Child poverty and academic skills at the national level: A longitudinal analysis of four waves of PISA

Abstract

Much research has examined school-related sources of cross-national variation in academic skills, but we know far less about the role that broader socioeconomic conditions play. In this study, we examine how nations’ child poverty rates associate with their overall scores on international assessments of academic skills over time. We first establish theoretically why and how child poverty might undermine overall academic skills at the national level. We then use pooled time-series data on 40 nations that participated in four recent waves of the Programme for International Student Assessment (PISA) to analyze the impact of changes in child poverty rates on changes in nations’ overall academic skills (N = 160 nation-years). Two-way fixed-effects models find support for two hypotheses and, to a lesser extent, a third: As child poverty increases within nations over time, (1) mean academic skills decrease, (2) the percentage of low-skilled students increases, and (3) the percentage of high-skilled students decreases (for math only). We conclude by discussing the theoretical and political implications of the findings.