THE GENDER WAGE GAP IN THE NETHERLANDS IN 2012

(1)

THE GENDER WAGE GAP IN THE

NETHERLANDS IN 2012

WHY DO WOMEN STILL EARN LESS THAN MEN WHILE

THEIR PRODUCTIVITY DISADVANTAGE DISAPPEARS?

Master Thesis, Msc Human Resource Management

University of Groningen, Faculty of Economics and Business

June, 5, 2016

BAS KNIKHUIS

S2034255

Korreweg 71a

9714AC Groningen

Email: B.J.H.Knikhuis@student.rug.nl

Word Count: 8657

Supervisor

Peter van der Meer

(2)

ABSTRACT

Although the gap in characteristics that have positive effect on the wages of men and women is diminishing, international research shows that in many countries the gender wage gap still exists. However, one can argue that the gender wage gap in the Netherlands is declining because of the increase in female human capital and the contribution of women in the labor market. Based on former research in 1996 in which the gender wage gap was approximately 20%, this research aims to answer the question whether the gender wage gap in the Netherlands declined between 2006 and 2012. It also examines if and why men and women have different age-earning profiles and if men and women have different reward package preferences. In line with the expectations, the results showed a declining gender wage gap. The data in the OSA Labour Supply Panel show a gender wage gap of 6,8% in 2012 compared to 8,3% in 2006. Strikingly, while women are more productive than men in almost every age-earning profile, men’s net wage is higher. It seems that this difference is primarily caused by the underpayment of women. This thesis confirms that men generally have steeper age-earning profiles and that it cannot be said that women have different reward package preferences.

(3)

INTRODUCTION

Background

The overall differences between men and women in productivity and education have been declining in recent decades. Why then do income differences between men and women still exist? This is a frequently asked question around the world, but one that is rather difficult to answer. For example, in the Dutch market not only the position, but also the contribution of women in the labor market has significantly changed. The female working population increased from 3,4 million (2005) to 3,8 million (2015), an increase of 11,8% (CBS, 2015). Besides this, women in the Netherlands occupy more and more high-level jobs. It is well known that gender wage differences exist, but has the gap declined in recent years?

One would expect that the gender wage gap in the Netherlands would be declining. However research provided by Van der Meer (2008) showed no declining wage gap in the Netherlands between 1985 and 1996. In later years, researchers did find a decline in the gender wage gap in the Netherlands from 20% in 1996 to 18% in 2006 (Fransen, Plantenga & Vlasblom, 2011). Women’s labor participation has increased and their economic status has changed in recent years. Besides the educational differences between men and women are negligible. Differences in the gender wage gap in certain age earning profiles might also exist. Generally, after graduation men and women start with a comparable salary that is relatively low. When becoming more experienced wages start to increase. Since nowadays more women complete higher education programs, this could lead to a narrowing gender wage gap. Besides women benefit from completing higher education programs, more women become more experienced.

(4)

According to Van der Meer (2008), one explanation for the wage gap is the higher proportion of men in top-level functions. This theory is still applicable, however the Central Bureau of Statistics (2014) found an increase in the percentage of women in top-level functions from 7% in 2007 to 15% in 2013, which might explain a declining wage gap between men and women in the Netherlands. In this paper research is presented that elaborates on the research of Van der Meer (2008) about the gender wage gap in the Netherlands.

Traditionally, researchers mention several causes of the gender wage gap. First, when looking at women’s labor market behavior, Van der Meer (2008) implied that women nowadays tend to withdraw less often from the labor market due to marriage or children than they did 10 years ago. Additionally women increasingly occupy professional and high-level positions (Kanter, 1977; Mandel, 2012, 2013) and work more hours than they used to do.

Second, human capital is often mentioned as an important cause of the gender wage gap. Since the lifetime-work expectations of women have changed, which means that they invest more in their own human capital such as skills, education, and experience, it could be that the gender wage gap declines. However, there seems to be a difference in skills, such as negotiation skills in the job hiring process. This could explain lower wages for women (Bowles & Babcock, 2007)

The third often mentioned cause of the gender wage gap is occupational segregation. Women tend to work more in female-dominated, low-paid occupations, such as healthcare. Traditionally, only a few women work in high-paying masculine sectors. Regardless of the sector, women tend to be treated differently than men (Van der Meer, 2008; Mandel & Semyonov, 2005; England, 1992).

According to Becker (1957), employers have a “taste” for discrimination. In his view, minority workers (e.g. women) have two options; they can “compensate” their employers by being more productive at a given wage or they have to accept lower wages for identical productivity.

There is currently an ongoing discussion on whether women are discriminated against in the labour market. Do they not just prefer different reward packages than men? In this case one might conclude that the wage difference is not a discriminating factor (Fransen et al., 2011).

(5)

impact of each of the sources might change over time. Therefore, it is possible that they play a different role in the current gender wage gap.

Examining the theoretical and practical contributions, no research has systematically examined whether the decline in the gender wage gap has been different in different age categories in the Netherlands. Van der Meer (2008) found smaller differences among young people than old, but did not specify these differences. Looking at the before mentioned trends for women about human capital, occupational segregation, labour market behavior, discrimination, and different preferences, this might imply that gender wage differences would decrease. Knowing this would be helpful both for young female graduates to better understand what their future potentials are when applying for a job and for employers in their recruitment process.

Research Question

Besides the fact that the participation of women in the labour market increased and that their economic status has changed, it might be that the increasing proportion of women in top-level functions might indicate a decline. Furthermore the increase in their lifetime-work expectations that leads to higher levels of human capital might cause a narrowing of the gender wage gap. The central question of this piece of research that therefore arises is:

Did the gender wage gap in the Netherlands decline between 2006 and 2012?

Based upon the research question there are three sub questions that will be answered: 1. What are the causes of the wage gap in the Netherlands?

2. What is the influence of different age-earning profiles on the gender wage gap? 3. Do women have other reward-package preferences?

Thesis Structure

(6)

THEORY

Introduction

This chapter defines key terms and reviews the literature on the gender wage gap. The gender wage gap can be defined as the difference in men’s and women’s median earnings divided by men’s median earnings. The gender wage gap implies that women earn less than men for the same productivity or outcomes (Schippers, 1987).

Most studies divide the gender wage gap into two components: the “explained” and the “unexplained” gender wage gap (Blau & Kahn, 1997; Tijdens et al., 2002; Van der Meer, 2008). Economists define the “explained” gap as the differences in work characteristics or price differences between men and women. This research also tries to include whether women had different preferences (that are expressed in the cost of labor) over the years. For example, one can imagine that women’s market position has improved due to emancipation, the availability of well-paid part-time jobs, and their availability for labor. Therefore, wage differences may decline. Furthermore different gender preferences with respect to reward packages are not much concluded in income inequalities. This thesis tries to fill that gap. If women attach less importance to wages than to secondary working conditions, it might initially seem that they are discriminated against, but in fact their employer meets their needs. It must be kept in mind that a reward package not only contains wages but also non-monetary benefits.

The “unexplained” gap is often associated with discrimination in the market (Weichselbaumer & Winter-Ebmer, 2005; Becker 1957). Blau and Kahn (2006) attribute the unexplained gap to a decrease in labor market discrimination of women and to a decline in the unmeasured characteristics.

Explained Gender Wage Differences

According to Van der Meer (2008), women’s participation in the labor market increased. Whereas women previously wanted to stay at home to take care of their children, nowadays an increasing number of women also want to work (Hakim, 2002). This might look like it would cause a declining wage gap, although according to Van der Meer (2008), higher participation of women does not necessarily lead to a decrease in the gender wage gap until 1996. One explanation for this could be that women tend to work in medium- and low-paying industries.

(7)

recent years, which may decrease the gender wage gap. Another evolution in the Dutch labor market is the so-called “Papa dag”. This law was passed in 2000 and made it possible for men to combine their job with their caring responsibilities. It might seem that that this only benefits men, but in practice it also benefits women. When husbands stay at home to take care of the children, wives are able to work more hours, and thus become more experienced at their jobs. Therefore, there is reason to believe that the aforementioned changes in the labor market caused a decline in the gender wage gap.

In addition to the changing labor market behavior, human capital also has to be taken into account when explaining wage differences. The literature contains various definitions for human capital. De la Fuente and Ciccone (2002) stress the knowledge and skills obtained throughout educational activities. A further definition is given by Sheffrin (2003), who describes human capital as the stock of skills and knowledge embodied in the ability to perform labor so as to produce economic value. Building on these definitions, this thesis defines human capital as the skills, education and experience of individuals. An excellent example of a human capital theory refers to the male-female lifetime work expectations (Polacheck, 2004). This theory suggests that a person’s incentive to invest in training (i.e., education) is proportional to their lifetime working expectations. Since on average men work more hours than women over their lives, one would expect women to invest less in human capital. Traditionally, this was true. However, women’s rising labor force participation in relation to men’s implies that their human capital investments have increased. In the Netherlands, more women (42%) than to men (36%) possess advanced degrees in 2009. Women’s increasing labor participation and experience combined with their higher educational level would lead to the expectation that the gender wage gap will be smaller now than in the past.

(8)

of the lack of men in these sectors, men tend to receive higher rewards for working in such occupations even when they do not have a productivity advantage.

The above factors tend to broaden the wage gap between men and women. However, the growing integration of women into new occupational domains alters some of these factors. Women’s increased participation in new professional and high-level positions and occupations will therefore probably narrow the wage gap (Kanter, 1977; Mandel, 2012, 2013). Where women traditionally were under-represented or even absent in these occupations, new insights provided by the Central Bureau of Statistics (2014) showed an increase in the number of women in top-level functions in the Netherlands from 7% in 2007 to 15% in 2013.

Unexplained Gender Wage Differences

Employers’ discrimination is mentioned as an important cause of the gender wage gap. Discrimination is a broad term, and it can arise in a broad variety of ways. In Becker's (1957) model, discrimination is defined and caused by the discriminatory tastes of employers, co-workers or customers. However, these tastes should diminish over time. Throughout this thesis, discrimination is defined as a negative treatment towards members of a minority group compared to members of a majority group, when both have the same productive characteristics. One might wonder whether this minority theory is still applicable to the Dutch labor market, as more and more women are entering the workforce. However, according to the CBS (2014), women are still underrepresented in top-level functions and in occupations at the top of the wage distribution. Besides, women in the Netherlands believe that women are disadvantaged and discriminated against. It therefore cannot be ruled out that there is some sort of discrimination against women in the Netherlands.

As mentioned earlier, according to Becker (1957), employers have a “taste” for discrimination. Translating this to the minority group of women in this paper, this means that they are less likely to be hired by employers. Also “statistical discrimination” is a term often associated with discrimination. These models describe the process of discrimination against minorities who are more productive than the average worker in a company.

(9)

woman quitting is higher than that of a man.” One can imagine that this unexplained factor thus is of great importance in the gender wage equation. Recent research observed a significant decline in the unexplained portion (e.g. discrimination) of the gap in the United States (Blau & Kahn, 2006). It is interesting to investigate whether the unexplained portion of the gap is also declining in the Netherlands. Because of the above, it cannot be ruled out that Becker’s discrimination theory is still applicable.

Empirical findings in previous research about age-earning profiles suggest that the gender wage gap is smaller in early carriers. After graduating, men and women start with the same base salary, which increases, as they get older and more experienced. This is confirmed by Lazaer (1974), who found that age is an important factor in the wage increases of men and women, especially between 14 and 24 years old. The gender wage gap might increase with age due to for example differences in on the job training perspectives and gender investment in human capital. Generally, women also have more often interruptions in their career, which causes lower level of experience.

(10)

(11)

METHOD

Data Collection

Data for this research was gathered from the labor supply panel of OSA (Organisatie voor Strategisch Arbeidsmarkt onderzoek) of 2006 and 2012. This is a labor market survey held among approximately 5000 respondents of the current Dutch labor force. The first survey was held in 1985 and the latest data of 2012 is recently published. Data was gathered every two years and the panel survey contains 150 questions. OSA provides descriptive data about age, experience and job tenure. Furthermore, it provides work-related information such as supervision responsibilities, firm size, the net monthly income, and working hours. The net monthly income and working hours were used to calculate the wage per hour. Participants were recruited within a broad variety of occupations, ranging from agriculture to healthcare. The OSA labor supply dataset is used to increase comparability with former research about the gender wage gap that also used this data archive (Van der Meer, 2008).

Data Analysis

Because this research was aimed at gender wage differences, a few criteria for the final dataset have been met. The final dataset only consists of observations of those who are currently employed at an employer and are between 16 and 65 years old. Cases with missing or incorrect information such as hourly wages below 6 euro were excluded. From the 8610 (N = 8610) approached participants (2006), the number of response was 5563. This is a response rate of 64%. Controlling for incorrect information and missing values, the final data set consisted of 2600 persons. Participants’ characteristics for 2006 were as follows: 1365 participants were male, 1235 were female. Their age was ranged from 16 to 65 years old with a male mean age of 44 years (SD =11,2) and a female mean age of 41 (SD = 10,2) Men’s tenure ranged from 1 to 49 years and women’s from 1 to 44 years.

(12)

There is a broad variety of ways to measure wage differentials between men and women. In this research the wage equation for men and women has been estimated by comparing their average hourly wage and their logarithms. The difference between this estimation is the so-called “gender wage gap”. This can be done by the model presented below where Wm is the

hourly wage of men and Wf the hourly wage of women.

(1) Ln (Gwf + 1) = Ln (Wm) – Ln (Wf)

However, because the above wage differential does not adjust for productivity differences between men and women, one should add more variables in the regression equation. By adding more variables it became visible how much of the former estimated gender wage gap can be explained by the productivity differences between men and women. Information about the unexplained gap (e.g. discrimination) is obtained by differences in production characteristics. By combining differences in gender between the wage equations and production characteristics, information about the unexplained gap of the wage differences (e.g. discrimination) could also be obtained.

There are several techniques for decomposing wage gaps by using regression equations. One of the most common procedure is proposed by Oaxaca and Blinder (1973) in which they separate linear regression models for men and women into a portion that is explained by gender differences in work-related characteristics and a residual part that cannot be explained. The first step in the decomposition is as follows:

(2) Ln (Gwf + 1) = βm (xm - xf) + xf (βm – βf)

To estimate the “true” wages equation the parameters should be weighted. This research elaborated on the weighting scheme of Oaxaca and Ransom (1994). In this weighting scheme the precision of the parameter estimates act as weight. By doing so, higher variation in the parameters leads to lower weight. The next step in the decomposition was:

(3) Ln (Gmf + 1) = Xm (βm – β*) + (β* - βf) + (Xm – Xf)’ β*

This model shows the overpayment of men, the underpayment of women and the productivity difference on the right hand side. β* stands for the “true” parameter. In addition to the

(13)

categories. The used categories are; 16 to 24 years old, 25 to 34 years old, 35 to 44 years old, 45 to 55 years old and 55 to 66 years old.

To estimate the wage equations, linear regression has been used. The used variables in the regression analysis are real tax after hourly wage, log hours worked, educational level, gender and marital status, job level, total labor market experience, performing unpaid overtime, being a supervisor, log firm size, having changed employer in the last two years, having changed the job at the current employer, having a permanent position, and having small children at home. In order to investigate the changing preferences (e.g. price

differences) of men and women over the years with regard to reward packages working at home, working at home to take care for children, hours worked, and career perspectives are included. The variable career perspectives is composed out of three variables; I am satisfied with career development possibilities at my current employer, I see enough career perspective in my job, I can develop myself. For 2006 Cronbach Alpha was .826 (α = .826) and for 2012 Cronbach Alpha was .824 (α = .826).

The preference variables could be a cause for a declining wage gap between men and women because the career perspective of women is evolving and by giving them the opportunity to work at home, more women are able to work more hours. To test whether a difference of preference between men and women exists, interaction effects between gender and those variables have been calculated. By comparing the parameters it could be concluded if there is a significant difference in preferences between men and women.

(14)

RESULTS

It was predicted that the wage gap between men and women declined between 2006 and 2012. Linear regression was conducted to estimate the wage equations and to answer this question. This chapter discusses the results of this research, which are concentrated in several tables.

The analysis in this chapter clarifies that the gender wage gap declined from 8,3% in 2006 to 6,8% in 2012. The absolute difference in pay per hour for men increased by 14,2% (€1,66) between 2006 and 2012 and in that same period the absolute difference in pay per hour for women increased by 16,7% (€1,79).

The descriptive statistics are presented in Table 1, divided by gender. It can be seen that most of the variables show a significant difference between men and women. Thus men attain higher job levels than women, are more experienced, have higher tenure and work more unpaid overtime. One remarkable finding is that the mean for men performing a supervisory role is approximately twice as large as for women. Next to this, men work in larger firms, more often have children, and more often have a partner, and work longer hours. Men perceive more career perspectives than women. It can be seen that in 2006 women had a small educational advantage, while in 2012 men had a small educational advantage. In 2006, women changed employers more often than men, while this rate is approximately equal in 2012. In contrast with this, the mean of men and women that changed jobs internally in 2006 was equal, while there was a significant difference in 2012, where women changed their job internally more often.

TABLE 1

Descriptive Statistics by Gender in 2006 and 2012 (Mean and SD)

2006 2012

Men Women Men Women

Mean SD Mean SD Mean SD Mean SD

(15)

Tenure 4,38* 1,14 4,11 1,01 4,54* 1,14 4,35 1,11 Sq. Tenure 20,19* 9,91 17,93 8,24 21,86* 9,91 20,16 9,43 Working at Home 0,20* 0,41 0,18 0,38 0,21 0,41 0,19 0,39 Working home care 0,03 0,22 0,03 0,18 0,05 0,22 0,04 0,20 Career Perspectives 9,50 2,63 9,36 2,77 9,59* 2,63 9,34 2,72 Log Wage 2,46* 0,16 2,37 0,12 2,59* 0,18 2,52 0,13

Absolute Wage 11,70 10,70 13,36 12,49

Nr of Cases 1365 1235 1260 1256

Source: OSA Labour Supply Panel, own calculations Note: * p>0,05

(16)

TABLE 2

Regression Results of Log (Wage) for Men in 2006 and 2012

2006 2012 Parameter SD Parameter SD Education 0,079 0,008 0,043 0,011 Job Level 0,065 0,008 0,097 0,011 Experience -0,017 0,039 0,097 0,043 Sq. Experience 0,002 0,008 -0,013 0,008 Unpaid Overtime 0,086 0,015 0,044 0,015 Supervisor 0,043 0,013 0,063 0,015

Log Firm Size 0,017 0,003 0,021 0,004

Change Employer 0,041 0,023 -0,008 0,023

Internal Job Change 0,004 0,018 -0,008 0,002

Permanent Contract 0,083 0,027 0,038 0,024

Children at Home 0,003 0,016 0,01 0,018

Partner 0,036 0,021 0,027 0,021

Log Hour Work -0,329 0,029 -0,457 0,029

Tenure 0,341 0,080 0,264 0,082

Sq. Tenure -0,029 0,010 -0,026 0,009

Working at Home 0,016 0,018 0,039 0,018

Working Home to Care 0,023 0,039 0,09 0,032

Career Perspectives 0,016 0,002 0,016 0,003

Constant 1,812 0,158 2,571 0,176

Adj. R2 0,462 0,484

SE 0,2278 0,237

N 1365 1260

Source: OSA Labour Supply Panel, own calculations

(17)

women in 2012 compared to men in 2012. The negative effect on wages of working part-time is smaller for women. In 2006, the effect of working at home was larger for women than for men. Women working at home had 4,3% higher wages in 2006 and this percentage increased to 6,6% in 2012. The effect for having career perspectives is larger for men compared to women, although this effect is small.

TABLE 3

Regression Results of Log (Wage) for Women in 2006 and 2012

2006 2012 Parameter SD Parameter SD Education 0,083 0,008 0,033 0,011 Job Level 0,077 0,008 0,125 0,011 Experience 0,079 0,034 0,004 0,039 Sq. Experience -0,007 0,008 0,006 0,008 Unpaid Overtime 0,012 0,014 -0,026 0,015 Supervisor 0,023 0,015 0,063 0,016

Log Firm Size 0,015 0,003 0,007 0,003

Change Employer 0,003 0,019 0,031 0,024

Internal Job Change 0,008 0,018 0,028 0,018

Permanent Contract 0,017 0,022 0,095 0,024

Children at Home 0,042 0,016 0,035 0,017

Partner -0,011 0,016 -0,033 0,017

Log Hour Work -0,146 0,016 -0,193 0,018

Tenure 0,063 0,060 0,238 0,064

Sq. Tenure -0,006 0,007 -0,025 0,007

Working at Home 0,043 0,018 0,066 0,019

Working Home to Care 0,006 0,034 0,048 0,033

Career Perspectives 0,005 0,002 0,006 0,002

Constant 1,837 0,116 1,829 0,13

Adj. R2 0,352 0,380

SE 0,208 0,225

N 1235 1256

Source: OSA Labour Supply Panel, own calculations

(18)

aged 45 to 54 and 55 to 66 are in the advantage of men, although not by much. Table 4 also shows that the wage differences in 2006 are not caused by the overpayment of men, but that it seems to be caused by the underpayment of women. A surprising result, and perhaps an outlier, is that women in 2006 between 35 and 44 years old were underpaid by 68%. Only women between 16 and 24 years old were not underpaid. This seems to be a reason that the wage gap in this age category is decreasing. Looking at 2012, it also can be seen that women were underpaid by approximately 15%, varying between 6% and 40% in certain age categories.

TABLE 4

The Gender Wage Gap and its Decomposition

Age Category Net Wage Gap

Overpayment of Men Underpayment of Women Productivity Differences Whole sample 2006 0,083 -0,048 0,147 -0,015 16-24 -0,145 -0,051 -0,023 -0,071 25-34 0,008 -0,148 0,171 -0,015 35-44 0,053 -0,528 0,683 -0,102 45-54 0,100 -0,106 0,187 0,019 55-66 0,134 -0,605 0,713 0,026 Whole sample 2012 0,068 0,004 0,122 -0,058 16-24 0,001 0,025 0,064 -0,088 25-34 -0,022 -0,250 0,302 -0,074 35-44 0,038 -0,246 0,397 -0,113 45-54 0,131 -0,090 0,279 -0,058 55-66 0,107 0,020 0,098 -0,011

Source: OSA Labour Supply Panel, own calculations

Table 5 contains results on the question of whether women have other preferences than men with regard to their reward packages. The interaction effect for the variables included in the analysis to test differences in preferences in 2006 for working at home and working at home to take care of children show a difference in their parameters, but this difference is not significant. This seems logical, because male and female means do not show significant differences (Table 1).

(19)

log hours worked and having career perspectives, we see that women working more hours earn higher wages and that women have less career perspectives compared to men.

TABLE 5

Regression Results Preference Variables of Log (Wage) and Gender Interactions

2006 2012

Parameter p-value Parameter p-value

1. Working at Home 0,016 0,336 0,090 0,027

2. Log hours Worked -3,290 0,000 -0,457 0,000

3. Working at Home to Care 0,023 0,542 0,016 0,004

4. Career Perspectives 0,016 0,000 0,016 0,000

5. Gender*Working at Home 0,027 0,284 0,027 0,288 6. Gender*Log Hours Worked 1,830 0,000 0,264 0,000 7. Gender*Working at Home to Care -0,017 0,743 -0,043 0,358 8. Gender*Career Perspectives -0,011 0,001 -,010 0,005

(20)

DISCUSSION

This research examined the gender wage gap in the Netherlands. A question is posed of whether the gender wage gap declined between 2006 and 2012, elaborating on research in 1996, when the gender wage gap was 18% (van der Meer, 2008). This question has been posed because labour market differences between men and women declined over the years. Besides, the female working population increased and there were more women in high-level positions. It also seems that productivity differences between men and women narrowed, which should cause a narrowing of the wage gap. Examining different age categories should clarify whether there are different age earning profiles and whether the gender wage gap is smaller for younger workers. Furthermore, it is expected that some part of the gender wage gap could be explained by a difference in women’s preferences with regard to their reward package. For example, in the view of the Dutch society women are expected to combine their job with their caring responsibilities, thus if women prefer social rewards such as working at home more than economic rewards it seems that they are discriminated, but in fact the employer does meet their needs.

In line with these expectations, it seems that women’s current labour market behavior has led to a substantial decrease in the gender wage gap since 1996. The results of the OSA Labour Supply Survey, show that the gender wage gap declined to 8,3% in 2006 and 6,8% in 2012. Decomposition indicates that this decline in the gender wage gap is mainly caused by a decrease in the underpayment of women from 14,7% in 2006 to 12,2% in 2012. Indeed, it seems that women have a productivity advantage of approximately 6% in 2012. The largest part of the gender wage gap can therefore be explained by price differences, primarily the underpayment of women.

(21)

The question then arises, why are women underpaid and why does the gender wage gap still exist? It seems that the remaining gender wage gap is mainly caused by the underpayment of women of approximately 12%. This is in line with the expectation that productivity differences in the advantage of men will have narrowed over time. Indeed, it seems that women have a productivity advantage of approximately 6% in 2012. Comparing the findings in Table 1, one can see that the return on education for women dropped more drastically than for men. A possible explanation could be the collapse of the economy around 2008. Because as the demand for labour decreases, the relative supply of labour increases, employers could choose from more applicants. Because men tend to work more hours and because female careers are more often interrupted, employers prefer hiring men. Also more men work in larger firms and since larger firms pay higher rewards in the Netherlands, this could maintain the wage gap (Oosterbeek & Praag, 1995).

Another explanation for the gender wage gap could be the differences in negotiating skills between men and women. Various researchers acknowledge a difference in negotiating outcomes (Amanatullah & Morris 2010; Galinsky, Kray, & Thompson, 2001). Whereas men tend to act assertively and in their self-interest based on internal abilities or beliefs, women tend to behave defensively and behave more socially strategic.

The tendency for women to work in different kinds of occupations and industries continues to be an important source of the gender wage gap (Blau & Kahn, 2007). Women are entering more male-dominated occupations, such as law, medicine and engineering, but data from the Central Bureau for Statistics in the Netherlands (CBS, 2016) shows that there is still a significant difference in well-paid sectors such as Information Communication Technology (ICT) and technical occupations. Strikingly, it seems that women’s anticipated work-life balance and their future planning of having kids largely affects women’s job choices for certain sectors (Barbulescu & Bidwell 2013).

(22)

economic crisis indeed causes an increase in the gender wage gap. However it has to be taken into account that this is the unadjusted gap, thus that factors that influence the gender wage gap such as education and type of job etc. are not included. Therefore a proportion of the increase can be explained by a different kind of decomposition of the gender wage gap or due to a cohort effect in this thesis.

Results confirm that men generally have steeper age-earning profiles. An explanation for this could be that most women have an interruption at a crucial moment in their career to have children when they are about 34 years old. Paull (2008) found that the change of becoming a parent is greater for women than for men. Women typically interrupt and reduce their employment while men begin to work even more. This interruption often has damaging consequences for women’s careers and thus their age-earning profiles. Therefore, it can be said that the transition to parenthood still might be an important factor in the gender wage gap for men and women (Schober, 2013).

There have been several researches that investigate whether men and women have different reward package preferences, but results are inconsistent. Gender role socialization might be a reason for a difference in preferences for non-economic rewards. Dutch society emphasizes caregiving for children and other family members as a more central role for women than for men. However, I did not find that women prefer working at home, and working at home to care more than men. This seems logical, because the results in Table 1 do not show significant differences in men and women’s means.

(23)

Theoretical Implications

This research has important theoretical implications. First, this research contributes to the literature on gender wage gaps in the Netherlands. There have been numerous researches on the gender wage gap worldwide, however, in the Netherlands little specific research has been conducted in recent years. Especially the effect of the economic downfall in the gender wage gap is interesting to see. Second, this research also contributes to the literature about different age-earning profiles between men and women in the Netherlands, which has not been done in recent years.

Practical Implications

The results have implications both for the supply-side and the demand-side of the labour market. On the supply-side, this thesis can be particularly useful for women who want to have a business career. To improve their “glass-ceiling”, they might consider postponing having children to later in their career. After all, the early thirties are a crucial time in getting promotions and growing in one’s job.

A demand-side implication could be that employers should stop arguing that women are less productive than men. Of course one can argue that men are more productive in some jobs, but this thesis makes clear that in general, women are more productive. Arulampalam et al. (2007) found that the gender wage gap is smaller on the bottom of the wage distribution. Given women’s higher productivity, hiring more women for top-level positions should therefore both benefit employers and decrease the gender wage gap. Employers should take this into account in their decision making process for hiring new employees.

(24)

Strengths, Limitations and Future Research Implications

Focusing on different age-profiles strengthens the findings of this research. It provides more detailed information and allows for more detailed conclusions than only looking at the wage gap without taking age categories into account. Another strength of this research is that it consists of quantitative data, conducted by the OSA labour Supply Panel. OSA has a lot of experience in conducting quantitative data and recoding it into SPSS, which makes the data very consistent, precise and reliable. Results are therefore representative for the Dutch society.

The conducted research is limited by the fact that the scale used to measure the segregation level of men and women changed between the two periods investigated. The scale used in 2006 (SBC’92), was no longer used to measure segregation in 2012 because of a transition to an international definition of the labour force. Therefore, data about segregation could not be included in this research, while it is known from literature that segregation can have great effects on the gender wage gap. Furthermore, the dataset only consists of after-tax wages. The progressive Dutch tax system leads to a more equalized income after taxes. As women in households could pay lower taxes than men, the before-tax wage gap might be larger. The before-tax wages are not available in the dataset and therefore could not be used. However, it can be assumed that a decline in the wage gap after tax also indicates a decline in the wage gap before tax.

(25)

Conclusion

(26)

REFERENCES

Arulampalam, W., Alison L. Booth, A.L., & Bryan, M.L. 2007. Is there a glass ceiling over Europe? Exploring the gender pay gap across the wage distribution. Industrial & Labor Relations Review, 60.2: 163-186.

Amanatullah, E. T., & Morris, M. W. 2010. Negotiating gender roles: Gender differences in assertive negotiating are mediated by women’s fear of backlash and attenuated when negotiating on behalf of others. Journal of

personality and social psychology, 98(2), 256.

Barbulescu, R., & Bidwell, M. 2013. Do women choose different jobs from men? Mechanisms of application segregation in the market for managerial workers. Organization Science, 24(3), 737-756.

Becker Gary, S. 1957. The economics of discrimination. The University Of Chicago Press, Chicago.

Bekker, S., Kerkhofs, M., Roman, A., Schippers, J., Voogd-Hamelink, M., & Wilthagen, T. 2007. Trendrapport

Aanbod van arbeid 2007. OSA Publicatie A234, OSA, Tilburg.

Blau, F. D., & Kahn, L. M. 1997. Swimming upstream: Trends in the gender wage differential in the 1980s.

Journal of labor Economics, 1-42.

Blau, F. D., & Kahn, L. M. (2000). Gender differences in pay (No. w7732). National Bureau Of economic Research.

Blau, F. D., & Kahn, L. M. 2006. The US gender pay gap in the 1990s: Slowing convergence. Industrial & Labor

Relations Review, 60(1), 45-66.

Blau F.D., & Kahn L.M. 2007. The Gender Pay Gap. Econ Voice 4:4.

Blau, F. D., Brummund, P., & Liu, A. Y. H. 2013. Trends in occupational segregation by gender 1970–2009: Adjusting for the impact of changes in the occupational coding system. Demography, 50: 471–492. Bowles, H. R., Babcock, L., & Lai, L. 2007. Social incentives for gender differences in the propensity to initiate

negotiations: Sometimes it does hurt to ask. Organizational Behavior and Human Decision Processes, 103(1), 84-103.

CBS. 2009. Emancipatiemonitor 2008, Retrieved November 20, 2015, from; http://www.cbs.nl/nl- nl/menu/themas/arbeid-sociale-zekerheid/publicaties/artikelen/archief/2009/persbericht-emancipatimonitor.html

CBS. 2014. Emancipatiemonitor 2014: Meisjes kiezen steeds vaker voor techniek, meer vrouwen dan eerst in topfunctie, Retrieved November 19, 2015, from;

http://www.cbs.nl/nl-NL/menu/themas/dossiers/vrouwen-en-mannen/publicaties/artikelen/archief/2014/2014-080b-pb.html

CBS. 2016. Werkzame beroepsbevolking; beroep. Retrieved 15 April, 2016 , from;

http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=82808NED&D1=0&D2=a&D3=0&D4=0%2c6% 2c17%2c27%2c37%2c54%2c73%2c84%2c118%2c125%2c132%2c150%2c161%2c171&D5=19%2c49&VW =T

De la Fuente, A., & Ciccone, A. 2002. Human capital in a global and knowledge based economy. Institute for Economic Analysis.

De Ruijter, J., Van Doorne-Huiskes, A. & Schippers, J. 2003. Size and causes of the occupational gender wage-gap in the Netherlands, European Sociological Review, 19: 345–60.

Echtelt, P., Vlasblom, J.D, & Voogd-Hamelink, M. 2014. Trendrapport aanbod van arbeid 2014, SCP Publicatie 2015-3, Sociaal Cultureel Planbureau, Den Haag

(27)

Eurostat 2014. Tackling the gender pay gap in the European Union. Retrieved January 25, 2016, from: http://ec.europa.eu/justice/gender-equality/files/gender_pay_gap/140319_gpg_en.pdf

Fransen E., Plantenga , J., & Vlasblom J.D. 2012. Why do women still earn less than men? Decomposing the Dutch gender pay gap, 1996–2006, Applied Economics, 44:33, 4343-4354

Hakim, C. 2002. Lifestyle preferences as determinants of women’s differentiated labor market careers, Work and

Occupations, 29: 428–59.

Heilman, M. E. 1980. The impact of situational factors on personnel decisions concerning women: Varying the sex composition of the applicant pool. Organizational Behavior and Human Performance, 26(3): 386-395.

Hoeben, J., & Venema, P. M. 2004. De arbeidsmarktpositie van werknemers in 2002: een onderzoek naar de

verschillen in beloning en mobiliteit tussen groepen werknemers. Arbeidsinspectie.

Kanter, R. M. 1977. Men and women of the corporation. New York: Basic Books.

Kray, L. J., Thompson, L., & Galinsky, A. 2001. Battle of the sexes: gender stereotype confirmation and reactance in negotiations. Journal of personality and social psychology, 80(6): 942.

Lazear, E. 1776. Age, Experience and Wage Growth. American Economic Review, 66: 548-558.

Mandel, H., & Semyonov, M. 2005. Family policies, wage structures, and gender gaps: sources of earnings inequality in 20 countries, American Sociological Review, 70: 949–67.

Mandel, H. 2012. Occupational mobility of American women: Compositional and structural changes, 1980–2007.

Journal of Social Stratification and Mobility, 30(1): 5-16.

Mandel, H. 2013. Up the Down Staircase: Women's Upward Mobility and the Wage Penalty for Occupational Feminization, 1970-2007. Social Forces, 0:18.

O'Neill, J. E., & O'Neill, D. M. (2005). What do wage differentials tell us about labor market discrimination? (No. w11240). National Bureau of Economic Research.

Oaxaca, R. (1973). Male-female wage differentials in urban labor markets. International economic review, 693-709.

Oaxaca, R. L., & Ransom, M. R. 1994. On discrimination and the decomposition of wage differentials. Journal of

econometrics, 61(1); 5-21.

Oosterbeek, H., & Van Praag, M. 1995. Firm-size wage differentials in the Netherlands. Small Business

Economics, 7(3); 173-182.

Paull, G. 2008. Children and women's hours of work. The Economic Journal, 118(526): F8-F27.

Polachek, S. W. 2006. How the life-cycle human-capital model explains why the gender wage gap narrowed. The declining significance of gender. IZA Discussion Paper, 102-124.

Schippers, J. J. 1987. Beloningsverschillen tussen mannen en vrouwen – een economische analyse, Thesis Utrecht

University, Wolters-Noordhoff, Groningen.

Schober, P. S. 2013. The parenthood effect on gender inequality: Explaining the change in paid and domestic work when British couples become parents. European Sociological Review, 29(1): 74-85.

Sheffrin, S. M. 2003. Economics: Principles in action. Upper Saddle River, New Jersey, 7458, 551.

(28)

Van der Meer, P. H. 2008. Is the gender wage gap declining in the Netherlands? Applied Economics, 40(2): 149-160.

Weichselbaumer, D., & Winter‐Ebmer, R. 2005. A Meta ‐Analysis of the International Gender Wage Gap. Journal

(29)

APPENDIX A

SPSS Syntax 2006

* Encoding: UTF-8. fre ca001_06. select if (ca001_06 eq 1). fre ca001_06. fre aa004_06. select if (aa004_06 gt 19). fre aa004_06. fre db051_06

/stat /format=notable /histogram. fre db004_06.

DELETE VARIABLES inkomen. compute inkomen=0. if (db004_06 = 1) inkomen=db051_06. if (db004_06 = 2) inkomen=db051_06/4. if (db004_06 = 3) inkomen=(db051_06*12)/52. if (db004_06 = 4) inkomen=db051_06/52. fre inkomen

/stat /format=notable /histogram. missing values inkomen (0). fre inkomen

/stat /format=notable /histogram. fre ea104_06 ea105_06 ea106_06 /stat.

compute uurloon=inkomen/ea104_06. fre uurloon

/stat /format=notable /histo

if (not missing (ea106_06)) uurloon=inkomen/ea106_06. fre uurloon

/stat /format=notable /histo. select if (uurloon ge 6).

compute lnuurloon=ln(uurloon). fre lnuurloon uurloon

/stat /format=notable /histo.

(30)

compute gender=aa001_06-1. compute leeftijd=aa004_06/10. compute lfkw=leeftijd*leeftijd. Compute Education=ba016_06. compute LogHwork =LN(ea104_06). compute Marketexp=cc001_06/10. Compute SqmarketEXp=Marketexp*Marketexp. compute PermC=0. if (eb002_06 eq 1) PermC=1. compute marstP=0. if ( aa006_06 LE 2) marstP=1. compute ChildrenH=0. if (ga026_06 eq 1) ChildrenH=1. compute supervisor=0. if (ec003c_06 GE 2) supervisor=1. Compute Joblevel=ec023_06. compute lnfirmsize=LN(ee004_06). Compute Upotime=0. if (ea010_06 GT 0) Upotime=1. FRE UPOTIME.

fre ca039_06 ca003_06.

CROSSTABS ca039_06 by ca003_06. compute verandering=0.

if (ca039_06 eq 1 and ca003_06 le 3) verandering=1 . if (ca039_06 eq 1 and ca003_06 gt 3) verandering=2 . fre verandering /stat.

compute Chcjob =0.

if (verandering eq 1) Chcjob=1. compute Chempl=0.

if (verandering eq 2) Chempl=1. fre Chempl Chcjob

compute Parttime=0.

(31)

compute PartP =0.

if (ea050_06 EQ 2 and ea051_06 LE 2) partP=1. compute Thwerken=0.

if (ed220_06 EQ 1 and ed221_06 LE 2) Thwerken=1. compute ThwerkKid=0.

if (ed220_06 EQ 1 and ed227_06 EQ 4) Thwerkkid=1. COMPUTE CarPers=ed203_06 + ed215_06 + eh101_06 . EXECUTE.

*Reliability loopbaanperspectieve RELIABILITY

/VARIABLES=eh101_06 ed203_06 ed215_06 /SCALE('ALL VARIABLES') ALL

/MODEL=ALPHA /STATISTICS=CORR /SUMMARY=TOTAL. SORT CASES BY aa001_06.

SPLIT FILE LAYERED BY aa001_06. FREQUENCIES VARIABLES=Chempl

/STATISTICS=STDDEV MINIMUM MAXIMUM MEAN MEDIAN MODE SKEWNESS SESKEW

/ORDER=ANALYSIS. REGRESSION

/DESCRIPTIVES MEAN STDDEV CORR SIG N /MISSING LISTWISE

/STATISTICS COEFF OUTS R ANOVA /CRITERIA=PIN(.05) POUT(.10)

/NOORIGIN

/DEPENDENT lnuurloon

/METHOD=ENTER Education Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chcjob PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers.

*************************************************************************** ***********

REGRESSION

(32)

/DEPENDENT lnuurloon /METHOD=ENTER gender. T-TEST GROUPS=gender(0 1) /MISSING=ANALYSIS /VARIABLES=uurloon lnuurloon /CRITERIA=CI(.95). T-TEST GROUPS=gender(0 1) /MISSING=ANALYSIS

/VARIABLES=Education Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chcjob PermC ChildrenH marstP LogHwork leeftijd lfkw

Thwerken ThwerkKid CarPers /CRITERIA=CI(.95).

DATASET ACTIVATE DataSet1. USE ALL.

COMPUTE filter_$=(aa005_06 = 1).

VARIABLE LABELS filter_$ 'aa005_06 = 1 (FILTER)'. VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'. FORMATS filter_$ (f1.0).

FILTER BY filter_$. EXECUTE.

REGRESSION

/NOORIGIN

/METHOD=ENTER Education Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chcjob PermC marstP LogHwork leeftijd lfkw

Thwerken CarPers.

REGRESSION

(33)

/NOORIGIN

REGRESSION

/NOORIGIN

/METHOD=ENTER Gender Education Joblevel Marketexp SqmarketEXp Upotime

supervisor lnfirmsize Chempl Chcjob PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers.

REGRESSION

/NOORIGIN

(34)

VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'. FORMATS filter_$ (f1.0).

REGRESSION

/NOORIGIN

/METHOD=ENTER Education Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chcjob PermC marstP LogHwork leeftijd lfkw

Thwerken ThwerkKid CarPers. Compute Geedu= Gender*Education. Compute GEJobl= Gender*Joblevel. Compute Geexp=Gender*marketexp. Compute GeExpS = Gender*SqmarketEXp. Compute Geage= Gender*Leeftijd.

Compute GeAgesq= Gender*lfkw. Compute GEUpotime= Gender*Upotime. Compute GeSuper= Gender*Supervisor. Compute GELnfisize= Gender*Lnfirmsize. Compute GeChempl= Gender*Chempl. Compute GeChjob=Gender*Chcjob. Compute GpermC= Gender*PermC. Compute GeChH= Gender*ChildrenH. Compute GmarstP=Gender*marstP. Compute GeThw=Gender*Thwerken. Compute GeLhW= Gender*LogHwork. Compute GETTwKid= Gender*ThwerkKid. Compute GecarP=Gender*CarPers.

REGRESSION

/NOORIGIN

/METHOD=ENTER Gender Education Joblevel Marketexp SqmarketEXp Upotime

supervisor lnfirmsize Chempl Chcjob PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers Geedu GEJobl Geexp GeExpS Geage GeAgesq

(35)

T-TEST GROUPS=aa001_06(1 2) /MISSING=ANALYSIS

/VARIABLES=lnuurloon Education Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chcjob PermC ChildrenH marstP LogHwork leeftijd lfkw

(36)

APPENDIX B

SPSS Syntax 2012

* Encoding: UTF-8. fre ca001_12. select if (ca001_12 eq 1). fre ca001_12. fre aa004_12. select if (aa004_12 gt 19). fre aa004_12. fre db051_12

/stat /format=notable /histogram. fre db004a_12.

DELETE VARIABLES inkomen. compute inkomen=0. if (db004a_12 = 1) inkomen = db051_12. if (db004a_12 = 2) inkomen=db051_12. if (db004a_12 = 3) inkomen=db051_12/4. if (db004a_12 = 4) inkomen=(db051_12*12)/52. if (db004a_12 = 5) inkomen=db051_12/52. fre inkomen

/stat /format=notable /histogram. missing values inkomen (0). fre inkomen

/stat /format=notable /histogram. fre ea104_12 ea105_12 ea106_12 /stat.

if (db004a_12 = 1) uurloon=inkomen.

if (db004a_12 >= 2) uurloon=inkomen/ea104_12. fre uurloon

/stat /format=notable /histo

if (db004a_12 >= 2 AND not missing (ea106_12)) uurloon=inkomen/ea106_12. fre uurloon

/stat /format=notable /histo. select if (uurloon ge 6).

compute lnuurloon=ln(uurloon). fre lnuurloon uurloon

/stat /format=notable /histo.

fre aa001_12 aa004_12 eb002_12 ee003_12 ee021_12 ec035_12 ec023_12 ee004_12 ec003c_12

(37)

compute gender=aa001_12 - 1. compute leeftijd=aa004_12/10 . compute lfkw=leeftijd*leeftijd. compute Educ=ba016_12.

compute LogHwork =LN(ea104_12). compute Marketexp=cc001_12/10. Compute SqmarketEXp=Marketexp*Marketexp. compute PermC=0. if (eb002_12 eq 1) PermC=1. compute marstP=0. if ( aa006_12 LE 2) marstP=1. compute ChildrenH=0. if (ga026_12 eq 1) ChildrenH=1. compute supervisor=0. if (ec003c_12 GE 2) supervisor=1. Compute Joblevel=ec023_12. compute lnfirmsize=LN(ee004_12). Compute Upotime=0. if (ea010_12 GE 1) Upotime=1. FRE UPOTIME. compute Thwerken=0.

if (ed220_12 EQ 1 and ed221_12 LE 2) Thwerken=1. compute ThwerkKid=0.

if (ed220_12 EQ 1 and ed227_12 EQ 4) Thwerkkid=1. Compute Chjob=0.

If (ca039_12 EQ 1 and ca003_12 LE 3) Chjob=1. Compute Chempl=0.

if (ee025_12 EQ 2) Chempl=1.

(38)

*Reliability loopbaanperspectieven RELIABILITY

/VARIABLES=eh101_12 ed203_12 ed215_12 /SCALE('ALL VARIABLES') ALL

/MODEL=ALPHA /STATISTICS=CORR /SUMMARY=TOTAL. SORT CASES BY aa001_12.

SPLIT FILE LAYERED BY aa001_12. REGRESSION

/NOORIGIN

/METHOD=ENTER Educ Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chjob

PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers. fre Educ Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chjob PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers /stat

*************************************************************************** **************

REGRESSION

/STATISTICS COEFF OUTS R ANOVA /CRITERIA=PIN(.05) POUT(.10) /NOORIGIN /DEPENDENT lnuurloon /METHOD=ENTER gender. T-TEST GROUPS=gender(0 1) /MISSING=ANALYSIS /VARIABLES=uurloon lnuurloon /CRITERIA=CI(.95).

(39)

Compute GEJobl= Gender*Joblevel. Compute Geexp=Gender*marketexp. Compute GeExpS = Gender*SqmarketEXp. Compute Geage= Gender*Leeftijd.

Compute GeAgesq= Gender*lfkw. Compute GEUpotime= Gender*Upotime. Compute GeSuper= Gender*Supervisor. Compute GELnfisize= Gender*Lnfirmsize. Compute GeChempl= Gender*Chempl. Compute GeChjob=Gender*Chjob. Compute GpermC= Gender*PermC. Compute GeChH= Gender*ChildrenH. Compute GmarstP=Gender*marstP. Compute GeThw=Gender*Thwerken. Compute GeLhW= Gender*LogHwork. Compute GETTwKid= Gender*ThwerkKid. Compute GecarP=Gender*CarPers.

COMPUTE filter_$=(aa005_12 =1).

VARIABLE LABELS filter_$ 'aa005_12 =1 (FILTER)'. VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'. FORMATS filter_$ (f1.0).

REGRESSION

/NOORIGIN

PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken CarPers. DATASET ACTIVATE DataSet1.

USE ALL.

(40)

/MISSING LISTWISE

/NOORIGIN

PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers. DATASET ACTIVATE DataSet1.

USE ALL.

REGRESSION

/NOORIGIN

/METHOD=ENTER Gender Educ Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chjob

PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers. DATASET ACTIVATE DataSet1.

USE ALL.

REGRESSION

/NOORIGIN

/METHOD=ENTER Gender Educ Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chjob

(41)

USE ALL.

REGRESSION

/NOORIGIN

/METHOD=ENTER Gender Educ Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chjob PermC ChildrenH marstP LogHwork leeftijd lfkw

Thwerken ThwerkKid CarPers Geedu GEJobl Geexp GeExpS Geage GeAgesq

GEUpotime GeSuper GELnfisize GELnfisize GeChempl GeChjob GpermC GeChH GmarstP GeThw GeLhW GETTwKid GecarP.

DATASET ACTIVATE DataSet1. T-TEST GROUPS=aa001_12(1 2) /MISSING=ANALYSIS

/VARIABLES=lnuurloon Educ Joblevel Marketexp SqmarketEXp Upotime supervisor lnfirmsize Chempl Chjob

PermC ChildrenH marstP LogHwork leeftijd lfkw Thwerken ThwerkKid CarPers /CRITERIA=CI(.95).