Compromising conventions: attitudes of dissonance and indifference towards full-time maternal employment in Denmark,Spain,Poland and the UK

Abstract

The article examines cross-national variations in attitudes towards gender roles and the extent to which they map onto regime types. It explores intra-national variation in attitudes to non-traditional gendered behaviour drawing on the theoretical approach of the ‘economy of conventions’, informed by feminist perspectives from comparative research. Data from the European Social Survey are used to map where there is a strong degree of resonance or dissonance between societal and individual attitudes and how these are attenuated by sex and employment status. The results expose unexpected national and intra-national similarities and differences. Societies characterized by a traditional male breadwinner model, such as Spain, indicate a higher degree of permissive values than expected; more liberal countries like the UK show high degrees of indifference, as well as a strong element of traditionalism. Dissonance and indifference compromise traditional gendered conventions and illustrate underlying tensions at the individual and societal level in resolving gender conflicts.

Keywords

attitudes conventions Denmark employment regimes European Social Survey gender roles maternal employment Poland Spain UK

Introduction: gender conventions and attitudinal research in comparative perspective

Comparative employment research has been embedded in an analysis of the institutional and policy effects of welfare and employment regime types (Daly and Rake, 2003). Where attitudes have been examined, these often correspond to the characteristics of national regimes shaped by historical institutionalism (Alwin et al., 1992; Pfau-Effinger, 2004), with a relative neglect of intra-country differences. There are a few important exceptions of researchers who have unpicked differences in attitudinal disposition related to class (Warren, 2000) or partnership status and ethnicity (Duncan et al., 2003). Nevertheless, it is suggested that the dominant paradigm of analysis based on countries as regime types has tended to blind scholars to internal differences within societies when investigated from a comparative perspective.

This article examines where gendered attitudes vary both within and across four European countries. Using the économie des conventions approach, informed by the comparative feminist literature on employment regimes and attitudes, the article illustrates the extent of cross-national differences and similarities, the level of dissonance or resonance to prevalent norms and the characteristics of those holding these views. This analysis confirms where attitudes support regime types and where dissonance is evident. It is argued that looking for evidence of cognitive dissonance suggests where people may contest the legacy of regime policies for managing gender conflicts.

Conventions

Gender differences are experienced and expressed in the way household members negotiate, conform, contest or transgress societal norms about the division of labour around paid work and care responsibilities (West and Zimmerman, 2009). Attitudes to gender equality create ‘gendered conventions’ about acceptable, normal practices that shape the way families organize their daily lives. These conventions can be understood as societal ‘compromises’ (Thévenon, 2006). They become established from attempts to coordinate family demands and women’s employment in terms of individuals’ aspirations and their capabilities to realize these (Hobson, 2011). The degree to which institutional provisions support gender equity shape the reference point and attitudinal expectations about fairness at home and in society at large (Greenstein, 2009; Thompson, 1991).

Letablier (2004) and Thévenon (2006) discuss ‘gender conventions’ as underpinning specific societal contracts understood as gendered societal compromises (O’Reilly and Spee, 1998). Thévenon (2006) focuses on the coordination decisions around job opportunities and family formation in different welfare states. He argues that this approach provides a bridge in understanding macro regulation and the consequences at the micro level on the decisions of households to manage this coordination problem. However, compromises are always unstable. They require justification and legitimation in the eyes of those who make them; otherwise they are open to contestation and change.

Thévenon’s approach is based on the French school of regulation theory known as the économie des conventions (EC) (Boltanski and Thévenot, 2006; Eymard-Duvernay, 2006a, 2006b). The key problem posed by theorists interested in the concept of conventions is to understand how actors solve coordination problems under circumstances of limited rationality and uncertainty. There are a number of ways in which conventions have been defined and used. Game theoretical approaches emphasize instrumental individualism (Lewis, 1969). More holistic approaches have a closer bearing on the concept of Verstehen, where individuals are the authors of the meaning they attribute to their social position and where they simultaneously occupy plural identities (Gilbert, 1992, 2008). Game theorists see conventions as solutions to conflicts, providing ways of regulating competing interests in attempts to cooperate. In contrast, the EC approach understands conventions as providing temporary stability ‘through a range of consistent interpretations and practices that agents assume to be shared (to some degree) by other participants in the interaction’ (Latsis et al., 2010: 550). Rather than focus on overt behaviour, EC scholars emphasize identifying normative frameworks of values, or ‘orders of worth’, about the way rules of society are integrated into the coordination of daily life. Actors make evaluative judgements about acceptable behaviour and justice, drawing on concepts of correctness, value and quality.

Boltanski and Thévenot (2006) argue that there is a plurality of ‘orders of worth’, or justifications. They identify six ‘worlds’ of justification: inspiration, domestic, fame, civic, market and industrial. While each dimension provides a different set of criteria to evaluate and justify cooperation between actors, these different worlds of justification co-exist. Compromising for the ‘common good’ requires ‘the possibility of, bounded in space and time, making two or more orders of worth compatible and to encompass them, with these limits, in an overarching, unifying qualification’ (Thévenot, 2009: 800). By ‘qualification’ Thévenot means a common and accepted understanding of different categories in the social order. Thévenot acknowledges that the failure of recognition and remuneration of care in public forms of justification ‘has been borne by professional care workers’ (2009: 808) in undervalued jobs. The undervaluation of an individual’s contribution to the household, for example in relation to care provision, leaves them little means to survive independently outside the household (Sen, 1990). This dependency compels them to cooperate with existing dominant conventions.

The EC approach gives prominence to competing value frameworks used to appraise the worth of different forms of action; it recognizes that individuals can simultaneously have multiple evaluative frameworks, or different ‘worlds of worth’. This approach also points to the instability of legitimation emanating from different worlds of evaluation. While this multi-layered perspective is attractive in understanding the complex and sometimes contradictory rationales actors draw on, their approach lacks a systematic feminist analysis that could be applied to understanding work-life conflicts and the competing demands on parental time.

Feminist approaches to comparative research

More feminist approaches to understanding how normative evaluations of family and working life are made have been developed by Duncan and Edwards using the concept of ‘gendered moral rationalities’. Their contribution is situated as a critique of the mainstream economic conception of ‘rational economic man’ from the new household economics and the over individualized accounts of how families negotiate in late modernity (Duncan et al., 2003). They emphasize the contextualization of tastes and preferences about what is morally right and socially acceptable and how these are ‘collectively negotiated, sustained, modified and changed’ (Duncan and Edwards, 1997: 33). Their approach was initially developed to understand intra-societal differences between lone mothers, ethnic and class differences in the UK; it was later applied to cross-national comparisons of lone mothers (Duncan and Edwards, 1999). They suggested that a concept of ‘genderfare’ encapsulates the interaction between welfare regimes and gender contracts derived from gender relations and the way different groups of women navigate their existence between these. Their approach illustrates how ‘economic rationality’ varies between and within states, with its origins in political and ideological conceptions of society.

This argument is developed by Warren’s (2000) comparison of Denmark and the UK. She argues that accepted ideologies concerning the appropriate role for mothers shape gender contracts among different classes of women in each society. This affects both the form and levels of women’s employment within and between countries. Evaluative frameworks in terms of class need ‘unpicking’ to reveal similarities between countries and how institutional provisions mediate to produce different outcomes.

Analysing maternal employment requires a multi-level approach. Distinguishing between three levels of analysis, Le Feuvre (2010) discusses the work orientations and practices of professional British and French women in terms of the gender contract, gender regimes and gender experiences. Her analysis of the gender contract represents an accepted societal ideology about appropriate roles for men and women at the macro level. This ideology is reinforced through policy choices made by governments and assumedly supported by their electorate. This captures the dominant practices and policies in a given society and how they change. Her analysis of gender regimes at a meso level focuses on mediating institutions such as paid work, social welfare, education, trade unions and political parties. These organizations ‘mediate’ gender relations in particular fields and articulate ‘solutions’ to gendered conflicts over competing demands for women’s labour. These ‘solutions’ can often result in a plethora of contradictory policies (Saraceno and Keck, 2010). One consequence of this can result in germinating future gender conflicts and compromises in attempts to resolve coordination problems between the micro and meso level (Leira, 2002).

Gender conflicts come about because of the inability to resolve coordination problems, in this case between paid work and care (Hirdman, 1994; O’Reilly, 2009). Even where attitudes have changed, opportunities to realize them have not. Opportunities created, or constrained, by care policies and jobs, or the lack of them, alongside the domestic division of labour to realize these changing attitudinal dispositions have not resulted in societal cohesion. Rather, they have created a new series of work-life conflicts across Europe (McGinnity and Whelan, 2009). This conflict is apparent from the research of Crompton et al. (2005) who found that while there is evidence of growing support for more egalitarian attitudes, these are not matched with a corresponding increase in men’s participation in household work. They attribute this not only to differences in state welfare policies and changing attitudes, but also to the intensification of paid work, making it more difficult for men to participate in domestic responsibilities.

These conflicts are discussed by Schober and Scott (2012) in terms of cognitive dissonance in reconciling pre- and post-natal parenting roles. They propose combining both psychological as well as structural factors to understand how more, or less, traditional attitudes to parenting and employment emerge. Evidence of how attitudes adapt to changed behaviour among peer groups has been provided by Himmelweit and Sigala (2004). They show how, as the number of mothers working has increased in Britain, those believing this is harmful for children have declined. Changing one’s opinion reduces the conflict between attitudes and behaviour, which can be understood as resolving cognitive dissonance.

Cognitive dissonance arises when the consequences of one’s actions produce unwanted results (Cooper, 2007). The originator of the theory, Festinger (1957), argued that there was a drive for individuals to resolve inconsistencies between their attitudes and their actions. They could do this either by changing their attitudes from what they had originally stated as a preference, or by providing a new alternative explanation for why the outcome they had expected had not materialized. This approach has received considerable attention in social psychology over the past 50 years, but limited attention in economic sociology. It is suggested that this approach could provide a fruitful direction for future research that goes beyond static regime typologies.

Regime typologies can be too restrictive and fail to address the multifarious experiences of women’s work and welfare in different countries (Daly and Rake, 2003). One of the limitations of regime typologies has led researchers to the question: if societies are constrained by path dependency and historical institutionalism, then how does change come about (Streeck and Thelen, 2005)? In the face of revolutionary change, as exemplified by the collapse of communism in Europe, Schmitt and Trappe (2010) ask how this has affected gender relations in the East. They argue that by combining an examination of institutional factors with the role of norms, preferences and historical social legacies one can begin to understand the persistence of differences between post-Soviet societies and other parts of Europe. One can also begin to identify how catalysts of discontent might contribute to the way different forms of maternal employment are supported or stigmatized.

Here the article sets out to examine the attitudinal evidence underlying family-work arrangements in four European countries: Denmark, Spain, Poland and the UK. These four countries can be seen as illustrative of different regime types: liberal, social democratic, Mediterranean and East European. They also represent a spectrum from strong to weak male breadwinner models. In terms of historically inherited gender conventions these four countries provide some interesting similarities and differences. Both Poland and Spain share a strong Catholic and conservative tradition; at the same time they represent two very different political traditions about the role of women in employment in countries formerly subject to communist and fascist governments (Cousins, 1994; Haas et al., 2006; Pollert, 2003). Denmark and the UK are also often considered to be quite distinct welfare models: social democratic and liberal. Nevertheless, they also share common characteristics of relatively high levels of female employment and part-time work as well as a more permissive attitude to gender equality (Gash, 2008; Steiber and Haas, 2010; Warren, 2000). In this sense, the choice of these four countries for comparison permits the use of the comparative method discussed by Crompton and Lyonette (2006) related to most similar and most different cases.

The aim of this article is to investigate what evidence there is of cognitive dissonance between the views held by individuals and their perceptions of those held by others in their society towards a range of non-traditional family and work conventions. The aim is to identify the factors associated with holding more traditional or permissive attitudes; and to compare how these are distributed both within and between respondents in each of these four countries, drawing on evidence from the European Social Survey.

Comparative regime theory would suggest the hypothesis, first, that more traditional attitudes will be found in Spain and Poland and more liberal attitudes in Denmark and the UK. These findings would support the characterization of their regime type as strong or weak male breadwinner societies. Second, class differences might be expected to affect attitudes. Using education as a proxy for class, more highly educated women across all countries might be expected to have more liberal attitudes than the less well qualified. This could be a result of their higher levels of expectations resulting from an investment in their human capital and their expectations about gender equity and the household division of labour (Greenstein, 2009). This hypothesis addresses the propositions of Warren (2000) that class still matters in shaping attitudinal dispositions, as do employment and partnership statuses (Duncan et al., 2003).

Data

The European Social Survey (ESS-2006)¹ is a particularly apposite source to examine these questions for three compelling reasons. First, individuals’ attitudes to a range of questions about families and maternal employment can be systematically compared between countries. Second, it is possible to compare how these individuals see themselves in comparison to those around them, which is interpreted as a measure of societal values. Third, one can differentiate between attitudes held by and towards men and women in each country. In this way the similarity and differences to gender norms can be compared within and across societies, at an individual and societal level and by gender. This allows co-existing conventional frameworks to be identified in different societies as well as the characteristics of individuals holding these views, by gender, educational attainment and social status.

Clearly there are limitations to using cross-sectional data to examine these issues, as previous researchers have indicated (Alwin et al., 1992: 16). This type of analysis cannot explain how and why these values have emerged; this would require a more qualitative, longitudinal or historical approach. But it can, nevertheless, provide evidence about the comparative strength of these attitudes both across societies and within them.

The structure of the ESS (Round 3) asks respondents about their own values and also how they think ‘most people’ would react to a number of questions concerning non-traditional gender practices around families and work. Individuals were asked:

‘Do you approve or disapprove if a woman/man:

… lives with a partner without being married to him/her?

… has a child with a partner she/he lives with but is not married to?

… gets divorced while she/he has children aged under 12?

… chooses never to have children?

… has a full-time job while she/he has children aged under 3?’

The same five dimensions were also examined at a societal level with the question:

‘Apart from your own feelings, how do you think most people would react if a woman/man they knew well did any of the following?’

Although the same dimensions were included at both levels, responses were recorded on a different scale. At the individual level responses were recorded on a five-point scale:

Strongly disapprove

Disapprove

Neither approve nor disapprove

Approve

Strongly approve

Societal attitudes were recorded on a four-point scale:

Most people would openly disapprove

Most people would secretly disapprove

Most people would not mind either way

Most people would approve

Harrison and Fitzgerald (2010) argue that there might be some discrepancy in the different scoring of these responses. Therefore, to harmonize comparisons between the individual and societal questions, the responses were recoded to generate a three-score scale: approval, indifference and disapproval. Comparing the responses to these questions allowed individuals’ responses to be compared with their perception of how others in their society would react. Additionally, it was possible to distinguish between attitudes held by and towards men and women because of the split ballot method.²

Findings

The findings are presented in two stages. First the descriptive statistics are provided to indicate the distribution of responses to the questions about families and employment using aggregate scores. The second part focuses solely on the question of attitudes to parental employment. Using a multinomial logit model, men’s and women’s attitudes were compared, accounting for differences in age, education, family and employment status for all four countries.

Approval and disapproval

The summary descriptive findings are presented in Table 1 for all five dimensions concerning family and working practices at the societal and individual levels and for women and men. (The highest aggregate score was 3 indicating strong non-traditional attitudes).³ Across all countries aggregate individual responses always scored more highly than respondents’ perceptions of ‘most people’, suggesting that individuals tended to see themselves as more permissive than those around them in all the countries.

Table 1.

Average individual and societal approval of family and work norms.

	Aggregate score		Cohabitation		Children born outside marriage
	Individual	Society	Individual	Society	Individual	Society
Denmark	2.78 (1)	2.44 (1)	2.88 (1)	2.58 (1)	2.80 (2)	2.56 (1)
Men about women	2.75 (2)	2.42 (2)	2.88 (2)	2.59 (3)	2.81 (3)	2.54 (3)
Men about men	2.79 (2)	2.44 (2)	2.91 (2)	2.56 (3)	2.88 (2)	2.57 (3)
Women about women	2.77 (2)	2.43 (2)	2.88 (2)	2.57 (3)	2.77 (3)	2.57 (3)
Women about men	2.78 (2)	2.46 (2)	2.86 (2)	2.61 (3)	2.73 (3)	2.55 (3)
Spain	2.24 (1)	1.75 (1)	2.40 (2)	1.79 (1)	2.36 (2)	1.74 (1)
Men about women	2.17 (3)	1.71 (2)	2.37 (3)	1.80 (3)	2.31 (4)	1.72 (3)
Men about men	2.25 (2)	1.80 (2)	2.44 (3)	1.83 (3)	2.38 (4)	1.80 (3)
Women about women	2.27 (3)	1.70 (2)	2.44 (3)	1.73 (3)	2.38 (3)	1.68 (3)
Women about men	2.25 (3)	1.80 (2)	2.36 (4)	1.82 (3)	2.35 (4)	1.77 (3)
United Kingdom	2.06 (1)	1.84 (1)	2.11 (1)	1.92 (1)	2.01 (1)	1.83 (1)
Men about women	1.96 (2)	1.76 (1)	2.10 (3)	1.92 (2)	2.02 (3)	1.85 (2)
Men about men	2.12 (2)	1.92 (1)	2.12 (3)	1.95 (2)	2.03 (3)	1.84 (2)
Women about women	2.04 (2)	1.80 (1)	2.13 (2)	1.90 (2)	2.00 (3)	1.81 (2)
Women about men	2.12 (1)	1.90 (1)	2.09 (2)	1.93 (2)	2.00 (3)	1.81 (2)
Poland	2.04 (1)	1.71 (1)	2.10 (2)	1.63 (2)	2.15 (2)	1.60 (2)
Men about women	1.98 (3)	1.63 (2)	2.14 (4)	1.62 (3)	2.19 (4)	1.58 (3)
Men about men	2.08 (3)	1.76 (2)	2.13 (4)	1.65 (3)	2.17 (4)	1.61 (3)
Women about women	2.00 (3)	1.67 (2)	2.02 (4)	1.60 (3)	2.07 (4)	1.59 (3)
Women about men	2.09 (3)	1.78 (2)	2.13 (4)	1.66 (3)	2.18 (4)	1.63 (3)

	Choosing to be childless		Divorce with children under 12		Working full-time with children under 3
	Individual	Society	Individual	Society	Individual	Society
Denmark	2.84 (1)	2.33 (2)	2.58 (2)	2.18 (2)	2.76 (2)	2.52 (2)
Men about women	2.82 (3)	2.33 (4)	2.61 (4)	2.27 (4)	2.62 (4)	2.38 (4)
Men about men	2.83 (3)	2.40 (3)	2.45 (4)	2.02 (4)	2.90 (2)	2.63 (3)
Women about women	2.86 (2)	2.25 (4)	2.72 (3)	2.31 (4)	2.59 (4)	2.42 (4)
Women about men	2.84 (3)	2.35 (3)	2.54 (4)	2.14 (4)	2.93 (2)	2.66 (3)
Spain	2.09 (2)	1.71 (1)	1.94 (2)	1.51 (1)	2.40 (2)	1.99 (2)
Men about women	2.09 (4)	1.73 (3)	1.91 (4)	1.49 (3)	2.15 (4)	1.80 (3)
Men about men	2.03 (4)	1.71 (3)	1.79 (4)	1.47 (3)	2.62 (3)	2.20 (3)
Women about women	2.19 (3)	1.70 (3)	2.08 (4)	1.53 (3)	2.27 (3)	1.83 (3)
Women about men	2.03 (4)	1.71 (3)	1.96 (4)	1.54 (3)	2.56 (3)	2.15 (3)
United Kingdom	2.14 (1)	1.87 (1)	1.86 (1)	1.60 (1)	2.17 (2)	1.98 (1)
Men about women	2.16 (2)	1.85 (2)	1.81 (3)	1.59 (2)	1.71 (3)	1.56 (2)
Men about men	2.13 (2)	1.91 (2)	1.72 (3)	1.53 (2)	2.57 (2)	2.36 (3)
Women about women	2.19 (2)	1.88 (2)	2.02 (2)	1.74 (2)	1.85 (3)	1.64 (2)
Women about men	2.09 (2)	1.85 (2)	1.85 (2)	1.55 (2)	2.55 (2)	2.37 (2)
Poland	1.71 (2)	1.63 (1)	1.71 (2)	1.49 (1)	2.50 (2)	2.20 (2)
Men about women	1.71 (4)	1.59 (3)	1.74 (4)	1.48 (3)	2.17 (4)	1.93 (3)
Men about men	1.72 (4)	1.69 (3)	1.61 (4)	1.45 (3)	2.79 (3)	2.42 (3)
Women about women	1.74 (4)	1.59 (3)	1.88 (4)	1.60 (3)	2.26 (4)	1.99 (3)
Women about men	1.67 (4)	1.66 (3)	1.61 (4)	1.43 (3)	2.79 (2)	2.46 (3)

Score for individual approval: 1 = Strongly disapprove or disapprove; 2 = Neither approve nor disapprove; 3 = Approve or Strongly approve.

Score for society approval: 1 = Most people would openly disapprove or Most people would secretly disapprove; 2 = Most people would not mind either way; 3 = Most people would approve

Note beside each average refers to the standard error, so the confidence interval at β=95% will be: (1) = mean +/- 0.01; (2) = mean +/- 0.02; (3) = mean +/- 0.03; (4) = mean +/- 0.04.

Country differences at the individual level generated a hierarchy, with Denmark consistently having the highest levels of positive approval on all dimensions: the aggregate individual mean score was 2.78. Spanish respondents (2.24) had a higher score than those in the UK (2.06) who were closer to Poland (2.04). At the societal level this ranking changed slightly with the UK moving closer to Denmark. While the strong non-traditional attitudes held in Denmark may not be very surprising, it is the more differentiated attitudes in the other countries that are intriguing. In particular in Spain, there was the largest gap between individuals with permissive attitudes and the perception of more conservative values held at the societal level; the reverse appeared to be true for the UK where the ranking changed between the individual and societal level.

The differences in these rankings are illustrated for example around opinions to cohabitation: approval at the aggregate individual level was higher in Spain than in Poland or the UK; but in the UK there was a higher perceived societal acceptance than in Poland and Spain. Acceptance of children born outside marriage reflected a similar pattern. The UK had the lowest aggregate scores at the individual level, but when comparing societal acceptance across countries the scores were higher in the UK than in Poland or Spain. Choosing to be childless received the lowest approval in Poland, followed by Spain and the UK, with Denmark scoring the highest rates of approval. On all three of these dimensions gender differences were not very pronounced, except for Poland: in Poland women’s attitudes to women cohabitating or having children outside marriage were more traditional compared to men’s attitudes, or to women’s attitudes about men.

Gender differences were more evident when examining divorce. Divorcing with children under the age of 12 received much stronger disapproval directed by and at men than by or at women, across all countries. This could be interpreted as an underlying assumption that fathers should be providers and if they divorce when the children are young they are abdicating this responsibility.

Where considerable gender and cross-national differences were visible was in attitudes to working full-time with children under the age of three years. Overall, the highest rates of approval for maternal full-time employment with children under three were found in Denmark (2.76), followed by Poland (2.50), Spain (2.40) and the UK (2.17) (mean aggregate individual scores). But in all countries there was more approval for men, than for women, working full-time where small children were present. The strength of disapproving attitudes held by men and women within a country tended to be of a similar proportion. However, men tended to disapprove more than women of women working full-time when small children were present; this was statistically significant in Spain and the UK, but not in Denmark and Poland.

Indifference

Using aggregate mean scores provides a summary measure of attitudes on a spectrum. However, this can obscure the importance of those who hold ‘indifferent’ attitudes. Harrison and Fitzgerald (2010) argue that high levels of indifference indicate a lack of traction for a particular value or norm. Overall, looking only at indifference scores, these were found to be higher at the societal than at the individual level. However, there were striking cross-national differences. The UK stood out as having the highest levels of indifference at both the societal and individual levels across all the dimensions examined. It was only on the measure of mothers working full-time that indifference changed to disapproval in the UK. In Denmark levels of indifference among individuals were very low. Danish respondents, in general, held very strong attitudes supporting non-traditional gender conventions. In Poland, traditional family conventions still had traction as reflected in respondents having the lowest levels of indifference to questions about cohabitation and children. In Spain, societal responses were similar to those in Poland, but a much higher proportion of individual Spanish people were indifferent to the choice of being childless, suggesting evidence of a move away from traditional societal conventions. National differences appeared to outweigh gender differences, except in relation to men divorcing and where mothers with small children were working full-time.

The first hypothesis suggested that one might have expected dominant stereotypes associated with more conservative and Catholic attitudes to be stronger in countries like Poland and Spain than was in fact evidenced from the data. Spain scored slightly higher on non-traditional gender values at the individual level than was the case in the UK and Poland. This evidence suggests a tension existed in Spain that questioned the legacy and dominant convention of a traditional breadwinner society. In the UK there was evidence of stronger conservative attitudes and higher levels of indifference reflecting different kinds of tensions to those in Spain.

Comparing attitudes to maternal and paternal full-time employment

Societal comparisons

The ‘economy of conventions’ approach suggests that one can expect to find competing value systems co-existing. To grasp this complexity and variety a nine-category typology was developed to measure the degree of dissonance, indifference and resonance between individuals and their perception of societal (dis)approval (see Table 2). This created three main categories: traditional, indifferent and permissive attitudes. Within each of these three categories the study distinguished between those who felt that societal attitudes accorded with, were indifferent to, or opposed to, their own position. The focus was on responses from both men and women to women’s employment as previous analysis indicated that disapproval rates were much higher for women than for men; the distribution of these attitudes are presented in Figure 1.

Table 2.

Typology of attitudes to maternal full-time employment with children under 3 present (individual and societal opinions).

Label (individual/societal)	Definition
1. Traditional Dissonance (dis/app)	Individual disapproves, feels most people approve
2. Traditional Assertive (dis/ind)	Individual disapproves, feels most people indifferent
3. Traditional Resonance (dis/dis)	Individual disapproves, feels most people disapprove
4. Indifference (ind/dis)	Individual indifferent, feels most people disapprove
5. Indifference (ind/ind)	Individual indifferent and so are most other people
6. Indifference (ind/app)	Individual indifferent, feels most people approve
7. Permissive Resonance (app/app)	Individual approves, feels most people approve too
8. Permissive Assertive (app/ind)	Individual approves, feels most people indifferent
9. Permissive Dissonance (app/dis)	Individual approves, feels most people disapprove

Figure 1.

Attitudes to maternal full-time work if small child(ren) (ind/soc) by country.

The UK had the lowest degree of support for mothers working full-time where there were small children present, compared to all the other countries. It also had the highest proportion expressing indifference. Denmark had the highest levels of individual and societal resonance, with nearly 50 per cent of Danish respondents in the category of Permissive Resonance and over 70 per cent in total expressing some degree of approval. Poland, followed by Spain, had a much higher degree of support for female full-time employment than was the case in the UK. Maternal full-time employment in Poland was accepted, even if it might not fit the more traditional conservative model of the family. This finding was possibly due to the economic necessity of working where the gender poverty gap was high; to the availability of part-time work being low; or due to the communist legacy encouraging female employment on a full-time basis (Braun et al., 1994; Steiber and Haas, 2010). Permissive dissonance was felt most strongly in Spain: a small but significant group, just under 10 per cent, individually approved but felt that full-time maternal employment with small children was stigmatized by the social attitudes of others.

Research on cognitive dissonance suggests that when respondents know they are comparing their responses to an external comparator they moderate these so that there is closer ranking between the two responses (Cooper, 2007). Given this finding, it might be assumed that the cases of both traditional and permissive dissonance recorded here were underestimated. Nevertheless, the differences between the countries are intriguing and do not support the initial hypothesis that more traditional attitudes would be found in Spain and Poland.

Individual level comparisons

To identify the individuals’ characteristics associated with these attitudes a multivariate analysis with a multinomial logit model was conducted on all nine categories (see Table 3). Attitudes of both men and women to maternal and paternal full-time employment with small children present were compared. The second hypothesis was that more highly educated women would have higher levels of approval for full-time maternal employment with small children compared to the less well qualified.

Table 3.

Multinomial logit model: approval, indifference or disapproval of individuals’ work when small child(ren) (individual/societal) present.

Work approval for parents with small child(ren) base outcome (app/app=ref)	Traditional Dissonance (dis/app) 1.	Traditional Assertive (dis/ind) 2.	Traditional Resonance (dis/dis) 3.	Indifference (ind/dis) 4.	Indifference (ind/ind) 5.	Indifference (ind/app) 6.	Permissive Assertive (app/ind) 8.	Permissive Dissonance (app/dis) 9.
Men (women=ref.)	0.40	−0.20	0.11	−0.48**	−0.02	−0.11	−0.01	0.03
asked about mothers (fathers=ref.)	1.41*	2.95***	3.13***	2.32***	0.98***	0.84	0.49*	2.75***
Men x asked about mothers (fathers=ref.)	0.05	0.62**	0.63***	0.82***	0.19	−0.63**	−0.14	−0.36
Age (centered at 45)	−0.02	−0.03***	−0.02***	−0.03***	0.00	0.00	0.00	0.00
Age (at 45) x asked about mothers	0.02	0.04***	0.03***	0.02*	0.00	−0.04***	−0.01	0.01
Age2/100 (at 45)	0.03	0.02	0.09***	0.07**	−0.01	0.03	−0.01	0.04
Age2/100 (at 45) x asked about mothers	0.01	−0.05	−0.07*	−0.02	−0.02	−0.05	−0.05*	−0.09**
Religiosity (0–10)	0.02	0.02	0.05	−0.03	−0.06***	−0.03	−0.05***	−0.03
Religiosity (0–10) x asked about mothers	0.14*	0.07	0.08**	0.02	0.10***	0.07	0.08***	0.01
Currently married (single=ref.)	0.14	0.15	0.22	0.28	0.04	−0.03	−0.19**	0.00
Currently married x asked about mothers	0.08	−0.50	−0.50	−0.19	−0.16	0.59	0.05	0.18
Cohabiting (single=ref.)	−2.10	0.16	−0.09	−0.07	−0.37**	−0.39	−0.15	−0.01
Cohabiting x asked about mothers	2.90	−0.38	−0.47	0.15	0.27	0.20	0.18	0.24
Ever had child(ren) in hh. (childless=ref.)	1.17*	0.75*	−0.14	−0.39	−0.09	0.34	−0.06	−0.09
Ever had child(ren) in hh x asked about mothers	−0.38	−0.17	0.67*	0.84*	0.68***	0.51	0.69***	−0.17
Currently children in hh	−0.51	−0.39	0.48*	0.37	0.13	−0.18	0.08	−0.34
Currently children in hh x asked about mothers	−0.12	0.16	−0.76**	−0.83**	−0.55***	−0.30	−0.51**	0.15
Educ.: primary or lower (secondary=ref.)	−0.35	−0.38*	0.81***	0.36*	0.07	−0.58***	−0.08	0.08
Educ.: primary or lower x asked about mothers	1.42***	0.70**	−0.54*	−0.55**	0.21	0.72***	0.11	−0.23
Educ.: tertiary (secondary=ref.)	0.62	−0.64*	−0.08	−0.27	0.11	0.07	−0.06	0.19
Educ.: tertiary x asked about mothers	−0.11	0.56	−0.35	0.47	0.17	−0.47	0.34*	−0.12
Work: part–time (empl. full-time=ref.)	−1.64	0.23	−0.44	0.49	−0.10	0.82***	0.13	−0.25
Work: part–time x asked about mothers	1.42	−0.11	1.04**	−0.08	−0.01	−1.00*	−0.10	0.28
Work: unemployed (empl. full-time=ref.)	1.38*	−0.59	−1.14	1.31***	0.43**	−0.86	−0.08	0.61*
Work: unemployed x asked about mothers	−1.12	0.33	1.58**	−0.99**	−0.29	1.43*	0.11	−0.61
Work: in education (empl. full-time=ref.)	0.78	0.96**	0.09	1.06***	0.79***	0.52	0.39**	1.23***
Work: in education x asked about mothers	−1.12	−0.48	0.22	−0.67	−0.22	−0.15	−0.09	−0.24
Work: not in employment (empl. full-time=ref)	1.52***	0.59	0.03	0.60**	0.04	−0.07	−0.16	−0.20
Work: not in employment x asked about mothers	−1.67***	−0.45	0.48	−0.30	0.02	0.54	0.11	−0.21
DK	−1.66	−0.56	−0.71	−1.41	−1.56***	−0.62	−0.61***	−0.92
DK x asked about mothers	1.01	−1.15	−1.09	−0.53	−0.53	−0.17	−0.75***	−0.96
ES	−0.86*	1.42***	2.04***	1.17***	1.27***	1.45***	0.55***	1.22***
ES x asked about mothers	−0.27	−1.33***	−1.25***	−0.09	−0.29	−0.92**	−0.22	−0.55**
UK	−0.13	0.20	0.44	0.99***	1.56***	1.83***	−0.22**	−0.26
UK x asked about mothers	0.73	1.64***	2.62***	1.94***	1.20***	−0.80*	0.96***	1.23***
Constant	−5.56***	−4.09***	−5.06***	−4.10***	−1.65***	−3.60***	0.14	−2.87***

Legend: *p<.1; ** p<.05; *** p<.01

Table 4.

Sample distribution.

	Split ballot		Total Sample
	Ask about women	Ask about men
Denmark	734	771	1,505
Male	368	370	738
Female	366	401	767
Spain	959	917	1,876
Male	446	456	902
Female	513	461	974
UK	1,213	1,181	2,394
Male	547	532	1,079
Female	666	649	1,315
Poland	863	858	1,721
Male	412	403	815
Female	451	455	906

Note: These figures are un-weighted and correspond to the exact sample size.

Source: http://www.europeansocialsurvey.org/

The likelihood was compared of belonging to one of the nine groups outlined in Table 2; the reference group was ‘Permissive Resonance’, i.e. where there was approval at both the individual and societal level (category 7). The variables employed controlled for sex, age, religion, family, education and employment status and country effects (with dummy variables). Gender was used to measure differences in men and women’s attitudes to mothers’ and fathers’ full-time employment when children were under three (the reference group was women) and the split ballot (the reference category was ‘asked about fathers’). A series of interaction effects between all covariates and the split ballot permitted contrasting the effects of individuals’ characteristics (and country specific effects) on individuals’ opinions about mothers’ and fathers’ full-time work where young children were present.

To examine whether young people had less traditional attitudes compared to older people a quadratic control for age was used, centred around age 45. To see if those with strong religious beliefs were more traditional, an indicator of religiosity (scale) was used ranging from 0 (not religious) to 10 (very religious). The results were controlled for being married or cohabiting compared to single people (including widowed, divorced and separated). Parenthood with children still living at home and households where children had left were contrasted with those being childless.

Educational attainment distinguished between completed primary or lower education, completed lower secondary school (reference category) and upper secondary or tertiary education. The employment situation of the individuals distinguished between working part-time (where working full-time was the reference category), being unemployed, being in education or in another form of non-employment (comprising housework, retirement and other forms of unpaid activities). Important controls were the country dummy variables, where Poland was the reference category. All analyses applied both design and population weights to allow for valid comparison across countries.

Across all countries both men and women expressed more traditional attitudes towards mothers’ work than towards fathers’ work. However, men were more traditional than women. Overall, middle aged groups were more traditional than younger or older generations; the reverse was true for attitudes to fathers’ work. Being married or cohabiting had no statistically significant effect on attitudinal disposition. Parents were more likely to have traditional or indifferent attitudes to women’s work, but if their children were still living at home this effect diminished. Students’ responses were distributed across the ideational spectrum, with no significant difference between attitudes to women’s or men’s work.

Evidence for hypothesis two was found where lower levels of education were associated with holding more traditional attitudes towards maternal employment. A computation of predicted probabilities revealed that those with lower levels of education were more likely to be more conservative, particularly in Spain and the UK; this effect was marginal in Poland and not evident in Denmark. Traditional attitudes were strong in the UK and to a lesser extent in Spain, compared with Poland or Denmark.

In contrast, individuals with higher levels of education and parents whose children had left home expressed more permissive attitudes towards mothers’ full-time employment. Indifference was more widespread among parents whose children no longer lived in the household, the unemployed, the non-employed, among UK and to a lesser extent Spanish respondents.

In accounting for country differences these results reinforce the descriptive analysis discussed earlier, but also illustrate some significant nuances in the distribution of attitudinal dispositions within countries. Denmark had the highest proportion of respondents in the reference category of permissive resonance.⁴ Traditional attitudes were more common in the UK than in Poland. Spain also revealed some support for traditional attitudes, but these were lower than in the UK. Indifference was particularly high in the UK and also evident in Spain. A significant group of those in the more permissive categories (8 and 9) in the UK was partly due to the very small numbers in Britain found in the reference category of permissive resonance (see Figure 1). High levels of indifference suggest that traditional values have limited traction (Harrison and Fitzgerald, 2010). Opinions in the UK and Spain appeared to be more dispersed across the three key areas of traditional, indifferent and permissive than opinions in the other countries, suggesting that there was a greater tension between individual attitudes and the perception of high levels of societal stigma in these countries.

Conclusions and discussion

Using the framework of the ‘economy of conventions’, informed by comparative feminist research, this article has argued that this analysis can identify competing evaluative frameworks about the acceptability of non-traditional gendered conventions around the family and employment. The value of the EC framework encourages a search for competing rationales or ‘justifications’. The approach used allows expressions of ‘gender conflicts’ between attitudinal dispositions and perceptions of societal conventions to be identified, alongside the characteristics of those who experience these most strongly. Schober and Scott (2012) have suggested that focusing on cognitive dissonance is an important dimension in understanding emerging patterns of coordinating paid work and care that encompass psychological as well as institutional factors. Here, comparative cross-sectional data have been drawn on to address the question of how far attitudes support the characteristics of regime typologies. The analysis indicates that the conventions underlying a traditional male breadwinner gender contract in these societies are being compromised, but they are also being reinforced.

They are being compromised in the sense that the traction of traditional conventions is being undermined either by approval of non-traditional behaviour or through indifference. A notable minority of women, who were more likely to be better qualified, was contesting traditional conventions in Spain, Poland and to an extent in the UK. In Spain permissive values around the family and maternal employment proved more extensive than might have been expected, especially in a society where it is harder for mothers to remain in employment, where welfare support is weak and part-time work is low. Even in Poland more support was found for maternal full-time employment than might have been expected for a society with strong traditional family values. This could in part be due to the economic necessity for Polish women to work, together with the more limited availability of part-time employment.

If indifference is understood as an indication of the limited traction of traditional conventions, the UK and to a lesser degree Spain illustrated this. Finally, Denmark stood out as a society with a very high degree of resonance in support of non-traditional gender conventions. National institutional contexts contribute to establishing individuals’ evaluative frameworks about legitimate and appropriate behaviours with respect to gender equity (Greenstein, 2009). However, importantly, the analysis identified areas of dissonance that compromised traditional conventions.

Nevertheless, the reinforcement of traditional conventions was also evident. Even in Denmark, full-time maternal employment received less support than was the case for fathers. The UK stood out in this comparison as having quite strong traditional attitudes. Gash (2008) and Steiber and Haas (2010) argue that these attitudes are related to the relatively poor infrastructure for family policy and the pervasiveness of part-time employment. Under these conditions it may not be surprising that part-time work was seen as an acceptable way for mothers to meet competing demands on their time for paid and unpaid labour. In this sense part-time employment can become established as a conventional practice that allows women to participate in paid work, while still taking the majority burden for domestic work (O’Reilly and Fagan, 1998). In countries where part-time work is less popular, the only way for mothers to work is on a full-time basis, which may indicate why full-time maternal employment was less stigmatized in Poland and Spain (Haas et al., 2006).

Cognitive dissonance in the analysis was found on the extreme spectrums of traditional and permissive attitudes. Indifference may have dampened out resentment and become an emerging, dominant characteristic in countries like the UK and to a lesser degree in Spain. The approach reported here indicates one way of exploring this relatively neglected question of how attitudes map onto regime types. The contribution of this research is to unveil the landscape of these dispositions both within and between societies. In doing so it reveals potential catalysts for contesting traditional conventions and the groups of people most likely to hold these positions. A future research agenda exploring these aspects could include more qualitative, longitudinal and/or historical perspectives for specific countries. It would be of interest to follow up the consequences of cognitive dissonance and how these were resolved, or not, at an individual, meso and macro level. The results presented here indicate that for particular groups of women the discord between the realities of working and societal acceptance of this is still very deep.

Footnotes

Acknowledgements

The authors are very grateful for the critical contributions provided by the reviewers and in particular from Prof. Jenny Tomlinson.

Funding

This research was funded by a Leverhulme Major Research Fellowship for Prof. J. O’Reilly, Ref 20090128.

Notes

Jacqueline O’Reilly is Professor for Comparative Employment Relations and Human Resource Management at the University of Brighton Business School in the UK. She is Director of the Centre for Research on Management and Employment (CROME) and was a Leverhulme Major Research Project Fellow (2009–12). She previously worked at the Wissenschaftszentrum Berlin; was a Jean Monnet Fellow at the European University Institute, Florence; and UMIST, Manchester. She has studied at the Observatoire Sociologique du Changement, Paris and Nuffield College, Oxford. She is co-editor of Part-Time Prospects with Colette Fagan. She is coordinator of the EU FP7 style-research.eu project examining Strategic Transitions for Youth Labour in Europe (2014-7).

Tiziana Nazio is a lecturer in sociology at the University of Turin and Affiliate Fellow of the Collegio Carlo Alberto. She is author of Cohabitation, Family and Society (Routledge, 2008). She has participated in a number of EU projects including ‘Analysis of Models of Economic Support in a European Perspective’, ‘Housing Policy and Family Dynamics in Turin’ and ‘Evaluation of Local Welfare in Turin’ (1998–2000). She was principal investigator of the ESRC funded project ‘Are Storks Striking for a Contract Renewal? Childbirth under Changing Employment, Family and Welfare Arrangements’ (2006–8) at Nuffield College, Oxford University.

José Manuel Roche is a postdoctoral researcher in the Department of International Development at the University of Oxford and holds a DPhil in sociology from the University of Sussex. His research focuses on global comparative analysis of distributional aspects of wellbeing and on human development and the capability approach. He is Education Officer and Member of the Executive council (elected 2012–15) of the Human Development and Capability Association (HDCA) and coordinator of the Quantitative Research Thematic Group at the HDCA (since 2009). He was awarded the 2007 Wiebke Kuklys Prize and is a Chevening Alumnus (2004/6).

References

Alwin

Braun

Scott

(1992) The separation of work and the family: attitudes towards women’s labour-force participation in Germany, Great Britain, and the United States. European Sociological Review 8(1): 13–37.

Boltanski

Thévenot

(2006) On Justification: Economies of Worth. Translated by Porter

Catherine

. Princeton, NJ and Oxford: Princeton University Press.

Braun

Scott

Alwin

(1994) Economic necessity or self-actualization? Attitudes toward women’s labour-force participation in East and West Germany. European Sociological Review 10(1): 29–47.

Cooper

(2007) Cognitive Dissonance: 50 Years of a Classic Theory. London: Sage.

Cousins

(1994) A comparison of the labour market position of women in Spain and the UK with reference to the ‘flexible’ labour debate. Work, Employment and Society 8(1): 45–67.

Crompton

Lyonette

(2006) Some issues in cross-national comparative research methods: a comparison of attitudes to promotion, and women’s employment, in Britain and Portugal. Work, Employment and Society 20(2): 403–14.

Crompton

Brockmann

Lyonette

(2005) Attitudes, women’s employment and the domestic division of labour: a cross-national analysis in two waves. Work, Employment and Society 19(2): 213–33.

Daly

Rake

(2003) Gender and the Welfare State: Care, Work and Welfare in Europe and the USA. Cambridge: Polity Press.

Duncan

Edwards

(1997) Lone mothers and paid work: rational economic man or gendered moral rationalities? Feminist Economics 3(2): 29–61.

10.

Duncan

Edwards

(1999) Lone Mothers, Paid Work and Gendered Moral Rationalities. Basingstoke: Macmillan.

11.

Duncan

Edwards

Reynolds

Alldred

(2003) Motherhood, paid work and partnering: values and theories. Work, Employment and Society 17(3): 309–30.

12.

Eymard-Duvernay

(2006a) L’économie des conventions, méthodes et résultats. Tome 1, Débats. Paris: Editions La Découverte.

13.

Eymard-Duvernay

(2006b) L’économie des conventions, quinze ans après. Tome 2, Développements. Paris: Editions La Découverte.

14.

Festinger

(1957) A Theory of Cognitive Dissonance. Stanford, CA: Stanford University Press.

15.

Gash

(2008) Preference or constraint? Part-time workers’ transitions in Denmark, France and the United Kingdom. Work, Employment and Society 22(4): 655–74.

16.

Gilbert

(1992) On Social Facts. Princeton, NJ: Princeton University Press.

17.

Gilbert

(2008) Social convention revisited. Topoi 27(1/2): 5–16.

18.

Greenstein

(2009) National context, family satisfaction, and fairness in the division of household labor. Journal of Marriage and Family 71(4): 1039–51.

19.

Haas

Steiber

Hartel

Wallace

(2006) Household employment patterns in an enlarged European Union. Work, Employment and Society 20(4): 751–71.

20.

Harrison

Fitzgerald

(2010) ‘A chorus of disapproval?’ European attitudes to non-traditional family patterns. In: Park

. (eds) British Social Attitudes: The 26th Report. London: Sage, 135–58.

21.

Himmelweit

Sigala

(2004) Choice and the relationship between identities and behaviour for mothers with pre-school children: some implications for policy from a UK study. Journal of Social Policy 33(3): 455–78.

22.

Hirdman

(1994) Women – from Possibility to Problem? Gender Conflict in the Welfare State. The Swedish Model, Research Report 3. Stockholm: Arbetslivscentrum (The Swedish Centre for Working Life, now Institute of Working Life).

23.

Hobson

(2011) The agency gap in work-life balance: applying Sen’s capabilities framework within European contexts. Social Politics 18(2): 147–67.

24.

Latsis

De Larquier

Bessis

(2010) Are conventions solutions to uncertainty? Contrasting visions of social coordination. Journal of Post Keynesian Economics 32(4): 535–58.

25.

Le Feuvre

(2010) Feminising professions in Britain and France: how countries differ. In: Scott

. (eds) Gender Inequalities in the 21st Century: New Barriers and Continuing Constraints. Cheltenham: Edward Elgar, 126–49.

26.

Leira

(2002) Updating the gender contract? Childcare reforms in the Nordic countries in the 1990s. NORA – Nordic Journal of Feminist and Gender Research 10(2): 81–9.

27.

Letablier

M-T

(2004) Work and family balance: a new challenge for policies in France. In: Giele

Holst

(eds) Changing Life Patterns in Western Industrial Societies. Oxford: Elsevier, 189–210.

28.

Lewis

(1969) Convention: A Philosophical Study. Cambridge, MA: Harvard University Press.

29.

McGinnity

Whelan

(2009) Comparing work-life conflict in Europe: evidence from the European Social Survey. Social Indicators Research 93(3): 433–44.

30.

O’Reilly

(2009) Conflits de genre, contrats et statuts: une exploration de nouveaux cadres pour l’analyse comparative. In: Nicole-Drauncourt

(ed.) Conciliation Travail-Famille: Attention Travaux. Paris: L’Harmattan, 29–44.

31.

O’Reilly

Fagan

(1998) Part-Time Prospects: International Comparisons of Part-Time Work in Europe, North America and the Pacific Rim. London: Routledge.

32.

O’Reilly

Spee

(1998) The future regulation of work and welfare: time for a revised social and gender contract? European Journal of Industrial Relations 4(3): 259–81.

33.

Pfau-Effinger

(2004) Development of Culture, Welfare States and Women’s Employment in Europe. Aldershot: Ashgate.

34.

Pollert

(2003) Women, work and equal opportunities in post-communist transition. Work, Employment and Society 17(2): 331–57.

35.

Saraceno

Keck

(2010) Can we identify intergenerational policy regimes in Europe? European Societies 12(5): 675–96.

36.

Schmitt

Trappe

(2010) Introduction to the special issue: gender relations in Central and Eastern Europe – change or continuity? Journal of Family Research/Zeitschrift für Familienforschung 22(3): 261–5.

37.

Schober

Scott

(2012) Maternal employment and gender role attitudes: dissonance among British men and women in the transition to parenthood. Work, Employment and Society 26(3): 514–30.

38.

Sen

(1990) Gender and cooperative conflicts. In: Tinker

(ed.) Persistent Inequalities. Oxford: Oxford University Press, 123–49.

39.

Steiber

Haas

(2010) Begrenzte Wahl – Gelegenheitsstrukturen und Erwerbsmuster in Paarhaushalten im europäischen Vergleich [Limited choice structures of opportunity and employment patterns in European couple households]. Koelner Zeitschrift fuer Soziologie und Sozial Psychologie 62(2): 247–76.

40.

Streeck

Thelen

(2005) Introduction. In: Streeck

Thelen

(eds) Beyond Continuity: Institutional Change in Advanced Political Economies. Oxford: Oxford University Press, 1–39.

41.

Thévenon

(2006) Régimes d’état social et convention familiale: une analyse des régulations emploi-famille. Economies et Sociétés, Série Socio-économie du Travail 27(6): 1137–71.

42.

Thévenot

(2009) Governing life by standards: a view from engagements. Social Studies of Science 39(5): 793–813.

43.

Thompson

(1991) Family work: women’s sense of fairness. Journal of Family Issues 12(2): 181–96.

44.

Warren

(2000) Diverse breadwinner models: a couple-based analysis of gendered working time in Britain and Denmark. Journal of European Social Policy 10(4): 349–71.

45.

West

Zimmerman

(2009) Accounting for doing gender. Gender and Society 23(1): 112–22.