BREAD Working Paper, April. Willa Friedman, Michael Kremer, Edward Miguel, and Rebecca Thornton (2011)

This follow-up study of a randomized intervention that provided merit scholarships to rural Kenyan girls unearths some surprising findings about their long run social and political attitudes.  The program increased political knowledge but also identification with one’s ethnic group and led to a broad rejection of authority, both at home in terms of parental involvement in marriages and domestic violence and in politics.

In this paper, Friedman, Kremer, Miguel, and Thornton use a randomized scholarship competition for Kenyan female teens to explore the social and political consequences of improved education.

The Girls Scholarship Program was launched by the Dutch NGO ICS in 2001.  It provided a merit scholarship to girls in the top 15% of their sixth grade class.  Totalling approximately US$38, the scholarship covered school fees and other school expenses in the last two years of primary school.  The scholarship was randomized across 24 treatment schools and 35 control schools.

Previous studies indicate that the scholarship increased test scores for girls across the distribution of initial test scores.  (It also increased test scores for boys, perhaps because of peer effects and increased teacher effort.)

In this follow up evaluation, the authors examine the long term impact of the program as measured by outcomes for program girls 4-5 years later, now aged 17 to 21.

They find that the in the long run, the program has a positive but statistically insignificant impact on subsequent educational attainment.  Views on women’s rights and gender roles (already pro-equality) did not significantly change.  However, treatment led to much stronger condemnation of domestic violence, and increased control over the choice of a spouse by girls.  The program did not have a significant impact on fertility, contraception knowledge, marriage age, or spouse characteristics.

Surprisingly, and against standard modernization theory, they did not find any evidence that the scholarship helped achieved the social and political goals of the Kenyan education system, such as national over ethnic attachments, secularism, and democracy.  Most surprising of all, the program led to girls expressing greater willingness to accept the use of violence in politics. 

Girls voiced strong attachment to their ethnic group, with only 11% not considering ethnicity “very important.”  The scholarship reduced ethnic attachment by a statistically insignificant 3.3 percentage points.  The scholarship also did not significantly change trust in members of other ethnic groups or more support for secularism.  It did not significantly affect a range of measures of democratic attitudes.  The authors conclude that “Overall, we find no support for the hypothesis that education promotes ‘modern’ attitudes including weakening of ethnic attachments, secularization, or greater belief in democracy” (22).

However, the program did increase political knowledge, with program girls spending less time listening to the radio, more time reading newspapers, becoming more likely to have a favorite newspaper, and becoming more knowledgeable about political figures.  However, programs girls reported sharply reduced satisfaction with political authority.

Overall, the authors identify a peculiar phenomenon: “the increased knowledge and reduced satisfaction with authority generated by the program does not seem to translate into greater perceived individual political efficacy or more participation in democratic politics or community affairs.  Instead, there is increased acceptance of the use of violence in politics” (25).

This may have to do with perceptions of the role of the individual in politics.  The authors report that at the time, Kenyan politics “was characterized by byzantine backroom deal-making among ethnically-based political leaders…  While treatment made the young women in our sample less satisfied with the Kenyan political situation, it apparently did not lead to any illusions about their own ability to directly change the situation” (25).  Increased education did not lead program girls to participate more in politics or community affairs relative to control girls.

These findings play into the debate on the role of education in politics.  While some argue that education should reduce political violence, others contend that education heightens expectations, and if left unmet, acceptance of violence may be the result.

The authors reason that “It seems plausible that education increases acceptance of violence because, as documented earlier, it increases the political knowledge of respondents and makes them less satisfied with the status quo, while apparently not simultaneously increasing their faith in their ability to achieve change through democratic means.  Since they are no more committed to democratic values as a result of education and, if anything, more ethnically identified (the key dimension of social conflict in Kenya), it may not be surprising that education increases support for political violence here” (26).