(WITHDRAWN) The Effects of Education Welfare on Low-Income Children's Stress, Self-Efficacy, and Aspirations: An Instrumental Variables Approach

Background: China's poverty alleviation is transforming towards a new era with developmental objectives. In view of this, promoting life chances of low-income children is one of the top priorities on national agenda, and thus education welfare as the intersection of education and anti-poverty domains is of great policy relevance. Education welfare aims at promoting economic well-being and education opportunities of low-income children. Studies on effects of education welfare have predominantly focused on academic attainments. However, how may the education welfare affect low-income children's socio-psychological functions remains understudied. The socio-psychological functions have been recognized as important predictors of children's achievements. Education welfare as a bundle may affect these aspects through both structural and cultural pathways, whereby the complex effects are largely unknown. This study sets out to fill this gap by empirically investigating the effects of education welfare on low-income children's perception of stress, self-efficacy, and aspirations.

Methods: Data and samples: We use data from the nationally-representative China Education Panel Survey (CEPS) of middle school students in 2013-2014 (baseline) and 2014-2015 (follow-up) waves. The CEPS dataset provides extensive information about students' demographic characteristics and education experiences, family background, as well as school resources and management patterns. We restrict our sample to children with low-income or low socio-economic status background (N=2831). Measures and methods: In order to identify the causal effects of education welfare on low-income children's stress, self-efficacy, and aspirations, we employ an instrumental variable (IV) approach and control for county-level fixed effects and a rich set of covariates. The instrumental variables are derived from school administrator reported indicators of school management and policy rules, such as school type (e.g. public school or private school), availability of education welfare for migrant children and local children respectively, available number of education welfare items, and migrant children's enrollment eligibility to the school. These indicators allow us to exploit policy variations at the school level and thus can rigorously address the endogenous relationships between education welfare participation and outcomes. The F-statistics for our analytic sample is 18.69, which is considered to be acceptable according to the cut-off of 10.

Results: Our preliminary results indicate that the receipt of education welfare reduced children’s perceived stress levels and improved their perceived economic status. Receiving education welfare also decreased children’s likelihood to find going school boring and reduced the incidence of change schools. However, education welfare did not have a significant effect in changing levels of self-efficacy (confidence and hardworking), sense of belonging to school environment, and aspirations to future life.

Conclusions and implications: Our findings yield concern that, although education welfare may financially support low-income children, it may not positively improve children’s self-efficacy and aspiration.