Inequality of Opportunity Through Effort in Education: An Application of Machine-Learning to Japan’s PISA Data (91479)

Session Information: Education / Pedagogy
Session Chair: Rolando Magat, Jr.

Friday, 16 May 2025 15:05
Session: Session 4
Room: Live-Stream Room 3
Presentation Type: Live-Stream Presentation

All presentation times are UTC + 9 (Asia/Tokyo)
– click here to convert to your timezone

Policymakers commonly aim at equalizing opportunity. This is because many hold that inequalities due to different levels of effort are acceptable, while those due to uncontrollable circumstances (e.g., ethnicity, sex, parental socioeconomic status) are objectionable. However, studies reveal that the amount of effort people exert is partly conditioned by such circumstances – an important caveat often overlooked in meritocratic societies. Then, as Roemer argued, inequality of opportunity (IOp), which policymakers aim to correct, should be defined as inequality due to the effects of circumstances, including their indirect effects through effort. This paper estimates the extent of IOp in educational attainment, including, importantly, the portion of it that manifests through effort. To this end, I operationalize Roemer’s model of IOp by applying the following methods to the data from OECD’s Programme for International Student Assessment (PISA): (1) a tree-based machine-learning techniqu and (2) a recently developed measure of students’ effort-level, which quantifies their performance decline during the assessment. This study is the first to conduct a machine-learning estimation of IOp for education. The study targets Japan because its effort-focused meritocratic culture combined with the post-2000 educational reform promoting self-motivated learning raises reasonable suspicion that IOp may have increased significantly. I also obtain estimates of other selected OECD countries for comparison. A high level of IOp and a significant increase between 2006 and 2012 are found in Japan, compared to other countries and to another study using IOp measure that does not account for IOp through effort.

Authors:
Yohei Yoshizawa, King's College London, United Kingdom


About the Presenter(s)
Yohei Yoshizawa is currently a Ph.D. student at the Department of Political Economy of King's College London.

See this presentation on the full scheduleFriday Schedule



Conference Comments & Feedback

Place a comment using your LinkedIn profile

Comments

Share on activity feed

Powered by WP LinkPress

Share this Presentation

Posted by Clive Staples Lewis

Last updated: 2023-02-23 23:45:00