicians (see Table 3, Career centrality, Career insight and Career ambition, Step 1). The first part of our second hypothesis (H2a) was therefore not corroborated by the data. However, the results did reveal that ideal-worker arrangements correlate positively – albeit fairly marginally – with women physicians’ career motivation (β = .07, p < 0.05, one-tailed for Career centrality, Step 1; β = .08, p < 0.05, one-tailed for Career insight, Step 1; and β = .10, p < 0.01, one-tailed for Career ambition, Step 1). Offering revising work-culture arrangements had no significant relationship with women physicians’ levels of career centrality and career insight, but they did have a significant positive relationship with women physicians’ levels of career ambition (β = .11, p < 0.01, one-tailed, Step 1). Overall, then, only one part of our second hypothesis (H2b) was corroborated by our data.
Finally, we turn to the outcomes regarding the third part of our research question. In our third set of hypotheses (H3a; H3b and H3c), we predicted that the relationship between women physicians’ goal frames and career motivation would be both positively and negatively moderated by gender-equality arrangements. In H3a, we posited that ideal-mother arrangements would temper the assumed negative relationship between a care goal frame and career motivation; however, this type of arrangement would also temper the assumed positive relationship between a career goal frame and switching goal frames and career motivation. As Table 3 shows, we found that ideal-mother arrangements only tempered the negative relationship between a care goal frame and career insight (β = .11, p < 0.01 one-tailed; see Career insight, Step 2) (see also Figure 2). Our hypothesis H3a is thus only weakly supported. Offering ideal-mother arrangements does not negatively affect the positive relationships between a career goal frame and switching goal frames and career motivation.
Interaction effect of ideal-mother arrangements on relationship between care goal frame and career insight.
In H3b, we anticipated that ideal-worker arrangements would strengthen the positive relationship between a career goal frame and switching goal frames and career motivation. However, we also predicted that they would strengthen the negative relationship between a care goal frame and career motivation. Table 3 shows that we only found a significant yet marginal moderation of ideal-worker arrangements on the positive relationship between a career goal frame and career centrality (β = .07, p < 0.05 one-tailed; see Career centrality, Step 2; see also Figure 3). It appears that respondents with a career goal frame had higher levels of career centrality when more ideal-worker arrangements are offered. We did not find that offering ideal-worker arrangements demotivated women physicians with a care goal frame. Although we found that ideal-worker arrangements strengthened the positive relationship between a career goal frame and career centrality, the hypothesis H3b was only partly supported.
Interaction effect of ideal-worker arrangements on relationship between career goal frame and career centrality.
The final part of our third hypothesis (H3c) predicted that revising work-culture arrangements would strengthen the positive relationship between a career goal frame or switching goal frames with career motivation, and would also temper the negative relationship of a care goal frame with career motivation. We did not find any significant interaction effects to support this hypothesis (H3c). Overall, we can conclude that offering gender-equality arrangements affects the relationship between goal frames and career motivation, but only marginally and slightly more for those with salient goal frames than those with switching goal-frames.
With respect to the control variables, we did not find any significant relationship between having a partner and career motivation. However, we found that the respondent’s years of work experience had a negative relationship with career centrality (β = –.24, p < 0.01) and career insight (β = –.22, p < 0.05); that the respondent’s age had a positive relationship with career centrality (β = .20, p < 0.05) and career insight (β = .17, p < 0.05); and that the presence of children had a negative relationship with career centrality (β = –.01, p < 0.01) and career ambition (β = –.08, p < 0.05). Also, medical speciality and being employed by a hospital were highly predictive for all dimensions of career motivation among our respondents. Specifically, physicians and surgeons had higher levels of career motivation than did general practitioners (e.g. family practitioners, insurance physicians).
Discussion
It seems that women physicians in the Netherlands are ‘caught between a rock and a hard place’ − working part-time because of social pressure to be a good mother, but working long hours and preferably full-time in order to meet expectations of being a good physician. The problem of the competing demands of ‘work’ and ‘life’ was noted by many researchers (e.g. Clark, 2000; Hakim, 2006; Nippert-Eng, 1996). However, despite this awareness and the gender-equality arrangements offered to these women to support them in this balancing act, the underlying societal ideologies do not appear to have changed and are being perpetuated through prevailing social practices, both at home and at work. Using goal-framing theory (Lindenberg and Frey, 1993), we have therefore introduced a typology of four goal frames through which we have investigated the cognitive aerobatics that women physicians have to go through when prioritizing or combining these conflicting ideals.
This article contributes to the academic and public debate on women’s labour force participation and career development in three ways. The first part of our research concerns how internalized role prescriptions for the ‘ideal worker’ and the ‘ideal mother’ influences career motivation among Dutch women physicians. From our results, we can conclude that women physicians’ goal frames do indeed affect their career motivation. Unsurprisingly, our results confirm the general idea that women physicians with a care goal frame have lower levels of career centrality and career ambition than those with switching goal frames. However, our results reveal that women physicians with switching goal frames are, in fact, as career motivated as women with a career goal frame. This contrasts earlier studies which seem unquestioningly to have adopted the underlying belief that ‘superwomen’ – those who want to do it all, or at least try to do it all – are less motivated in their career than those who focus solely on their career (e.g. Hakim, 2006; Schwartz, 1989). So far, researchers have acknowledged that women are confronted with, and disadvantaged by, the ideal worker norm because it is based on masculine traits (e.g. Acker, 1992). We believe that previous research may have overlooked the possibly ‘positive’ consequence of the fact that the ideal worker norm can also be internalized by women and that some women do seek to live up to this norm. Even though living up to the ideal worker norm usually puts them in an uphill fight, our study reveals how it may counterbalance the effect of the ideal mother norm when it comes to career motivation among women physicians. Moreover, so far studies on women’s ‘boundary work’ (e.g. Mills and Täht, 2010; Nippert-Eng, 1996) focused on how women – often on a daily basis – manage combining career with family life by either separation or integration of work and family life. Goal-framing theory (e.g. Lindenberg and Frey, 1993) has provided us with the underlying reason why women physicians do ‘boundary work’, which is to gain social approval from both the private and the public domain for their roles as mothers and physicians. This study has also revealed how these cognitive acrobatics in turn affect women physicians’ career motivation. In doing so, we have moved beyond stigmatizing women ‘who want to do it all’ as less focused and thus less motivated than those who (claim to) prioritize their career clearly over caretaking responsibilities.
The second part of our research question dealt with the relationship between gender-equality arrangements and career motivation among women physicians. Hitherto, other studies have revealed that these arrangements correlate with employees’ experiences of work−family conflict (e.g. Kelly et al., 2011) or women’s labour market participation (e.g. Himmelweit and Sigala, 2004). It could be argued that most gender-equality arrangements – at least in the Netherlands – were originally designed to persuade women to work more hours (e.g. SCP, 2009), leaving gender equality policy to the responsibility of single organizations. However, labour market participation no longer appears to be the central issue for women professionals such as women physicians. Although they work just as many hours as their male counterparts, they are still greatly underrepresented in the higher echelons of medical academia (Van Doorne-Huiskes and Van Beek, 2009). We have therefore focused on the relationship between gender-equality arrangements and women physicians’ career motivation. To our surprise, however, we found that ideal-mother arrangements had no significant (direct) relationship with career motivation. For example, although offering the option of a part-time contract or extra leave might benefit working women in terms of work−life balance (e.g. Thompson et al., 1999), such options did not seem to affect career motivation among women physicians. Ideal-worker arrangements, however, correlated positively with all three dimensions of career motivation, albeit only weakly. Revising work-culture arrangements, as the least gendered bundle of equality arrangements, appeared to be the most likely to improve women physicians’ career ambition. Nevertheless, further research is needed on how these revising work-culture arrangements are implemented and used in practice, particularly as other researchers have expressed their concerns that this type of flexibility (e.g. working from home) is a double-edged sword that may lead to a situation in which work never ends (Kelliher and Anderson, 2010) – once again feeding into ideal-worker norms. Overall, although we found significant positive relationships between gender-equality arrangements and career motivation among women physicians, the effect sizes were small. Considering the amount of attention, energy and resources that are channelled into gender-equality arrangements in the (Dutch) healthcare sector, which is already overrun with cost-containment strategies, the outcomes regarding the relationship between offering gender-equality arrangements and women physicians’ career motivation is somewhat disappointing. It could be that gender-equality arrangements only serve as a ‘hygiene factor’ (Herzberg, 1966): not offering gender-equality arrangements might discourage women from striving to achieve their career goals, but offering them does not imply higher levels of career motivation.
The third and final part of our research question addressed the extent to which gender-equality arrangements alter the relationship between women physicians’ goal frames and their career motivation. Our findings reveal that both ideal-mother arrangements and ideal-worker arrangements mainly seem to support those women physicians who prioritize one goal frame over the other. Perhaps not surprisingly, those with a care goal frame seem to be more receptive to, and supported by, the offered ideal-mother arrangements and those with a career goal frame seem to be particularly supported by ideal-worker arrangements. According to goal-framing theory, this could be the result of having a particular goal frame, which results in being more susceptible for information and cues that fit with one’s primary goal: either one’s career or one’s caretaking responsibilities. Contrary to our hypotheses, these arrangements are not ‘backfiring’ by negatively affecting the positive relationship between a career goal frame or switching goal frames and career motivation. Revising work-culture arrangements did not moderate the relationship between women physicians’ goal frames and their career motivation. More important, however, is the finding that women physicians with switching goal frames were not specifically supported by any of the gender-equality arrangements. Although they appear to be only marginally supported by ideal-worker arrangements and revising work-culture arrangements (as were women with a career goal frame); based on the results from this study they seem to be the ones that benefit the least from gender-equality arrangements overall.
All in all, it seems that gender-equality arrangements struggle to ‘protect’ women’s career motivation from the ubiquitous notion of the ideal mother, who prioritizes caretaking responsibilities over career, and that of the ideal worker, which still requires all workers to work overtime and be (or at least seem) available and dedicated team members.
Potential limitations
Inevitably, this study has a few shortcomings which should be addressed. First, although we found significant relationships between goal frames, gender-equality arrangements and career motivation among Dutch women physicians, some of these effects were rather weak. Moreover, the speciality in which a woman medical doctor is employed accounts for a considerable amount of variation in career motivation. This would suggest that more research is needed. For example, more insight is needed into the specific cultures in the different specialities and why family practitioners and medical doctors who work for insurance companies are less motivated to strive for their career than those working in a hospital setting. One explanation could be that women general practitioners believe they have fewer career prospects (in terms of upward mobility) than those working in hospitals. This would imply, in turn, that, according to some women physicians, career advancement is still interpreted in the narrow, traditional way, whereby climbing the hierarchical ladder is what constitutes career success.
Second, another important limitation is the cross-sectional research design of this study. Our descriptive results suggest that respondents’ age seemed to play an important role in this study, which may suggest that goal frames reflect ‘phases’ through which women physicians tend to move. Other researchers have reflected on the effects of generational differences in career motivation in which members of the Generations ‘X’ and ‘Y’ are often expected to prefer jobs that reflect their personal values and which have greater scope for work−life balance (Sullivan et al., 2009). However, in our study, the women physicians who wanted to live up to all gendered role prescriptions at the same time were the youngest category (Generation Y) and were most likely to be living and working in-house. It could be that younger women physicians, who are potentially on the threshold of both motherhood and a career as a physician and who often do not yet have a permanent contract of employment, are more susceptible to normative role prescriptions and – as a result – require more social approval from their environment. Longitudinal research should be carried out to investigate the dynamic nature of how highly educated women are affected by gendered role prescriptions throughout their career.
A final limitation concerns an issue that is commonly experienced by researchers when investigating men and women’s career motivation. As in almost all quantitative studies of career motivation, the construct of a person’s career motivation is based largely on the career experiences of men. Women are thus measured by men’s standards (Pringle and McCulloch Dixon, 2003). For example, the use of ‘career centrality’ could be inappropriate given that making one’s career central to one’s life implies that someone else needs to take care of all the other aspects of one’s life. While for men ‘career centrality’ could have a positive connotation, as a driven, ambitious and goal-oriented man will receive social approval for his behaviour (Eagly and Steffen, 1984), for a woman with children the same traits could have a negative connotation, as she would receive much less social approval for her making her career central in her life if there is no financial need to do so (Bagilhole, 2006; Mills and Käht, 2010; SCP, 2010). Nevertheless, we decided to use the career motivation construct because we were interested in how women physicians’ goal frames would affect (the admittedly gendered notion of) their career motivation. What is more, we deliberately focused solely on women physicians, and did not compare men’s and women’s goal frames and career motivation. However, for the sake of gender equality, a similar study should be conducted for male physicians, as with the increasingly prominent notion of the ‘involved father’, men too find themselves more and more caught between conflicting ideals (Uhlenberg, 2000). This, we believe, would require further investigation in career construct measurements, preferably supported by qualitative research to specify the switching and non-traditional goal frames and how these goal frames may or may not provide an example of this interpretive perspective in career theory.
Despite the shortcomings mentioned above, practitioners can draw some insight from our results. Interventions such as gender-equality arrangements are designed to change structures that produce inequality, but often with no corresponding intervention into beliefs that legitimize inequality (Ely and Meyerson, 2000). This study has revealed not only how internalized gendered beliefs regarding the ideal mother and the ideal physician affect career motivation among women physicians, but also how gender-equality arrangements mainly seem to support those who prioritize one dominant role prescription over the other. The ‘superwomen’, on the other hand – women (medical) professionals who identify themselves as both a dedicated professional and a dedicated mother – apparently are every bit as career motivated as women who focus mainly on their career. In a relatively egalitarian society such as the Netherlands, women who accept gendered ideals may still experience more life satisfaction than women who ‘resist’ them (Napier et al., 2010), but this does not absolve academics, (HR) practitioners and managers of the responsibility to find proper support for those who may, in the long run, pay a high price (e.g. a decline in their physical well-being). Perhaps the time is right for further investigation and experimentation when it comes to how equality programmes can support women who are trying to stretch the limits of normative ideals.
Footnotes
Acknowledgements
The authors would like to thank Samula Mescher, Associate Editor Professor Terry Beehr, and three anonymous reviewers, whose feedback resulted in a much more cogent article.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.