Skip to:


Counseling, Vouchers, and High School Matriculation

China's national government still struggles to keep students in school through high school

Compulsory education in China ends with grade 9. In the past five years, fiscal input by the national government has made grades 1 to 9 almost free. This effort—and other factors—have contributed to making compulsory education almost universal. But something is still missing...


While compulsory education (grades 1-9) is now almost universal, disadvantaged students across China still attend high school at alarmingly low rates. 

We have identified two potential barriers to high school matriculation in China: financial constraints/poverty, and imperfect information about the returns to schooling. 

  • High school tuition rates in China are the highest in the world. For a family living at or near the poverty line, the cost of high school attendance for one child may be the equivalent of 15 years of per capita income.

  • Another obstacle to attending high school might be a lack of trustworthy and accurate information. Poor children growing up with uneducated or migrated parents and attending under-funded schools may not get the advice and information they need about why attending high school is so important.

Differing strategies exist to address these challenges. For example, the impact of offering vouchers (or scholarships) has been studied before and is consistent with microeconomic theory. Our own work has shown that guaranteeing high school students financial aid for college has a significant impact on their educational choices (see REAP's financial aid project). However, the effect of similar interventions aimed at increasing high school matriculation rates among middle school students has not been studied.


This study aims to estimate the effects of financial vouchers, informational counseling, and a joint counseling-voucher intervention on the decision of poor rural students to attend academic high school. More specifically:

  • How would student education choices change if they are given a conditional cash transfer in the form of a voucher for academic high school?

How would choices change if students are informed of the different options for schooling and what the choices typically lead to, both in terms of career choices and wages? Using a randomized controlled trial approach, we randomly selected 90 treatment schools and 30 control schools in Hebei and Shaanxi provinces. Shaanxi is a Western inland province, whereas Hebei is near the coast (around Beijing), so there may be substantial differences between students in these two provinces, enabling us to test the efficacy of our programs across a more diverse population. We assigned treatment and control conditions at the school level to avoid spillover effects (for example, students receiving counseling may tell their friends who are in the control group about what they have learned).

The voucher intervention is a commitment from the research team that if awardees are accepted to academic high school at the end of 9th grade and choose to go, we will provide funding to cover the tuition for each of the three years that they are enrolled. We tested this voucher program among both 7th and 9th grade students.

The “short counseling” intervention is a half-hour session where students receive information on available educational choices and the returns to the different educational choices available. Students should leave this session with knowledge that students who do not attend high school tend to have lower earnings. We tested this intervention among 7th grade students.

A “long counseling” intervention spans the course of the school year. Students are encouraged to develop goals for the future, what they want to do with their lives, and consider how high school can help them achieve their goals. Students should leave with increased self-efficacy—an idea of what their goals are and a sense of ownership over their future. We tested this intervention among 7th grade students.

Out of 120 total schools, 30 schools received vouchers only; 15 schools received vouchers and short counseling, and another 15 received vouchers with long counseling. 15 schools received long counseling only, and another 15 schools received short counseling only. 30 schools will be assigned to the control condition. 

Once the schools were selected, a short 1-2 page baseline survey on the socio-economic characteristics of the students and their families was collected, and all students were tested with a grade-level appropriate standardized examination. 


We began the interventions in Month 1, then checked in on the sample schools in Months 11 and 23.  In schools with the voucher intervention, we reaffirmed our commitment to supply a tuition voucher to students who make it to high school. In months 32-35, we resurveyed schools, giving vouchers to those enrolling in academic high school as appropriate. 

In month 68 (time of High School graduation, 3 yrs after middle school), we resurveyed graduates of our sample schools to determine which students went on to university, which students pursued other education, and which students joined the labor force (and in what sectors).


The voucher program had no substantive effective. Using causal chain analysis, we argue that the competitiveness of the education system successfully screened out poorer performing students and promoted better performing studetns. Thus, by ninth grade, the remaining students were already committed to going to high school regardless of voucher support. Furthermore, we show that seventh graders appear to be engaged in wishful thinking--although their plans to attend high school improve when they learn about the voucher, few of these students actually end up enrolling in high school after they graduate junior high.

The short and long counseling interventions, on the other hand, did have a significant effect, although they actually acted to increase dropout and lower academic achievement, not increase high school enrollment. In our analysis of the causal chain, we conclude that financial constraints and the poor quality of education in junior high schools in poor, rural areas may be contributing to the absence of positive impacts on students outcomes from counseling. The negative effects of counseling may be due to the high and growing wages for unskilled labor (high opportunity costs) in China's transitioning economy. It is possible that when our counseling curriculum informed the students about the reality of how difficult the requirements for entering academic high school were to meet, it may have induced them to revise their cost-benefit calculations and drop out or work less hard in school. 


Thank you to the 3ie Foundation for funding this project.