Microsoft Word Socialized Choices 31-12 pod docx

Chapter 4 - The vital and stabilising role of work preferences 
127 
attitudes, especially in highly tolerant societies such as the Netherlands, as if all 
behaviours are regarded as acceptable (Hakim, 2003c, p.341). 
Education has a similar effect on a mother’s preferred number of work hours 
as on her actual work hours. However, the effect of age is the opposite. As the 
years pass, a mother prefers to work more hours, up till the age of 43, after which 
she prefers fewer work hours. The reason for this different impact of age on 
actual work hours and preferred work hours is not clear, and would be an 
interesting subject for future research. Possibly, it is related to a growing 
discrepancy between preferences and actual labour market opportunities as 
women become older. Remarkably, neither the presence nor the income of a 
partner relates to a mother’s preferred number of work hours.  
Finally, the analysis shows a profound effect of the presence of a working 
mother during the respondent’s childhood on the subsequent work preference and 
the gender and work attitudes of the now-adult daughter. Other recent Dutch 
studies already demonstrated that having had a working mother affects Dutch 
mothers’ work hours (Lut, Van Galen and Latten, 2010; Van Putten et al., 2008) 
and their participation decision (Cloïn, 2010). The analysis confirms these results. 
The effect of a mother in paid work seems especially mediated by her daughter’s 
work preference. Moreover, the results suggest that the presence of a working 
mother in childhood is more significant than previous studies have shown, since a 
working mother during childhood (up to the age of 12) also affects her daughter’s 
general gender values and personal gender and work attitudes, corresponding 
with work preferences. This makes the total effect of a working mother in 
childhood on the respondent’s work preferences significantly large. The result is 
not only interesting in and of itself, but also reveals the constant role played by 
preferences, which might otherwise be much more subject to changing 
circumstances. Based on the results, I assume that in the Netherlands, the 
preferred number of hours can be – to a certain extent – a valuable predictor of a 
mother’s employment activity. 

---

 
 
 

---

 
129
 
Chapter 5 
The social origins of Dutch mothers’ gender values and  
ideal family life
44
 
5.1 Introduction 
This chapter traces the origins of two kinds of gender attitudes, namely general 
gender values (i.e. the gendered division of labour that a mother deems 
appropriate for other people), and more personal lifestyle preferences (i.e. a 
mother’s own ideal family life). A mother’s general gender value is measured 
against a scale based on seven questions, which originate from the European 
Social Survey (2009). Her personal gender attitude, or her ideal division of labour 
with her own partner, is based on Hakim’s (2000) question regarding a mother’s 
ideal family life. The distinction between general values and personal ideals is 
important, because previous studies have shown that personal attitudes are a 
better predictor of a mother’s labour market behaviour than her general gender 
values (Hakim 2000, 2003c; Marks and Houston 2002a; Risman et al., 1999).  
Therefore, the main dependent variable of this study is ideal family life, while  
mothers’ general gender values are understood as underlying basic elements that 
guide a mothers’s ideal family life (Ajzen, 1991; Becker, Van Deth and 
Scarbrough, 1995, p.38; Van Enckevort and Enschedé, 1983, p.19-20). Firstly, 
this chapter investigates how mothers’ ideal family lives – directly or indirectly 
intermediated by her general gender values – are formed throughout their lives by 
specific parental messages. Secondly, I shall examine the influence of perceived 
career support from significant others, both at the onset of adulthood from 
teachers and later in life through parents, peers, partners and work colleagues on 
mothers’ general gender values and ideal family lives. 
Previous studies have shown that, besides micro-economic factors such as 
education and income, women’s different gender attitudes are important in 
understanding their diverse labour market patterns (Beets, Liefbroer and de Jong, 
1997; Bolzendahl and Meyers, 2004; Cloïn, 2010; Cunningham et al., 2005; 
Hakim, 2000, 2003c; Himmelweit and Sigala 2004; Hoffnung, 2004; Kan, 2007; 
Marks and Houston, 2002a; Risman, Atkinson and Blackwelder, 1999; Steiber 
and Haas, 2009).  Nonetheless, few studies have gone beyond current gender 
attitudes and addressed the socializing factors formed prior to the experience of 
                                                           
44
  This chapter is based on a paper that is provisionally accepted for publication by Sex Roles: a 
Journal of Research (Springer). 

---

Socialized Choices - Labour Market Behaviour of Dutch Mothers 
130 
motherhood and the combination of care and work commitments (Blair-Loy, 
2003).  
In 2010, 32.4 per cent of Dutch mothers were not in paid employment, 42.5 
per cent worked from 12 to 24 hours a week, 13.8 per cent worked 25 to 35 hours 
a week, and 11.3 per cent worked 36 hours or more (CBS Statline, 2011). The 
typical Dutch pattern is often related to the statutory law and collective labour 
agreements, which have created the opportunity for Dutch mothers to work part-
time (Tijdens, 2006). The relatively large variation makes the Netherlands an 
interesting case on which to base a study of the causes of individual labour 
market decisions by mothers. Why do some mothers have a full-time job, while 
most mothers work part-time, and others are not employed at all? 
The main research question is: To what extent can diverse socialization 
processes explain the variation of Dutch mothers’ gender values and ideal family 
life? 
I test the hypotheses of this study with structural path analyses of data 
collected in 2010 in a representative survey among 935 Dutch mothers, with at 
least one child below the age of 13 living at home. The advantage of a structural 
path model is that it enables us to examine the relationship between more than 
one dependent variable (general gender values and ideal family life) and a 
number of independent variables in a single regression analysis. However, the 
analysis has some limitations because of the cross-sectional research design.  
5.2 
Macro background: the specific case of the Netherlands  
For a long time, Dutch female labour participation, especially among married 
women, was very low. In 1960, 25 per cent of women (Tijdens, 2006), and 7 per 
cent of married women were employed, compared to 30 per cent of English and 
33 per cent of French married women (Kloek, 2009).  Since the 1960s, as in many 
other Western countries, Dutch society changed radically through processes of 
secularization, increased educational attainment of women, and the greater 
acceptance of non-familial roles for women and familial roles for men (Sullivan, 
2004), which have contributed to the changes in the “male-breadwinner family 
model”. As a result, the female participation level had increased to 35 per cent by 
1985 (Tijdens, 2006). This rise continued throughout the following decades, 
resulting in one of the highest levels of female participation among the Western 
countries, at almost 70 per cent in 2009 (OECD, 2013). However, this very strong 
rise in female labour participation was accompanied by an equally strong rise of 
part-time employment. Thus, most of the rise of female participation was in part-
time work (OECD, 2013).  
Specific national characteristics can explain the heterogeneous employment - 
yet dominantly part-time - pattern of Dutch women (Plantenga, 2002; Tijdens, 
2006; Van Doorne-Huiskes and Schippers, 2010). In the 1970s and 1980s, when 
female labour participation started to rise, there were few public childcare 
facilities, which forced mothers who entered the labour market to combine their 

---