We have another new PhD! Mayumi Uno successfully defended her dissertation at the University of Minnesota. The project is entitled: “National Institutional Context and Educational Inequality: A Multilevel Analysis of Variation in Family SES Effects on Academic Achievement across OECD Countries.”
Mayumi pursues a question that is of great interest to me: How does institutional context and the organization of national education systems affect schooling? In particular, she focuses on educational inequality, operationalized as the slope linking family background (SES) and achievement — which captures the extent that schools reproduce existing inequality. The dissertation examines a wide range of national-level institutional variables that might influence inequality, including differentiation, sources of funding, private tutoring, labor market incentives, and even welfare state variables.
The dissertation has a ton of interesting findings. For one thing, strong state control over the educational system is associated with less SES-based inequality. If the state controls all the funding and curriculum, and if there are more instructional hours in the school year, family effects are reduced.
That makes a lot of sense. The simplest way to eliminate the influence of family SES would be to send all kids to state-run boarding schools year-round. Obviously, no country does that… but there are real differences in the length of the school year, centralized funding of education, etc, which appear to be very consequential for inequality.
The dissertation also has some surprising results. For instance, the prevalence of tutoring/shadow education in a country is associated with less SES-related inequality. Mayumi expected the opposite, and so did I. You’d think that family resources would translate into much better tutoring. But, inequalities are greater in societies where there isn’t much tutoring. Perhaps tutoring has the biggest payoff in societies where few people can afford it? Like any good dissertation, the project raises new questions to be studied…
The study is a multilevel analysis of the PISA dataset, looking at math achievement among 15-year olds. The quantitative analysis is superb. Among other things, Mayumi carefully parses out the within-school vs. between school components of the SES effects… which yield a whole other set of interesting findings. For instance, the prevalence of private schooling is associated with bigger school-level (contextual) SES effects but smaller within-school SES effects. It makes sense one I started thinking about it… but I definitely didn’t know that before. VERY interesting.
I co-chaired the dissertation with Teresa Swartz. Jeylan Mortimer and Ann Hironaka were also on the committee. Mayumi also worked with Karen Bradley at Western Washington, where she did her MA.
Congratulations, Mayumi! You did a terrific job!