Organizer: Scott D. Rozelle, , USA Discussant: Li Han, Hong Kong University of Science & Technology, Hong Kong Social experimentation has swept through the field of development economics. Frustrated by traditional analytical methods that at most produce rough measures of correlations, in recent years economists have begun to turn to Randomized Control Trials (RCTs), methods commonly used in agricultural experimentation and drug testing as a way of identifying cause and effect relationships. With evidence-based research, academics have been trying to make a larger impact on policy. Research groups like the Poverty Action Lab in Harvard/MIT; Center for Global Development at Berkeley; and the Growth Center at Yale have websites that detail the findings from thousands of RCTs.
Interestingly, almost NONE of these social experiments have been done in China. This is ironic and perhaps tragic since arguable no where else in the developing world is there more potential for using policy-relevant results to upscale findings from a pilot project to a much larger set of agents. China’s government is not short of fiscal support. Officials mostly want to address the needs of those in poor areas. In fact, the government is in the middle of a concerted effort to invest in education, health, nutrition and governance. One of the most frequently heard complaints of leaders in recent years is that they are ready to implement new projects and programs if they knew how to spend their money effectively and efficiently.
It is for this reason that a new group of scholars (mostly economists—inside and outside of China) has begun to implement rigorous impact evaluations in China that are empirically based, policy-relevant and independently executed. The objectives of these teams of scholars are to identify programs and projects that work, produce clear and simple evidence-based research statements and communicate results to government decision makers who may be able to make a difference in the lives of many more.
In this proposed session, we will look at three such efforts: Presentation 1 examines how a training program can increase the participation of women and the “quality” of their votes in China’s village elections (in Fujian). Presentation 2 reports on the findings of a study that gives multivitamins with iron to poor students in Shaanxi, Ningxia and Qinghai provinces and examines how student health improves and test scores rise. Presentation 3 provided cash prizes and peer tutoring to students in poor migrant schools in the outskirts of Beijing and examines the impacts of these programs on the performance of some of the lowest performing students in China.
Although national voting statistics show that more than 90% of individuals in rural China vote, there is concern that the actual participation of women in village committee elections is lower. The goal of this paper is to provide an empirical basis for understanding the voting behavior of women in rural China and to evaluate whether or not if it is possible to increase the participation of women.
To meet this goal, we have undertaken a randomized control trial in 36 villages in Fujian Province. After running a baseline survey on a subset of ten women per village, three interventions occurred. In Group 1 a trainer using a standardized curriculum trained women about the problems facing women in being able to exercise their right to vote as well as their rights and responsibilities of voting. In Group 2 the trainer trained village leaders using the same curriculum. In Group 3 the trainers trained both the women and the village leaders. Group 4 was the control.
According to our data, the knowledge of women about their rights to vote was still low and the absence of knowledge is correlated with low voting. After the training in Groups 1/3, the knowledge of women about their voting rights rose (vs. Groups 2/4). When women learned more about their right to vote and how to vote, the rate of participation rose; the nature of their voting behavior improved. Policy-wise, the results show that China needs to increase its effort to promote elections—especially among women.
A significant share of students across rural China still has inadequate access to micronutrient-rich regular diets. Such poor diets can lead to nutritional problems, such as iron-deficiency anemia, that can adversely affect health as well as learning in school. The overall goal of this paper is to test whether simple nutritional interventions lower rates of anemia and improve educational performance among students.
To meet this goal we carried out a randomized control trial (RCT) involving over 3600 fourth grade students, mostly aged 9 to 12, from 66 randomly-chosen elementary schools in 8 of the poorest counties in Shaanxi Province in China’s poor northwest region. The design called for random assignment of schools to one of three groups: two different types of treatment/intervention schools; a non-intervention, control group. One intervention provided a daily multivitamin with mineral supplements, including 5 milligrams of iron, for 5 months. The other informed the parents of their child’s anemia status.
Some 38.3 percent of the students had Hb levels of below 120 g/L, the World Health Organization’s cutoff for anemia for children 9 to 12 years old. In the schools that received the multivitamins with mineral supplements, Hb levels rose by 0.2 standard deviations. The math test scores of the students in the schools that received the multivitamins also improved significantly. In schools that received the information treatment, only students that lived at home (and not students that lived in boarding schools and ate most meals at schools) registered positive improvements in their Hb levels.
Despite concerns that monetary incentives for learning may be detrimental in the long run, economists and policymakers are increasingly interested in experimenting with the idea of paying students to learn as a solution to the failing education situations in many parts of the world. Implicitly we assume that the motivations of students are central inputs into educational outcomes.
If we really take student motivation seriously, paying students to learn is only one possible channel of remedying education. Paying students to help each other may be another channel. Even though economists recognize peer learning as an important channel of peer effects, the large literature on educational peer effects takes student motivation to interact with each other as given.
In this paper we report on the design and results of an incentivized learning field experiment performed in China. In the field experiment we offered competitive prizes to under-performing students in 23 migrant children primary schools to encourage them to improve learning. In half of the program classes, treated students also got help from peer tutors whose bonus scheme was linked to tutee’s score improvement. The peer tutoring program improved the test scores of the under-performing tutees by 0.14 standard deviations over a semester, while the self motivation program without peer tutoring did not have a significant impact. Additional parental intervention boosted tutoring effect to about .2 sd, but was not effective in the self motivation program. Our program results suggest that improving teaching is crucial for remedying education.
|
