Abstract: Growing evidence suggests that teachers in developing countries often have weak or misaligned incentives for improving student outcomes. In response, policymakers and researchers have proposed performance pay as a way to improve student outcomes by tying concrete measures like achievement scores to teacher pay. While evidence from randomized experiments generally indicates that performance pay programs are effective at improving student achievement in developing countries, there has been considerable variation in how much these programs affect student achievement. The goals of this study are to: (1) examine the impacts of different teacher performance pay designs on student achievement, both for the average student and for students across the baseline achievement distribution; and (2) examine the mechanisms through which different teacher performance pay designs affect student achievement (for the average student and for students across the baseline achievement distribution). The sample includes a total of 8,892 students and their grade 6 mathematics teachers from 216 schools from 16 nationally-designated "poverty" counties in Yulin Prefecture (Shaanxi Province) and Tianshi Prefecture (Gansu Province) in rural, northwest China. To test the impacts of the different teacher performance pay designs, researchers designed a cluster-randomized controlled trial. In this trial, schools were randomly allocated to 4 different treatment arms: (1) control--no teacher incentive pay; (2) levels incentive--performance pay contract stipulating rewards based on student achievement levels on endline tests; (3) gains incentive--performance pay contract based on student achievement gains from baseline and endline tests; and (4) pay-for-percentile incentive--performance pay contract stipulating rewards based on student growth percentiles. Surveys were used to collect information from the students, teachers, and school administrators. Findings reveal that: (1) Only "pay-for-percentile" incentives had a positive, statistically significant impact on average student achievement; (2) Teacher incentives based on "levels" or "gains" were ineffective; (3) "Gains" incentives led teachers to only focus on certain types of students, which led to negligible learning (on average) across all students; and (4) Pay-for-percentile incentives led to score gains across all students (on average). The results of this study may have important implications for how Teacher Performance Pay Policy can be implemented in China and in other developing countries.

The education of poor and disadvantaged populations has been a long-standing challenge for education systems in both developed and developing countries. Drawing on data from two randomised controlled trials involving two cohorts of grade 3 students in poor rural minority schools in China’s Qinghai province, this paper explores the effects of computer-assisted learning (CAL) on student academic and non-academic outcomes for underserved student populations, and how interactions between the CAL programme and existing classroom resources affect the programme effectiveness. Results show that CAL could have significant beneficial effects on both student academic and non-academic outcomes. However, when the scope of the programme expanded to include a second subject (in this case, math – which was added on top of the Mandarin subject matter that was the focus of the first phase of the programme), some schools had to use regular school hours for CAL sessions. As a result, the phase II programme did not generate any (statistically) significant improvement over the first phase. 

Abstract: China's agricultural sector faces challenges because most farms are still small scale. China's policy is to encourage the consolidation of farms and promote farms that are larger in scale. A question that arises is: Are China's farms growing? The goal of the present paper is to determine whether large farms in China have emerged or if farms remain small. To meet this goal, we systematically document the trends in the operational sizes of China's farms and measure the determinants of changes in farm size. Using a nationally representative dataset, the study shows that in 2013 China's farming sector was still mostly characterized by small-scale farms. However, at the same time, there is an emerging class of middle-sized and larger-sized farms. Most large farms are being run by households but there is a set of large farms that are company/cooperative-run. Today, farmers on larger farms are younger and better educated than the average farmer.

Abstract: Recent attention has been placed on whether integrating Information Communication Technology (hereafter, ICT) into education can effectively improve learning outcomes. However, the empirical evidence of the impact of programmes that adopt ICT in schooling is mixed. Theory suggests it may be due to differences in whether or not the ICT pro-grammes are integrated into a teaching programme of a class. Unfortunately, few empirical studies compare the relative effectiveness of programmes that integrate ICT into teaching with the ones that do not. In order to understand the most effective way to design new programmes that attempt to utilize ICT to improve English learning, we conducted a clustered randomized controlled trial (RCT) with some schools receiving ICT that was in-tegrated into the teaching programme of the class; with some schools that received ICT without having it integrated into the teaching programme; and with other schools being used as controls. The RCT involved 6304 fifth grade students studying English in 127 rural schools in rural China. Our results indicate that when the programme is integrated into the teaching programme of a class it is effective in improving student test scores relative to the control schools. No programme impact, however, is found when the ICT programme is not integrated into the teaching program. We also find that when ICT programmes are inte-grated into teaching, the programmes work similarly for students that have either high or low initial (or baseline) levels of English competency. When ICT programmes are not in-tegrated with teaching, they only raise the educational performance of English students who were performing better during the baseline.

The debate over whether boarding school is beneficial for students still exists in both developing and developed countries. In rural China, as a result of a national school merger program that began in 2001, the number of boarding students has increased dramatically. Little research has been done, however, to measure how boarding status may be correlated with nutrition, health and educational outcomes. In this paper, we compare the outcomes of boarding to those of non-boarding students using a large, aggregate dataset that includes 59 rural counties across five provinces in China. We fi nd that for all outcomes boarding students perform worse than non-boarding students. Despite these differences, the absolute levels of all outcomes are low for both boarding and non-boarding students, indicating a need for new policies that will target all rural students regardless of their boarding status.

China’s rapid development and urbanization have induced large numbers of rural residents to migrate from their homes to urban areas in search of better job opportunities. Parents typically leave their children behind with a caregiver, creating a new, potentially vulnerable subpopulation of left-behind children in rural areas. A growing number of policies and nongovernmental organization efforts target these children. The primary objective of this study was to examine whether left-behind children are really the most vulnerable and in need of special programs. Pulling data from a comprehensive data set covering 141,000 children in ten provinces (from twenty-seven surveys conducted between 2009 and 2013), we analyzed nine indicators of health, nutrition, and education. We found that for all nine indicators, left-behind children performed as well as or better than children living with both parents. However, both groups of children performed poorly on most of these indicators. Based on these findings, we recommend that special programs designed to improve health, nutrition, and education among left-behind children be expanded to cover all children in rural China.

Nearly nine years following the release of the Center for Global Development’s When Will We Ever Learn? Improving Lives Through Impact Evaluation report, and almost a decade into increased focus on evaluation among global donors, many in the research community are reflecting on the state of the impact evaluation field, whether the development community is learning what was hoped to from impact evaluations and where the future of impact evaluation leads. 

As part of this reflection, in this paper we will explore the recent past, current status and future of impact evaluation of development interventions. 

Abstract: Independent reading—unassigned reading for personal pleasure—has been shown to be an important driver of reading skills and academic success. Children that commonly read for pleasure exhibit higher academic performance. However, little research has been done on independent reading in rural China, where the education system is charged with schooling tens of millions of students. Many rural students fall behind their urban counterparts in school, with potentially troubling implications for China’s ongoing development. This article explores the prevalence of independent reading and its associations with reading ability and academic performance among rural students. Using a mixed methods approach, we analyze quantitative data from a survey of 13,232 students from 134 rural schools and interviews with students, teachers, principals, and caregivers. We find that independent reading is positively and significantly correlated with reading ability as well as standardized math and Chinese tests scores. Despite such correlations, only 17 percent of students report reading for pleasure for an hour a day. Interview findings suggest that inaccessible bookstores, curriculum constraints, unsupportive home environments, low availability of appealing and level-appropriate books, and insufficient school investment in reading resources may explain the low prevalence of independent reading.

We estimate the impact of two early commitment of financial aid (ECFA) programs—one at the start and one near the end of junior high school (seventh and ninth grades, respectively)—on the outcomes of poor, rural junior high students in China. Our results demonstrate that neither of the ECFA programs has a substantive effect. We find that the ninth-grade program had at most only a small (and likely negligible) effect on matriculation to high school. The seventh-grade program had no effect on either dropout rates during junior high school or on educa- tional performance as measured by a standardized math test. The seventh-grade program did increase the plans of students to attend high school by 15%. In examining why ECFA was not able to motivate significant behavioral changes for ninth graders, we argue that the competitiveness of the education system successfully screened out poorer performing students and promoted better performing students. Thus by the ninth grade, the remaining students were already committed to going to high school regardless of ECFA support. In regards to the results of the seventh grade program, we show how seventh graders appear to be engaged in wishful thinking (they appear to change plans without reference to whether their plans are realistic).

The goal of this study is to examine whether promising a conditional cash transfer (conditional on matriculation) at the start of junior high school increases the rate at which disadvantaged students matriculate into high school. Based on a randomised controlled trial (RCT) involving 1418 disadvantaged (economically poor) students in rural China, we find that a CCT voucher has no effect on increasing high school matriculation for the average disadvantaged student. The CCT voucher also has no differential impact on students at any point in the distribution of baseline academic achievement. This result suggests that CCTs, while shown to be effective in many contexts, do not always work.