Meta-Analysis Of The Educational Return Rate In China

Education is a century-long project.Education provides high-quality labor force for economic construction and brings corresponding returns to the labor force.The reasonable estimation of the rate of return on education is a very important research field.In theory,the method of estimating the rate of return on education in our country is mainly based on adding some instrumental variables on Mincer equation.Different scholars estimate the rate of return differently due to different data and controlled different variables.From the actual published literature,286 papers on the rate of return on education in China were retrieved and 30 papers(58 independent studies)were selected according to the screening criteria.After preliminary analysis of 58 independent studies of these 30 articles,it was found that the lowest rate of return on education in China was only 0.023,and the highest rate was 0.335.This shows that the research on the rate of return on education in China is indeed controversial.In order to obtain a representative result of the rate of return on education in China,this paper uses meta-analysis to synthesize the conclusions of these literature studies to estimate the rate of return on education in China.Firstly,because of the controversy of the rate of return on education in China,this paper uses meta-analysis to synthesize it to get representative results.Secondly,because the heterogeneity test results of the meta-analysis in this paper show it is heterogeneous,this paper uses subgroup analysis and variance analysis from three factors: the types of papers publishing,the time of papers publishing and the source of data in papers to discuss the heterogeneity of the meta-analysis model in this paper.Then,because the meta-analysis in this paper appears the issue of publication bias,which will have a great impact on the results of meta-analysis,this paper uses the selection model and a simulation sensitivity analysis method to correct the publication bias.Finally,this paper makes some comparative analysis to further verify the correctness of the conclusions of this paper.The results of this study show that: Firstly,the meta-analysis result is 0.2067 when the publication bias is not corrected.Secondly,about heterogeneity,the types of papers publishing,the time of papers publishing and the source of data in papers are not the reasons for the heterogeneity of meta-analysis,which indicates that the meta-analysis of the rate of return on education in China can ignore the impact of these three factors on the heterogeneity of meta-analysis in the future.Thirdly,about the problem of publication bias,The meta-analysis result of corrected publication bias is 0.1916,which was about 7.3% lower than the bias.Fourthly,the results of comparative analysis further validate the rationality of the selection equation used to modify publication bias,explain the validity of the conclusion that publication bias is overestimated and the degree of overestimation is small,and further analyze the rationality of the result that the comprehensive effect value is 0.1916.