The effect of gender norms on the
education gender gap in Europe
Master Thesis Economics
University of Groningen
S.W.M. Fakkert
1s2370247
Supervisor: Dr. P. Milionis Abstract:This thesis tests the hypothesized relationship between gender norms and the education gender gap. Using data from 28 EU countries at different levels of NUTS classification I find that gender norms play a significant role in determining the education gender gap for the age group from 25-64 years. However, gender norms are not the sole determinant in explaining the education gender gap, female employment rate too is a determinant of the gap. Additionally, there is no effect of gender norms on the education gender gap for the age group 30-34 years, female employment rate is mostly the sole determinant of the education gender gap for this age group.
Keywords: education gender gap, gender norms, educational attainment JEL-code: A14, I21, J16
I. Introduction
The difference in levels of education between men and women have changed tremendously from the 1950s to the present (Pekkarinen, 2008; Buchman & DiPrete, 2006; OECD, 2011). In most Western countries women have surpassed men in terms of educational attainment, the education gender gap is now more often in the advantage of women. Since the 1990s, worldwide more women than men are enrolled in higher education (Shofer & Meyer, 2005). There is a vast amount of literature that have studied this development of the education gender gap. However, there has been little research done into the role of gender norms in determining the size of the education gender gap.
Educational attainment plays an important role in achieving the full economic potential of women (OECD, 2013; Lewis & Lockhead, 2008). Achieving the full economic potential of women could possibly lead to equality between men and women. Equality between men and women ultimately comes down to having equal opportunities, rights and obligations in all aspects of life. If this could be achieved through female educational attainment it is important to have knowledge of the determinants of female education and the determinants of the difference of educational attainment between men and women.
as a sociological factor in determining the differences in educational attainment between men and women.
Kluckhohn & Strodbeck (1961) find that values an attitudes can be seen as the principle which guide, channel or direct behaviour. Thus, gender norms can in theory influence the choice of education. The effect of gender norms on labour market behaviour of women has been studied in the past and several studies find that women who have traditional gender norms are less likely to supply labour to the market (Alebrecht et al., 2000; Hakim, 2000; Stam et al., 2014). Additionally, Stam et al. (2014) find that for women in the Netherlands gender norms are a determinant for the number of hours worked by women. Women with more traditional beliefs regarding gender norms work substantially less hours than women with egalitarian beliefs regarding gender norms. Hence, one can conclude that gender norms do affect certain decisions and in theory could therefore influence education levels.
thus have the highest level of education at this point in time. And since the labour force is up to 64 years old, 64 years was selected as the upper bound. The group 30-34 years was selected to focus on one both cohort rather than on a group of different cohorts. With these two different group one can understand the importance of gender norms and test if this still hold once the emphasis lays on one birth cohort.
Do gender norms affect the educational gender gap? Based on the results from this thesis the answer is: “Yes, to some extent.” The reason for this answer is that for the age group 25-64 years gender norms are a significant determinant of the education gender gap, for all NUTS levels tested, while this is not the case for the age group 30-34 years. However, gender norms was not the sole determinant in explaining the size of the education gap for the age group 25-64 years, female employment rate too is a determinant in explaining the education gap for this age group. Meanwhile, for the age group 30-34 years female employment rate is the sole determinant in some regressions.
The remainder of the thesis will be organized as follows: section II summarizes both the past research into the education gender gap and the theory behind gender norms. In section III the data that is used to perform this research is discussed in detail. The results of the regressions will be discussed in Section IV. Test that were performed to test the robustness of the model will the summed up in section V. In the last section, section VI, results will be discussed and some concluding remarks are formed.
II. Literature review
between levels of education between men and women is referred to as the education gender gap. Educational attainment can be measured in two ways, as a stock measure or as a flow measure (Pekkarinen, 2012). A stock measurement reflects the human capital at one specific point in time within an economy. The flow measurement focusses on the contributions of new cohorts to the total stock of human capital. Thus, the two are rather different. Changes will be more easily reflected when using the flow measurement than the stock measurement. In order to see a change in the stock measurement there will need to be a lot of new human capital accumulated in order to see a difference. Meanwhile, the flow measurement will be different for each cohort.
of 144 countries. Pekkarinen (2012) uses the Barro and Lee (2010) data to study birth cohorts in the Nordic countries2 and the US. Pekkarinen (2012) finds that in all of these
countries, except for Denmark and the US, women born in the cohorts in the 1950’s already caught up with men in terms of educational attainment. The catching up of women to men in terms of education did, in some cases, not stop at parity. Since the mid-1990s the gender gap has mainly been in the advantage of women (Vincent-Lancrin, 2008). However, variations between and within countries remain present (Hek et al., 2016; Baker and Jones, 1993; McDaniel. 2010).
Reasons for this rapid increase in the educational of attainment are based on different factors: economic, demographic, educational and sociological (Pekkarinen, 2012; Vincent-Lancrin, 2008). In discussing the economic factors that contributed to the rapid increase of female educational attainment education is seen as an investment. According to rational choice, an individual will balance the benefits and the costs when making decisions regarding education (Breen & Goldthorpe, 1997; Van de Werfhorst and Hofstede, 2007; Becker, 2009). The cost include the direct monetary costs and the forgone earnings when one would have been active on the labour market, and the effort used for attaining the education (Pekkarinen, 2008). The benefits of education is the earnings one will receive for a certain level of education, also education should increase productivity and thus increase the earnings of an individual. Acemoglu and Autor (2010) and Van Reenen (2011) found that the returns to education have been increasing in most of the OECD countries since the early 1980s. Increasing returns from education could be a motivation for individuals to get a certain level of educational attainment. However, this does not explain the differences in education between men and women, because if returns are increasing why would men not too
increase their educational attainment? This could be explained based on the thought that the expected return of education has developed differently for men and women. The literature that examines this aspect is somewhat divided. Trostel et al. (2002) for the US and Dougherty (2005) for European countries, find a rather consistent pattern that the average return to an additional year is slightly higher for women than for men. Nevertheless, it seems that the returns to university level of education have increased similarly for both men and women since the early 1980s (Pekkarinen, 2012).
Demographic factors facilitated women to have greater participation in
education and decreased drop-out rates of women. A decisive factor in demographic factors that allowed women to have more opportunities for education and explained the increase of female educational attainment is contraception. Research from the US find that since the introduction of oral contraception in 1960 participation of women in higher indication increased (Goldin and Katz, 2002). Women were now able to delay the age at which they had their first child and thus had more time to start and complete education. However, it is important here that one bears in mind the cultural factor, in some cultures this effect will be different, because of the norms and values within a culture. The fact that this factor is also influenced by culture could partly explain why there are variations between countries and or regions.
Educational factors also play a large role in the increase of female educational
Another important aspect are the school drop-out rates. Of the OECD member countries in 2008, only in Germany and Switzerland women more often left school without upper secondary education than men (OECD, 2011). As for the other OECD member countries men more often left school without upper secondary education than women. Thus, when zooming in on the group of school drop-outs one will find that the biggest part of this group will be male. The combination of a decreasing number of men receiving a upper secondary education and the increased presence of women in higher education partly explains changes in the education gender gap.
Sociological factors. These factors relate for example to the decline of
decimation in the labour market and the effects of living in a more egalitarian society (Vincent-Lancrin, 2008). Studies relate emancipatory contextual circumstance to be the reason for most of the cross-country and over-time variation in the education gender gap (Hek et al., 2016; Marks, 2008; Penner, 2008). One can link the emancipatory contextual circumstances to having certain beliefs of what kind of behaviour is seen as appropriate for women.
women too strive for a successful career instead of focussing on the family life. Do note here that in reality it is not as simple as it is described here, these views are two extremes and therefore not likely to be present in society in the form as described above. It is more likely that a society will be at some point in between these two extremes. Based on gender norms an individual will behave a certain way that is in line with the gender norms created by society. Therefore, values and attitudes can be seen as principle which guide, channel or direct behaviour (Kluckhohn & Strodbeck, 1961).
Gender norms are thus a cultural attribute that characterize groups of people. Gender norms are passed on by parents to their children (Witt, 1997), by peer groups (Coleman, 1961; Akerlof and Kraton, 2002) and by social institutions (Young men Brazil). Once norms are formed they are rather stable and resistant to change (Bolzendahl & Meyers, 2004; Blundson & Reed, 2007). However, within the society there will always be certain individuals that have different ideas regarding the gender norms, these individuals can be called non-conformist. Barker (2001) explains this by stating that individuals ‘reconstruct’ the gender norms and put their own ‘subjective spin’ on the gender norms. Gender norms are a socio-cultural construct (they are learned), society-made, variable (changes over time), systemic (differs with society), hierarchical (binds the person to certain roles and responsibilities, difficult to change (but not impossible).
society plays a role in the decision-making process. This is implied by word such as ‘perceived as being appropriate’, ‘self identity’ which is affected by society, and ‘gender roles’. Eccles (1987) states here that based on gender roles men and women place different value on long-range goals and adult activities, meaning activities an individual will undertake when he or she is adult, such as being employed. This once again emphasizes that gender roles determine ‘appropriate’ behaviour determined by society and the people in and individual’s environment (Kluckhohn and Strodbeck, 1961).
Sharpe (1974) studies girls aged 14 to 15 in the 1970s from the working class in London using interviews and surveys The surveys and interviews were surrounding the topic of priorities regarding love, marriage and career. The reason to interview working class girls was, because they had rather different future perspectives than girls from higher income parents. The priorities the girls had can be summed up as follows:
‘1. Love 2. Marriage 3. Husbands 4. Children 5. Jobs 6. Careers’. With priorities like
1970s to the 1990s change has found place in the gender norms held by young girls and women, a possible explanation for this can be the role of feminism.
a large role in the decline of gender discrimination (Brooks and Bolzendahl, 2004; McHugh and Frieze, 1997) which provides young girls and adult women more opportunities in education, but also in society in general.
There are several social gains from educating women, promoting female education is known to reduce fertility levels, reduce child mortality, levels and also promote the education of the next generation (Klasen and Lamanna, 2009). Bloom and Williamson (1998) discusses the ‘demographic gift’ after reduced fertility rates for about twenty years. The so-called demographic effect occurs when for a period of several decades the working-age population will grow much faster than the overall population, thus lowering dependency rates with positive repercussions for per capita economic growth. As previously mentioned the role of contraception was important in the catching up of women educational wise (Goldin and Katz, 2002).
In the twentieth century, global life expectancy at birth increased by more than 30 years (Riley, 2005). If education is considered as an investment, increased life expectancy will give individuals a chance to receive the returns of the investment over a longer time-period.
Economically active women can be a role model for young girls, encouraging them to get educated and strive for successful careers (Hek et al., 2016), which in turn will increase labour market participation of women. Moreover, this increased labour market participation of women has been found to increase labour productivity (Knowles, 2002) and this can increase income.
dominance in educational attainment is becoming stronger in the coming decades (Pekkarinen, 2012). OECD (2011) estimates that in 2025 there will approximately be some 1,4 female students for every male student in higher education.
III. Data and Methodology
The data set employed in this study was constructed using data from Eurostat. The advantage of using data from Eurostat is that the data uses the NUTS (Nomenclature of Territorial Units of Statistics) classification. The NUTS system runs from a level 0 to a level 3. A NUTS 0 level corresponds to a country as a whole, while a NUTS 3 level corresponds to the finest division of regions. The distinction between NUTS regions is in particular useful for this study, as literature suggests that there are not only differences in the education gender gap and gender norms between countries, but also between regions within the same country (Yazilitas et al., 2013; Hek et al., 2016; Baker and Jones, 1993).
The following variables are included in the data set that is used to perform the analysis: the education gender gap 25-64 years, the education gender gap 30-34 years, gender norms, disposable income, female employment rate, total population, fertility rate and female life expectancy.
Dependent variable
test scores has the advantage that we can study men and women from different education cohorts. If one would use test score it would not be able to study several cohorts as scores are almost always collected with students from primary or secondary school (Hek et al., 2016). Additionally if one would choose to use enrolment rates into education, one would again not study a range of cohorts, but solely the cohort enrolled at that point in time. The education gender gap was constructed by deducting the average years of schooling of women (25-64 years) from the average year of schooling of men (25-64 years). In the same way the second gap, for the population from age 30 to age 34 was constructed.
of women. Meanwhile, for the age group of 25-64 years we clearly see the existence of an education gender gap, only for 2012 and 2013 the gap is in advantage of women. However, we do see the gap closing up from the period 2000-2011.
As previously mentioned there are still large variations between countries and within regions. Figure 2 shows the gap per country3, it is clear that there are large
differences between countries. For example, Belgium has only a small gender gap in the advantage of men, while in Finland there is a rather large gap compared to the rest of the EU countries, in the advantage of women. If we would zoom in on NUTS2 level regions Figure 5 and Figure 6 (see Appendix) show that there are indeed also variations within countries.
Gender norms
The variable that reflects gender norms was created using the 2008 wave of the European Values Study (EVS) in a paper written by Beugelsdijk, Klasing & Milionis (2017). The data from the EVS are obtained by conducting interviews with a
representative and random sample of the adult population. In order to create the gender norms variable the focus was on a subset of all the questions in the EVS. Based on responses that the individuals gave during their interviews they were assigned a certain degree of gender norms. The gender norms variable ranges from 0 to 1, getting closer to 1 indicates having egalitarian views. Approaching 0 corresponds with having rather traditional views. The Figure above shows the values for the gender norms on a country-level. An important factor in using this variable for gender norms is that it is based only on the EVS of 2007 and is then used as being a constant variable for the time period 2000-2013. The reason that this should not be seen as an issue is, because once gender norms are formed they are stable and resistant to change (Bolzendahl and Meyers, 2004; Blundson and Reed, 2007).
when comparing to the other EU countries. Another thing that can be concluded is that in Northern and Western Europe4 the variable of gender norms is, on average, higher
than for Eastern and Southern Europe5 indicating that Northern and Western Europe
hold more egalitarian beliefs towards gender norms than Eastern and Southern Europe. One could possibly link the lower gender norms for the Eastern European countries to the history of the presence of communism (Eurydice, 2010). Thus, gender norms differ between countries, but when we focus on NUTS2 level regions, it can also be concluded that gender norms differ within countries (see Appendix Figure 7 and Figure 8).
Based on the literature reviewed in section II, the hypothesis is that holding more egalitarian views will lead to a smaller education gender gap, or perhaps even a gap in the advantage of women. While holding more traditional views will lead to a larger education gender gap, in the advantage of men.
Control variables
Disposable income - The disposable income per household is measured in
euro’s and is the net disposable income. Education is often expensive, therefore income could influence the decision to enter into education or to continue with education. Especially if we think of a family in which parents have to make the decision to finance their children education. A higher level of income increases the possibilities of getting education. Within the EU there are large variations between countries in the amount of the disposable income. The country with the highest disposable income is
4 Western: Austria, Belgium, France, Germany, Ireland, Luxemburg, Netherlands, United Kingdom.
Nothern: Denmark, Finland and Sweden
5 Eastern: Estonia, Latvia, Lithunia, Bulgaria, Czech Republic, Croatia, Hungary, Poland, Romenia,
Austria with an average over the time period 2000-2013 of €19.002 and Bulgaria has the lowest average disposable income at an amount of only €2.188. In the regressions performed the natural logarithm of disposable income is introduced.
Female employment rate - The female employment rate is the percentage of
Total population - This variable is introduced to control for differences in
population between regions and or countries, to see if there would possibly be an effect of being highly populated or not. In the regressions performed the natural logarithm of total population is introduced.
Fertility rate - The fertility here is measured as the mean number of children
born to a women during her lifetime. As discussed a social gain of increased female educational attainment is the decreased fertility rate (Klasen and Lamanna, 2009). Additionally, Goldin and Katz (2002) find that the introduction of oral contraception, which decreased fertility rates, attributed to the increased participation in education by women. Thus, lower fertility rates offer women more chances to enter or continue education, so that the average educational level of women will increase and this will have a negative effect on the education gender gap.
Female life expectancy at birth – This indicates the expectation of the maximum
age a women will reach in her lifetime from the point of birth. Having a higher expectation of the duration of life will make the investment in education a more attractive investment, as the returns will be received during a longer period.
Methodology
The following equation is used in the analysis to study the effect of gender norms on the education gender gap:
Gap it = β0 + β1 Gender Norms i + β2 ln (disposable income) it + β3 Female
employment rate it + β4 ln (total population) it + β5 Fertility rate it
+ β6 Female life expectancy at birth it + γ0 i.Countryi + γ1 i.Yeari + ε
The equation is performed three times: with data of the entire sample, using only NUTS1 regions and using only NUTS2 regions. Since differences in the education gender gap are not only between countries, but also between regions, which provides reasoning to perform this analysis at different NUTS levels.
an issue in this sample and the variables that were selected based on theory can all be included.
When working with panel data one should always take into consideration that it is possible that different errors terms do not have the same variance, i.e. there is heteroskedasticity. Using the standard OLS to evaluate the existence of heteroskedasticity with the help of a Breach-Pagan test it was found that heteroskedasticity forms a problem in this general data set6. When performing the
analysis the fixed effect model included robust errors clustered at the country-level and thereby overcoming the issue of heteroskedasticity.
In order to justify the use of fixed effects for this regression a Hausman test was conducted. The results of the Hausman test had a p-value of 1, meaning that both random effects and fixed effects would be suited to run a regression. For this reason the use of a fixed effect model is justified. However, the most important reason for using a fixed effects model here is that one can control for cross-country differences, by adding country-dummies.
IV. Results
To assess the hypothesized relationship between gender norms and the education gender gap, the education gender gap for the ages 25-64 years and the education gender gap for the ages 30-34 years were both separately regressed on the gender norms, in three separate regressions. In each regression the following variables were included as control variables: disposable income, female employment rate, female life expectancy, fertility rate and population. In order to take into account effects of
unobserved country-level developments, the regression is one with country fixed effects and the standard errors are clustered at the country-level and country dummies. As the EVS was published in 2008, and the interviews were conducted in 2007, we will focus only on the year 2007 to get to our baseline regression results.
Results based on the education gender gap ages 25-64 years
Table 3 reports the results for the regression for the education gender gap ages 25-64 years. Based on these results it becomes apparent that gender norms play a role in the education gender gap. In column 1, 2 and 3 the sign of the coefficient is negative, meaning that moving from one side of the spectrum (the traditional side) to the other side (the egalitarian side) will decrease the education gender gap in societies.
Results based on the education gender gap ages 30-34 years
gender gap. Again, the effect is significant, however the effect is small. As for the population this also has a positive effect on the gap, if population increases by one percent the gap is expected to increase by 0,001767.
A possible explanation for these results to be deviant from the results found for the education gender gap of the age group 25-64 years is that the gaps are completely different. For the age group 25-64 years the gap in 2007 was 0,1521, still positive, thus men had high average years of schooling than women did. Contrary, for the age group 30-34 years the education gender gap was -0,3544, so women have a higher average years of schooling than men.
V. Robustness
In this section the results that are used to test for the robustness of the model are based on the time period 2000-2013 (results in Table 5, column 1,3 and 5) rather than just period 2007, in which the gender data was generated. Theory suggests that norms are not subject to a lot of change over time (Bolzendahl and Meyers, 2004; Blundson and Reed, 2007), thus keeping it constant should not be an issue. Since we now use a time period of 13 years, year dummies are introduced to correct for the influence of aggregate trends in the time-series.
The first way the robustness of de model is test is by changing the dependent variable from a gap to a ratio8. In principle the results should be similar. However, if in
the measurement of the gap a coefficient is negative (positive), it should be positive
7Note here that this effect is close to the effect found for NUTS 1 level for the gap of the age group 25-64 years.
However, for the age group 25-64 years the variable was significant at a p<0,001 level and for the age group 30-34 years it is significant at a p<0,1 level.
(negative). Looking at the table for the age group of 25-64 years one can conclude that the results are mostly consistent when moving from the education gender gap as a real gap to the education gender gap as a ratio. The same test was performed for the age group 30-34 years and the same conclusion can be drawn, that the results are mostly consistent. Thus, the results do not change significantly when using a different manner of measuring the dependent variable.
Control variables
by total employment rate and female life expectancy by total life expectancy? The answer is not much9 (see Appendix Table 7). The variable for gender norms remains
to be significant. However, it is worth to note that the variables total employment and total life expectancy are not significant. This finding actually makes sense based on past research that stated that female employment rate attribute to female education levels (Hek et al., 2016) thereby having an effect on the gap.
VI. Discussion and Conclusion
Over the past 50 years, women have been catching up to men in terms of educational attainment and in some cases even surpassed men (Shofer & Meyer, 2005; Pekkarinen, 2012; Vincent-Lancrin, 2008; Hek et al. 2016). Literature provides several reasons for this, such as demographic factors, economic factors, sociological and educational factors (Vincent-Lancrin, 2008; Pekkarinen, 2012). Some studies allude to the fact that sociological factors do partly determine the size of the education gender gap (Hek et al., 2016; Nosek et al., 2009; Charles & Bradley 2012), which has been the main motivation for this thesis to focus on the influence of gender norms on the education gender gap.
The results show that gender norms do play a role in explaining the size of the education gender gap for the group of 25-64 years, indicating that moving from a traditional to a more egalitarian view will decrease the size of the educational gender gap. However, gender norms is not a significant determinant in explaining the educational gender gap for the age group 30-34 years. Other significant variables for
the age group 25-64 years are the female employment rate, which can be linked to the literature that states that women that are economically effective will encourage girls to strive for successful careers, thereby increasing the importance of education and eventually decrease the educational gender gap. Female life expectancy at birth was also found to be important, which is logical if one would consider education as an investment with returns over a certain time period. Higher life expectancy increases the time period in which returns can be received. Although it is perhaps important to note that the female life expectancy is only significant at the NUTS1 level, while female employment rate is significant for the regression using the entire data set and for the regression using only NUTS2 regions.
Remarkably, the results for the age group 30-34 years are completely different than the results for the age group 25-64 years. For example, gender norms is no longer significant, but mostly only female employment rate remains significant. The most obvious explanation here is that the educational gender gap for the two groups is completely different (see Figure 1). For the age group 30-34 years the educational gender gap is on average in the advantage of women and is also widening in the advantage of women.
Nevertheless, there are still men in education that are now starting to lack behind. It perhaps seems a bit odd that men are missing out on education that will benefit them, as there are high returns to education (Acemoglu & Autor, 2010). The future will tell if the decreased level of education of men respectively to the levels of education of women will become a reason for concern.
responses of the individuals that are interviews are taken at one point in time, rather than recorded frequently over a certain time period. However, several studies find that gender norms are, once they are formed, stable and resistant to change (Bolzendahl and Meyers, 2004; Blundson and Reed, 2007). So according to literature this would not form a threat in biasing the results if one focusses on the entire period from 2000 to 2013. Notwithstanding that if somehow these norms did change, this study does not take this into consideration.
An additional limitation is that there is no information regarding the changes that are made in the time period studied per country. It is possible that in the time period studied (2000-2013) there were changes made in the educational system or in government policy. Do bear in mind that we do control for these differences, but cannot analyse them fully. The results do not tell us which system, in which country or region is the most efficient in decreasing the educational gender gap, which can be seen as a limitation of this thesis.
As panel data is used in the analysis of the research one needs to evaluate the threat of endogeneity. Endogeneity in a model can arise in different ways omitted variables, measurement error and simultaneity and reverse causality. Omitted variables are often seen as the biggest threat to a model. Having omitted variables can seriously bias the results of a regression. However, in this model the threat is not a reason for concern. Firstly, because of the use of a fixed model, time invariant omitted variables are ‘caught’ by the use of this fixed effects model and the use of year and country dummies. Secondly, based on theory all relevant variables are included.
year 2007 they are on average rather low. When we move to the results for the entire time period from 2000 to 2013 the R-square do increase, but still remain below 0,5. However, we do find several significant variables in the regressions. Therefore, the low R-squares are not a reason for concern.
References
Acemoglu, D., & Autor, D. (2010). Skills, tasks, and technologies: Implications for employment and earnings. Handbook of Labor Economics, volume 4B, 1043-1171.
Akerlof, G.A., & Kranton, R.E. (2002). Idenitty and Schooling: Some Lessons for the Economics of Education. Journal of Economic Literature, 40(4), 1167-1201.
Albrecht, J. W., Edin, P., & Vroman, S. B. (2000). A cross-country comparison of attitudes towards mothers working and their actual labour market experience. Labour, 14(4), 591–608.
Badgett, L.M.V., & Folbre, N. (1999). Assigning care: Gender norms and economic outcomes.
International Labouw Review, 138 (3), 311-326.
Baker, D. P., & Jones, D. P. (1993). Creating gender equity: Cross-national gender stratification and mathematical performance. Sociology of Education, 66, 91–103.
Barker, G. (2001). Peace boys in a war zone: Identity and coping among adolescent men in a favela in Rio de Janeiro, Brazil. Unpublished doctoral dissertation, Erikson Institute, Loyola University, Chicago. Barro, R.J., & Lee, J-W. (2010). A new data set of educational attainment in the world, 1950 – 2010.
Journal of Development Economics, v104, 184-198.
Becker, G.S. (2009). Human capital: A theoretical and empirical analysis with special reference to
education (3rd ed.). Chicago, IL: The University of Chicago Press.
Beugelsdijk, S., Klasing, M., & Milionis, P. (2017). Value and diversity and regional economic development. The Scandinavian Journal of Economics.
Beugelsdijk, S., Klasing, M., & Milionis, P. (2017). Regional economic development in Europe: the role of total factor productivity. Regional Studies.
Bloom, D.E., & Williamson, J.G. (1998). Demographic Transition and Economic Miracles in Emerging Asia. World Bank Economic Review, 12(3), 419-455.
Blunsdon, B. and Reed, K. (2007). Changes in attitudes to mothers working: evidence from Australian surveys. Labour & Industry, 16, 2, 1030-1763.
Bolzendahl, C. and Meyers, D.J. (2004). Feminist Attitudes and Support for Gender Equality: Opinion Change in Women and Men, 1974-1998. Social Forces, 83, 2, 759-790.
Breen, R., & Goldthorpe, J.H. (1997). Explaining educational differentials: Towards a formal rational action theory. Rationality and Society, 9, 275-305.
Brooks, C., & Bolzendahl, C. (2004). The transformation of US gender role attitudes: Cohort replacement, social-structural change, and ideological learning. Social Science Research, 33(1), 106 133.
Buchmann, C., & DiPrete, T. A. (2006). The growing female advantage in college completion: The role of family background and academic achievement. American Sociological Review, 71, 515–541. Charles, M., & Bradley, K. (2002). Equal but spate? A cross-national study of sex segregation in higher education. American Sociological Review, 67, 573-599.
Coleman, J.S. (1990). Foundations of social theory. Cambridge, MA: Harvard University Press.
Dougherty, C. (2005), Why are returns to schooling higher for women than for men? Journal of Human
Resources, 40 (4), 969-88.
Eccles, J.S. (1987). Gender Roles and Women’s Achievement-Related Decisions. Psychology of
Eurydice Network. (2010). Gender Differences in Educational Outcome: Study on the Measures Taken and the Current Situation in Europe. Education, Audiovisual and Culture Executive Agency.
Fraser, N,. (2012). Feminism, Capitalism, and the Cunning of History. Paris: Fondation Maison des sciences de l’omme.
Freedman, E. B. (2002). No turning back: The history of feminism and the future of women. New York: Ballantine Books.
Goldin, C. (2004). The Long Road to the Fast Track: Career and Family. The Annals of the American
Academy of Political and Social Science, 596, 20-35.
Goldin, C., & Katz, L. F. (2002).The power of the pill: Oral contraceptives and women’s career and marriage decisions. Journal of Political Economy, 110 (4), 730-70.
Hakim, C. (2000). Work-lifestyle choices in the 21st century: Preference Theory. Oxford: Oxford University Press.
Hamilton, P., & Spongberg, M. (2017). Twente Years On: feminist histories and digital media. Women’s
History review, 26:5, 671-677.
Klasen, S., & Lamanna, F. (2009). The impact of Gender Inequality in Education and Employment on Economic Growth: New Evidence for a Panel of Countries. Feminist Economics, 15(3), 91-132. Klasing, M., & Milionis, P. (2017). The International Epidemiological Transition and the Education Gender Gap. Beiträge zur Jahrestagung des Vereins für Socialpolitik.
Kluckhohn, F. R., & Strodtbeck, F. L. (1961). Variations in value orientations. Oxford: Row, Peterson. Knowles, S., Lorgelly, P.K., & Owen, P.D. (2002). Are educational gender gaps a break on economic development? Some cross-country empirical evidence. Oxford Economic Papers, 54, 118-149.
Lewis, M., & Lockheed, M. (2008). Social Exclusion and the Gender Gap in Education. Washington, D.C.: The World Bank.
Marks, G.N. (2008). Accounting for the gender gaps in student performance in reading and mathematics: Evidence from 31 countries. Oxford Review of Education, 34, 89-109.
McDaniel, A. (2010). Cross-national gender gaps in educational expectations: The influence of national-level gender ideology and educational systems. Comparative Education Review, 54, 27–50.
McHugh, M. C., & Frieze, I. H. (1997). The measurement of gender-role attitudes: A review and commentary. Psychology of Women Quarterly, 21(1), 1–16.
Nosek, B. A., Smyth, F.L., Sriram, N., Lindner, N.M., Devos, T., Ayala, A.,…. Greenwald, A.G. (2009). National differences in gender-science stereotypes predict national sex differences in science and math achievement. Proceedings of the National Academy of Sciences, 106, 10593-10597.
OECD. (2004). Learning for tomorrow’s world: first results from PISA 2003. OECD Publishing OECD. (2011). Education at a Glance 2011: OECD Indicators, OECD Publishing.
OECD. (2012). Education at a Glance 2012: OECD Indicators, OECD Publishing. OECD. (2013). Trends Shaping Education. OECD Publishing.
Pekkarinen, T. (2012). Gender differences in Education. Nordic Economic Policy Review, 1, 165-197. Pekkarinen, T., (2008), Gender differences in educational attainment: Evidence on the role of tracking from a Finnish quasi-experiment. Scandinavian Journal of Economics, 110 (4), 807-825.
Riley, JC. (2005). Estimates of regional and global life expectancy, 1800-2001. Population and Development Reviews, 31(3), 537-543.
Sharpe, S. (1976) Just Like a Girl: How Girls Learn to be Women. Harmondsworth: Penguin.
Sharpe, S. (1994). Just like a Girl: How Girls Learn to be Women: from the Seventies to the Nineties. Harmondsworth: Penguin.
Shofer, E., & Meyer, J.W. (2005). The Worldwide Expansion of Higher Education in the Twentieth Century. American Sociological Review, 70(6), 898-920.
Stam, K., Verbakel, E., de Graaf, P.M. (2014) Do values matter? The impact of work ethic and traditional gender role values on female labour market supply. Social Indicators Research: An International and
Interdisciplinary Journal for Quality-of-Life Measurement, v116 n2: 593-610.
Trostel, P., Walker, I., & Woolley, P. (2002). Estimates of the economic return to schooling for 28 countries, Labour Economics, 9 (1), 1-16.
Van de Werfhorst, H.G., & Hofdstede, S. (2007). Cultural capital or risk aversion? Two mechanisms for educational inequality compared. The British Journal of Sociology, 59, 391-415.
Van der Lippe, T., & Van Dijk, L. (2002). Comparative research on women’s employment. Annual review
of sociology, 28, 221–241.
Van Hek, M., Kraaykamp, G., & Wolbers, M.H.J. (2016). Comparing the gender gap in educational attainment: the impact of emancipatory contexts in 33 cohorts across 33 countries. Educational
Research and Evaluation, 22(5-6), 260-282.
Van Reenen, J. (2011). Wage inequality, technology and trade: 21st century evidence. Labour
Economics, 18 (6), 730-741.
Vincent-Lancrin, S. (2008). The Reversal of Gender Inequalities in Higher Education: An On-going Trend. Higher Education to 2030, 1, 265-298. Demography, OECD Publishing, Paris.
Walters, M. (2005). Feminism: a very short introduction. New York, Oxford: Oxford University Press. WHO (2007). What do we mean by "sex" and "gender". http://www.who.int/gender/whatisgender/ Wilkinson, H. (1994). No turning back: Generations and the genderquake. London: DEMOS.
Witt, S.D. (1997). Parental influence on children’s socialization to gender roles. Adolescence, Summer, 32(126), 253-9.
