According to Turkish government figures, in the 2006-07 school year the overall rate of enrolment in voluntary secondary education (ie, schooling above the eight years that’s compulsory) was 56.5 per cent. This breaks down into rates of 60.7 per cent for boys and 52.2 per cent for girls. What causes this gap between male and female enrolment rates? In order to answer this and related questions, in 2009 researchers at Galatasaray University conducted a statistical analysis of data from TurkStat’s 2003 Household Budget Survey. You can find their report here (pdf; Turkish).
Interpreting the results of this analysis requires a familiarity with statistics that I don’t possess, but fortunately the implications of the research have been teased out in a report from the Education Reform Initiative (ERG), entitled Equity in Education: Policy Analysis and Recommendations. In the table below, I have summarised some of the key points relating to the the differential impact that a range of factors have on enrolment rates in secondary education for boys and girls.
The table only includes those factors which had a statistically significant impact on enrolment. One interesting result of this is that regional considerations don’t appear. This is in sharp contrast to the equivalent results for primary education, where residence in southeast Anatolia significantly reduces the odds of girls’ participation. This suggests that various factors not controlled for by the researchers’ socio-economic variables are at play in relation to compulsory primary education but not to voluntary secondary education — the ERG points to language issues, traditional values and problems stemming from internal displacement.
|Table 1: Factors affecting participation in secondary education (15-17 year olds)|
|Effect on the odds of participation|
|1-year increase in father’s education level||+15%||+10%|
|1-year increase in mother’s education level||—||+8%|
|Mother is the only parent in the household||—||-69%|
|The number of siblings increases by one||—||-15%|
|Social security membership||+40%||+26%|
|Household’s main source of income is agriculture||-53%||-63%|
Source: ERG, Equity in Education, 2009. Note: The term ‘odds’ has a specific meaning within the context of the regression analysis on which these figures are based. From the ERG report: “Although there is a close relationship between odds and probability. They are separate concepts. Odds are calculated by dividing the probability of an occurence by the probability of its non-occurence.”
The first point to note in relation to these secondary education data is that girls’ participation rates are much more sensitive to changes in household incomes than are boys’ rates. According to the ERG, this is because lower-income families already choose to send their sons rather than daughters to secondary school. Rising income allows more of these families to send their daughters too.
On the other broadly money-related variables the gap between effects on boys and girls is less pronounced. Households that depend on agriculture for their income are, unsurprisingly, less likely than others to send their children to school once it’s no longer compulsory. Conversely, the degree of financial security that’s provided by membership in the state’s social security arrangements increases the likelihood of participation. In each of these circumstances girls are slightly disadvantaged relative to boys. That is to say, relative to boys they see less of the the educational advantages of social security but more of the disadvantages of the agricultural sector.
Where the gender gap really starts to widen is when we look at household structure and parents’ education levels. In a household where the mother is the only parent, the odds of a daughter participating in secondary education are cut by as much as 69 per cent. What makes this figure all the more astonishing is that there is no equivalent effect on boys’ participation rates in these households. How many households are we talking about? According to the OECD, 7.2 per cent of Turkish households were in this category in 2007.
The most likely explanation for the huge effect of single parenthood on secondary school participation is that girls are required to take on domestic tasks which the mother no longer has time for as the household’s main breadwinner. This would seem to be confirmed by the indicator relating to the number of siblings in the house. For every additional sibling that a girl has, the odds of her participation in secondary school drop by 15 per cent. Again, there is no equivalent drop for boys, suggesting that sibling care duties fall very disproportionately on daughters rather than sons.
The final two indicators in the table above relate to parental levels of education. Here it is clear that mothers’ education levels are an important part of the gap between the participation rates of boys and girls. For every additional year of education that a mother has, her daughter’s odds of secondary participation increase by 8 per cent. In absolute terms, this increase is less than the 10 per cent increase which results for the daugher for each year of additional education that her father has. But in relative terms the mother’s education is more significant: boys benefit disproportionately from their fathers’ additional education (their odds of participation increase by 15 per cent), whereas their participation isn’t correlated at all with their mothers’ education.