Does shadow education contribute to inequality?

ABSTRACT

This study examines the role of shadow education as a contributor to educational inequality in South Korea. Using data from the Korean Educational Longitudinal Study (KELS), this paper analyzes the effects of private tutoring at 10^th and 12^th grade level on College Scholastic Aptitude Test (CSAT) scores. We reduce biases in prior research by using multi-time measures of private tutoring and instrumental variables approaches. Our findings show that shadow education increases the CSAT scores. We also find that shadow education significantly contributes to educational inequality even after controlling for various socio-economic characteristics.

KEYWORDS:

• Educational inequality
• private tutoring
• academic achievement

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. In the literature, the term “private tutoring” has been widely used instead of “shadow education”. We use “private tutoring” and “shadow education” interchangeably throughout this article.

2. As of 2021, 1,000 KRW is approximately equivalent to 0.83 USD.

3. KEDI has produced a sample weight but has decided not to release the weight to the public because the use of the weight does not make any significant difference for most analyses.

4. Naturally, the number of students taking the CSAT is smaller than the number of high school graduates. Some graduates may not have taken the CSAT because they planned not to go to college or had already been admitted, through early admission plans, to a college without the need for a particular CSAT score.

5. Furthermore, the KELS could not collect information regarding the CSAT scores from a small number of respondents because it failed to get their permission by mistake. To the extent that this administrative mistake arose randomly across respondents, however, it does not cause a bias.

6. We excluded watching study programs on public education broadcasts and attending after-school programs in a formal school setting, which were measures introduced by the government to reduce private tutoring.

7. Our instrumental variables estimation may suffer from an upward bias to the extent that students’ academic ability and parental support, affecting private tutoring and the CSAT scores, do not differ between subjects. As discussed in the text, we do not believe that the bias is serious. Given that our IV estimates may still suffer from the endogeneity bias, however, they may be accepted as an upper bound of $\beta$.

Additional information

Notes on contributors

Inhoe Ku

Inhoe Ku is a professor in the Department of Social Welfare, Seoul National University. He got his Ph. D from University of Washington. His research interests include poverty, inequality, and social policy.

Hyerim Lee

Hyerim Lee is currently a PhD student in the Department of Social Welfare, Seoul National University and is currently working as a researcher at the Seoul Institute.

Jung-Eun Kim

Jung-Eun Kim is an expert adviser in the Division of Social Welfare Policy Coordination, Ministry of Health and Welfare. She got her Ph. D from Case Western Reserve University. Her research interests include child welfare, inequality and social policy.