The school matters: Hong Kong secondary schools’ grade-retention composition, students’ educational performance, and educational inequality

ABSTRACT

In the face of Hong Kong’s high grade-retention rates, this study aimed to investigate how Hong Kong secondary schools’ grade-retention composition is associated with student performance and socioeconomic inequality in student performance. As the research questions involved analysis at the school and student levels, this study employed hierarchical linear modelling to analyse the Programme for International Student Assessment (PISA) 2018 data. While grade retention was often suggested to have a negative impact on repeaters’ performance in studies using the same-age comparison strategy, this study found that a higher proportion of retained students at school was not associated with a reduction in students’ performance. However, greater socioeconomic inequality in student achievement was found in schools with higher retention rates. In addition to providing plausible explanations for these findings, this paper discusses the potential role of the government’s retention policies in these respects.

KEYWORDS:

• Grade retention
• school retention composition
• educational inequality
• educational performance
• PISA 2018

Acknowledgements

We would like to express our sincerest gratitude to the two anonymous reviewers and the journal editors for their insightful comments, constructive advice, and generous support. This research has been conducted as a part of a larger project on Hong Kong education under Professor Stephen Wing-Kai Chiu.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Compared to reading and scientific abilities that are more likely to be learned outside school, mathematics skills are primarily instructed at school and are more affected by grade levels (Pong, 2009). Thus, mathematics performance is considered as a more reliable outcome variable in assessing the associations between school retention composition and educational achievement/inequality. However, although we focused on presenting the results based on students’ mathematics performance, supplementary analyses were also conducted with students’ reading/science performance as dependent variables. The results, which are not presented here but are available upon request, were consistent with our conclusions based on students’ math performance as the dependent variable.

2 Unlike Hong and Raudenbush’s (2006) research that measured schools’ retention composition by a dichotomous variable (i.e., high-retention vs. low-retention schools), schools’ retention composition was measured more precisely in our study by a continuous variable (i.e., schools’ percentage of retained students). Therefore, we did not adopt propensity-score matching like Hong and Raudenbush to account for the covariates associated with the difference in schools’ percentage of retained students, because propensity-score matching addresses the heterogeneity associated with receiving a treatment or not (i.e., a dichotomous scenario).

3 In HK, government and government-aided schools offer free education to students and instruct the curriculum recommended by the government.

4 By comparison, DSS schools enjoy greater flexibility in setting tuition fees and designing their own curricula.

5 In total, over 13% of 15-year-old HK students had repeated a grade during primary and/or secondary school in our sample.

6 Given the centring techniques adopted in this study, the slope of school retention composition also indicated its association with school mean performance.

7 Since the PISA score is built as a 0–1000 scale (with normal distributions) to capture the variation in students’ test results, it does not have substantive meanings (OECD, 2019). To facilitate the interpretation of PISA scores in substantive terms, the PISA 2018 math scale was classified as six skill proficiency levels, each of which corresponded to a range of 62.3 score points (with the exception of the highest level). As for smaller score-point differences, a comparison with some recognised benchmarks was recommended (Bloom et al., 2008). One benchmark was the average score gained by students within 1 year, which was estimated to be around 40 score points across PISA countries/economies (OECD, 2019). With the aid of these benchmarks, we interpreted that a 2.75-point reduction in students’ math scores associated with a 1% increase in school retention rate might not have large substantive meanings. This interpretation was supported by the small Cohen’s f² (= 0.11) for school retention composition, which is one measure of effect sizes in multilevel models (Lorah, 2018).

Additional information

Funding

We would like to acknowledge the partial support of the Education University of Hong Kong Start-up Research Grant (RG42/17-18R).

Notes on contributors

Nan Xiang

Nan Xiang is currently an assistant professor of Public Management at the School of Humanities and Social Sciences, Harbin Institute of Technology (Shenzhen). She received her doctoral degree in Sociology from The Chinese University of Hong Kong (HK), her master’s degree in Comparative Social Policy from the University of Oxford (UK), and her bachelor’s degree in English from Fudan University (China).

Stephen Wing-Kai Chiu

Stephen Wing-Kai Chiu is currently Chair Professor of Sociology at the Department of Social Sciences, Co-Director of The Academy of Hong Kong Studies, and Associate Dean of the Faculty of Liberal Arts and Social Science, The Education University of Hong Kong (HK). He obtained his doctoral degree from Princeton University (US) and his bachelor’s and master’s degree from The University of Hong Kong (HK).