ABSTRACT

Data from sub-Saharan Africa show that many students leave school without the skills they need. To address this issue, it is important to understand what factors influence learning. According to the Heyneman and Loxley effect established in 1983, in low income countries school quality influences how much students learn more than those students’ backgrounds. Recent research suggests that this influential conclusion no longer holds, without discounting the possibility that this trend could still be observed in very-low income countries. The present work investigates this possibility by using the PASEC dataset, which includes ten countries in West- and Central Africa. Improving on Heyneman and Loxley’s methodology by using general dominance analysis, this article finds no support for the ‘Heyneman-Loxley Effect’. Both school resource and student background account for around half of the explained variance in learning. This suggests that both family- and school-related factors are important sources of inequality of opportunity in low-income contexts.

摘要

来自撒哈拉以南非洲地区的数据显示，许多学生在没有掌握所需技能的情况下离开了学校。要解决这个问题，重要的是要了解影响学习的因素。根据 1983 年建立的海尼曼和洛克斯利效应，在低收入国家，学校质量对学生学习程度的影响大于对学生背景的影响。最近的研究表明，这一有影响力的结论不再成立，而且不排除这种趋势在极低收入国家仍然存在的可能性。目前的工作通过使用 PASEC 数据集调查这种可能性，该数据集包括西非和中非的十个国家。通过使用一般优势分析改进海尼曼和洛克斯利的方法，本文发现不支持“海尼曼-洛克斯利效应”。学校资源和学生背景都占解释学习差异的一半左右。这表明，与家庭和学校相关的因素都是低收入环境中机会不平等的重要来源。