Empirical evidence from developed countries supports the idea that parent-teacher interaction is highand improves student outcomes. The evidence from developing countries is, however, decidedly mixed.Using longitudinal data from nearly 6000 students and their 600 teachers in rural China, we show theprevalence of parent-teacher interaction is generally much lower than that of developed countries. Wealso show parent-teacher interaction, when it exists, can have positive effects on raising academicachievement and reducing learning anxiety. We demonstrate that the prevalence and effectiveness ofparent-teacher interaction in a developing country context varies considerably due to both demand-sideand supply-side factors.

Studies suggest that students’ prior performance can shape subsequent teacher evaluations, but the magnitude of reputational effects and their implications for educational inequality remain unclear. Existing scholarship presents two major perspectives that exist in tension: do teachers primarily use reputational information as a temporary signal that is subsequently updated in response to actual student performance? Or do teachers primarily use reputational information as a filter that biases perception of subsequent evidence, thus crystallizing student reputations and keeping previously poor-performing students stuck in place? In a field experiment, we recruited a random sample of 832 junior high school teachers from the second-most populous province of China to grade a sequence of four essays written by the same student, and we randomly assign both the academic reputation of the student and the quality of the essays produced. We find that (1) reputational information influences how teachers grade, (2) teachers rely on negative information more heavily than positive information, and (3) negative reputations are crystallized by a single behavioral confirmation. These results suggest that students can escape their prior reputations, but to do so, they must contradict them immediately, with a single confirmation sufficient to crystallize a negative reputation.

Although students in rural and migrant schools in China generally have not performed well, a share of each cohort has been able to thrive in school and to test into academic high school and college. To understand the origins of persistence, specifically, why some students learn more than do others, researchers have identified certain sources of the problem. Few studies, however, have paid attention to the role that low levels of cognitive development of students play in their academic performance. To address this gap, this study focuses on the role that cognition may play in terms of the academic achievement of rural students. We analyze data from more than 10,000 primary school students from private migrant schools in Beijing and Suzhou and from public rural schools in Henan and Anhui, using the Raven Standard Progressive Matrices test. Our results show high rates of developmental delay (about 33% of the students have Ravens scores that are less than one standard deviation lower than an international mean). Further, the rates of delay are large among all subgroups in the study, including rural children who attend migrant schools in cities and those who live in rural areas and attend rural public schools. The results also suggest that the cognition of students is highly correlated with their educational performance and, in fact, is by far the most important factor in their academic achievement.

Despite massive investments in teacher professional development (PD) programs in developing countries, there is little evidence on their effectiveness. We present results of a large-scale, randomized evaluation of a national PD program in China in which teachers were randomized to receive PD; PD plus follow-up; PD plus evaluation of the command of PD content; or no PD. Precise estimates indicate PD and associated interventions failed to improve teacher and student outcomes after one year. A detailed analysis of the causal chain shows teachers find PD content to be overly theoretical, and PD delivery too rote and passive, to be useful.

We assess and compare computer science skills among final-year computer science undergraduates (seniors) in four major economic and political powers that produce approximately half of the science, technology, engineering, and mathematics graduates in the world. We find that seniors in the United States substantially outperform seniors in China, India, and Russia by 0.76–0.88 SDs and score comparably with seniors in elite institutions in these countries. Seniors in elite institutions in the United States further outperform seniors in elite institutions in China, India, and Russia by ∼0.85 SDs. The skills advantage of the United States is not because it has a large proportion of high-scoring international students. Finally, males score consistently but only moderately higher (0.16–0.41 SDs) than females within all four countries.