论文发表,代写代发论文

　　
　　Table 4-1: Estimate results of Multinomial logit model on general management and professional and technical personnel
　　
　　Table 4-2: Estimate results of Multinomial logit model on enterprise ordinary employees
　　
　　
　　Table 4-3: Estimate results of Multinomial logit model on individual operators in transport, wholesale and retail and other industries
　　
　　From table 5, it can be seen that Cox & snell value and Nagelkerke value of the model reached 0.053 and 0.575 respectively. For the cross-sectional data, this suggests that the general fitting effect of the model is good.
　In multiple-linear regression model, null hypothesis "all coefficients except constant are equal to 0" is often tested with F-test. In Logit model, however, the same purpose is often reached with likelihood ratio test; whether Logit model is statistically significant is tested with the approximate obedience of likelihood ratio statistics to Chi-square distribution (Aldrich & nelson, 1984; Greene, 1990). From the test results in table 6, it can be seen that the likelihood ratio between the originally assumed model with all variables except Intercept equal to 0 and the model added with all explanatory variables was 337.644, and the overall regression result of the model rejected the original assumption at a significance level of 1%, indicating all explanatory variables generally possess a powerful explanatory power on the model.
　　
　　Table 5: Model R2 test
　　
　　5.Research Results
　　From the regression results in table 4, five aspects can be seen clearly. First, the estimate coefficients of educational background were very significant, keeping consistent with expectation. Second, the estimate values of age on management personnel, technicians, company employees and service personnel were negative, keeping consistent with expectation hypothesis. Third, in technique and management jobs (Table 4-1), gender played a powerful positive effect on migrant labors to become management and professional technical personnel, and the estimate coefficient was .258, opposite to expectation hypothesis. Fourth, the estimate values of the human capital variables (vocational training, professional skills and non-agricultural work experience) on migrant labors in management and technical positions were positive 2.117, 1.899, and .155 (Table 4-1) respectively, indicating vocational training, professional skills and non-agricultural work experience played a powerful strong positive effect on migrant labors in management and technical positions. Fifth, the estimate coefficients of the dummy variables of the groups from different regions were basically not significant, indicating regional differences were not the primary factors deciding the selection of the job types of migrant labor in the investigated samples of this paper. Sixth, health factor played a significant effect on migrant labor in management and technical positions, indicating these types of jobs propose a higher requirement on physical quality in comparison with the fifth-type job.
　　
　　6.Conclusion
　　Low human capital accumulation is an important factor restricting rural migrant labor to work in high-level industries of urban areas. Regardless of economy, institutions and policies, the effect of human capital accumulation on the employment of rural transfer labor is mainly reflected from three aspects. First, the educational background of China's rural migrant labors remains poor, and their average schooling years were 8.5 years. Second, vocational training is received only by a small number of migrant labors. Third, the generally low human capital forces a majority of rural migrant labors to flow to the fifth-type (dirty, tired and dangerous) jobs that urban people are unwilling to do. This conclusion is fully in line with the principle of market allocation of resources. In the long run, to speed up the human capital accumulation of rural population and enhance rural human capital is not only helping Chinese peasants to get non-agricultural employment and increase income, but upgrading urban industrial structure and meeting the needs of urbanization.
　7.Acknowledge
　　Fund Project: This paper is supported by National Social Science Foundation Project (No. 10BJY031).
　　
　　8.References
　　[1]Qiren Zhou. Enterprises in Market: A Special Contract between Human Capital and Non-Human Capital [J]. Economic Research, 1996 (6).
　　[2]Ling Yu. Study on Human Capital Constraints of China's Rural Labor Mobility. Zhejiang: Zhejiang University, 2002.
　　[3]Yaohui Zhao. China's Rural Labor Mobility and the Role of Education In It—Study based on Sichuan Province. Economic Research, 1997 (02): 37-42.
　　[4]Ling Zhu. Investment in Health and Human Capital Theory. Economic Perspectives, 2002 (8).
　　[5]Yang Du. Study on the Effect of Education on Non-agricultural Labor Supply of Peasant Households in Poverty-stricken Areas. Chinese Journal of Population Science, 1996 (6). (责任编辑：南粤论文中心)