The Education Performance of Migrant Students in Urban Public Schools in China
Yuanyuan Chen[1]
School of Economics, Shanghai University of Finance and Economics
Shuaizhang Feng
School of Economics, Shanghai University of Finance and Economics
Abstract
A large proportion of migrant children in Chinese cities entered public schools together with their local city children. This paper compares the standardized Math test score and student behavior between the two groupsusing data collected in Shanghai in 2010 and 2012.We found that for students within the same school, the two groups don’t have significant difference in the test score in mathematics, teacher evaluated behavior or class leadership. Across schools, student test score are highly negatively correlated with the fraction of migrant studentsin school. We interpret that schools with better quality are more likely to adopt students from families with better social-economic status. Given family social economic background, hukou is not an important sorting factor. We also find that within schools, the class ratio of migrant students do not matter for student performance.
Key words:migrant students; student ratio; public school
JEL Classification: I28; J15; O15
- Introduction
With the development of China’s urbanization, a large proportion of rural workers have been migrated to work in cities. According to 2010 census statistics, there are 220 million migrant workers in cities. Among these migrant workers, around 9 million are children are elementary or middle school students. The education of migrant students has been a major challenge in the process of urbanization.
Before 2001, migrant children are not entitled to enter local public schools as compulsory education is based on Hukou system. Only of a small proportion of children could manage to enter public schools, mostly by paying an expensive extra fee. The rest large proportion of migrant children had to enter private migrant schools which have much lower quality than public schools (See Chen and Feng，2013). Since 2001, the central government has urged the host city government to take the responsibility for the education of migrant children. As a result, the proportion of migrant children entering in public schools has been increasing. According to the survey of the Chinese State Department in 2012, a large proportion of migrant children have now entered in public schools in major host cities. In Beijing, 88% of migrant children are hosted by public schools; around 70% in Shanghai; 41% in Guangzhou and 46% in Shenzhen.
The education achievements of these migrant students in public schools are important for the following reasons. First, given that migrant children account a large proportion of children in urban cities, whether these children can obtained equal education opportunity will definitely determine the success of economic development and social sustainabilityin the next few decades. Second, the ultimate goal for the central government is to allow all migrant students in the city to have the equal education opportunity as urban students. Therefore, the number of migrant students studying in public schools is expected to grow. Third, the success experience of migrant students studying in urban public schools will encourage migrant parents to bring their children left-behind in the rural village to come along with their parents. There are around 60 million left-behind children and they have suffered a significant loss in their education, cognitive achievementsand psychological health because of the absence of parents (Zhang et al., 2014).
This paper aims to evaluate the performance of migrant students in public schools. Using our self-collected data from 11 public elementary schools, we find that within schools, migrant students perform as well as their local peers. Specifically, we don’t find significant difference between migrant and local students in terms of their standardized Math test scores, teachers’ evaluation, the probability of being class leaders and self-reported psychological behaviors. We find no evidence of discriminationtoward migrant students within schools.
Within school comparison, however, does not reflect the whole picture of the education situation of migrant students. The school entrance is not random. There might be sorting of students from different family background to schools with various quality. To see this, we construct a school quality index and confirm that for both migrant and local students, those with better family background are sorted into schools with better quality. This sorting is even severe for migrant students. In addition, we also find that migrant students are less likely to enter schools with better quality.This sorting is mostly because migrant students are from families with worse social economic situation. Given family background, hukou is not important in school sorting.
The sorting across schools occurs when parents with better social economic background make their effort to enroll their children into schools with better quality or schools with lower ratio of migrant students. The latter happens when parents, especially local parents, think that mixing with migrant students will give negative peer effects to their children. Since 2008, as shanghai government closed all migrant school in central district and public schools start to enroll a significant number of migrant students, some parents decide to transfer their children out of schools with higher ratio of migrant students. In our survey, about one third of local parents think that schools should separate local students from migrant students in terms of class assignment. There are also 32% of local parents think that admitting migrant students will lower the school quality. And around 10 percent of local parents consider transferring their children to schools with lower ratio of migrant students.
Hence, it is important to examine whether exposure to higher fraction of migrant students cause lower achievement. That is whether the segregation rate of migrant students matters for student achievement. Our paper is the first study that evaluates the segregation impact of rural to urban migrant students in urban schools in China. Taking into account the sorting of students from different family background into schools, we use class level ratio of migrant students to evaluate the impact of segregation on student achievement. We find that class level ratio of migrant students is not statistically related with student performance. Segregation rate do not matter for the achievement of either migrant students or local students.
- Background and Literature Review
(1)Background of enrollment of migrant students in urban public schools in China
One of the main drivers of China’s phenomenal economic growth during the past three decades has been the large-scale and persistent rural-to-urban migration. As a consequence, a half of China’s 1.3 billion people now live in cities, as compared to only twenty percent in the early 1980s (NBS, 2012). Nevertheless, researchers have expressed concerns over China’s so called “pseudo-urbanization” (Henderson, 2009; Yew, 2012). The majority of rural migrants in cities do not possess local household registration (HuKou), thus are excluded from full access to pension, health care, public education and other social benefits at the place they live. Based on the 2010 Census, there are around 210 million such “non-HuKou” migrants who work and live in a city other than where his/her official HuKou is registered (NBS, 2012).[2]
Among all non-Hukou migrants, approximately 20 million are children agedbetween 6 and 14 (NBS, 2012). Migrant children account for significant proportionsof students in major host cities. According to the survey result from the State Department, migrant students account for 29.1% of all students in elementary and middle schools in Beijing, around 37% in Shanghai and Guangzhou, and 53.4% in Shenzhen. Although all Chinese school-age children are entitled to a free and compulsory 9-year education by law, the Chinese educational system designates elementary education to local governments at the county/district level. Funding for elementary education is allocated by the number of children with HuKou and not portable across administrative units. Therefore, local authorities lack the incentives and financial resources to accommodate the educational needs of migrant children, despite directives from the central government that urge them to do so.
In the early 1990s, there were already a large number of children migrating with their parents and living in cities. But public schools at that time charged hefty “out-of-district” fees to anyone without a local HuKou, and required an array of documents and certificates which many migrant parents were unable to provide. Even if some migrantparentswere able to afford the tuition and to provide all necessary documentations, they might still not be able to enroll their children in a public school as spaces in public schools were limited. A lot of migrant children had to enroll in private run migrant schools with much lower quality (Chen and Feng, 2013).
Since 2001, the central government has urged the host city government to take the responsibility for education of migrant children. Gradually, the host cities started to release the school enrollment policy. Hukou is no longer restricted required for enrollment in public schools. Some cities require that students can enroll in public schools if their parents have residence permit or legal jobs in the city. Migrant students can enter public schools that are close to their residence and have seats available. Since then, the proportion of migrant students in public schools in urban cities has increase sharply. Nevertheless, policies in difference cities differ very much. Usually, the policy is more restricted for large cities with a lot of population pressure. In the following section, we will discuss the policy in Shanghai in more detail.
(2)Background of enrollment of migrant students in urban public schools in Shanghai
As the major city with the largest population in China, Shanghai hosts a large population of migrants. According the Shanghai census data, the migrants has increased from 3.87 million in 2000, to 5.81 million in 2005, to 8.97 million in 2010. In 2010, the migrants have account for 39% of the total population in Shanghai. According to the survey data from the State Department, for elementary and middle school, there are 0.44 million migrant students, accounting for 37.0% of all students in compulsory school age. The proportion is even higher for elementary school, because a lot of migrant student go back to home province since they cannot attend high school and college entrance exam in Shanghai. For elementary schools, there are 0.33 million students, accounts for 44% of all elementary students. Among these migrant students, there are 0.2 million students enrolling in public elementary schools.
We then turn to discuss the enrollment policy of public elementary schools in Shanghai. The enrollment is mostly based on school district. Both local and migrant students report their residence, hukou status and housing ownership to the education department in local district. The education department ranks students by the above information, and assign them to local schools. The first priority is given to students with both shanghai hukou and the home ownership within the school district. If the public school within the district has more seats available, the education department will assign them to other students living in the school district with the priority order as follows: students with home ownership and shanghai elite special resident permit, students without home ownership but with shanghai hukou from other districts, students with home ownership but without shanghai hukou. The lowest priority are for students without shanghai hukou and home ownership. These include students coming from other provinces or from abroad and their parents have legal shanghai residence permit. The schools will enroll students from the highest priority to the lowest, as long as the seats are available. For students who cannot enter the school in their school district, the education department will assign them to other schools as close as possible.
In this system, it is naturally that there exists sorting of students from different family background to schools with different quality. The housing price in different school district, to some degree, also reflects the quality of their public schools. Parents with higher socioeconomic status will buy house in the area with better quality public school resource. For students with shanghai hukou, if their parents cannot afford for higher housing price in better school district, they will stay in the school district regardless of its public schools district. For students without shanghai hukou, if their parents choose to buy house or live in better school district, they can also manage to enter better schools if the seats are not fully occupied by shanghai hukou students. As for top elite public schools, the competition of enrollment is fierce and they only enroll students with both shanghai hukou and local home ownership or students with strong social connection. These schools are unlikely to enroll migrant students and they are not in our public sample.
(3)Literatures on education of migrant children
The education of migrant children has received considerable attention by both Chinese and international scholars. To our knowledge, there is no paper quantitatively studying the performance of migrant students within public schools. Most of these studies are comparing the relative qualitybetween migrant schools and public schools. Chen and Feng (2013) use standard test scores to find that students in migrant schools perform considerably worse than their more fortunate counterparts in public schools in Shanghai. Lu and Zhou (2013) find similar result in Beijing that migrant children have poorerachievementand greaterloneliness in migrant schools than public schools. Liu et al. (2015) finds that mathematics achievements of migrant children in migrant schools are determined by a lot of family or school related characteristics while mathematics achievements of migrant children in public schools are only determined by family socioeconomic status.
Beside the comparison on the static quality between migrant schools and public schools, there are also a few studies that compare the dynamic changes of migrant schools. Chen and Feng (2015) find that the quality of migrant schools in Shanghai has been improved as government subsidies for migrant schools increased. Lai et al. (2015)compares migrant schools in Beijing with rural schools in Shaanxi. They find that students in migrant schools perform significantly better than their counterparts in rural public, but they made less progress due to the poorer quality of migrant schools in Beijing.
(4)Literatures on school segregation
The impact of segregation of children in minor group in schools is widely studied in international literatures. Segregation can affect the relative education achievement of students through several mechanisms. One of the most widely discussed channels is a peer exposure effect, arising from the fact that students’ outcomes depend on the performance of their peers, and from a presumed correlation between these characteristics and the composition of the group. Identifying whether exposure to a higher fraction of minor peers actually causes lower achievement is difficult, however, because individuals are not randomly assigned to schools. Minor student enrollment share at a school may be correlated with school quality (Boozer et al, 1992, Card and Krueger, 1992).
The existing literatures have not reached a consensus whether segregation have a negative impact. Card and Rothstein (2007)study the effects of school and neighborhood segregation on the relative SAT scores of black students acrossdifferent metropolitan areas and find that segregation increase the black and white gap in SAT score. Flores and Scorzafave (2014) find that in Brazil the higher the school segregation is, the higher the ratio of black white score differential. Hanushek and Rivkin (2008), Hoxby (2000), Hanushek et al. (2002) also identifylarge significant negative effects of exposure to black classmates. Using a school district assignment change, Billings et al.(2014) find that both white and minority students lower their test scores when they are assigned to schools with higher ratio of minority students. On the other hand, Angrist and Lang (2004) study a desegregation program called Metco that sends students from Boston schools to more affluent suburbs. They find that Metco has no impact on white students, but a modest effect on blacks, esp. girls.
Most of these research studies the impact of segregation between black and white students on children’s achievement. There is a small literature examining whether ethinic concentration affects the school performance of immigrants. Cortes(2006) finds that attending an enclave school (above 25 percent foreign-born) has no effect on students test score. Nordin (2013) findsa negative effect of having a large share of immigrant schoolmates.
- Data and measurement
3.1 Data