In countries other than China, student cadres are rarely used to describe the leaders in student organizations or groups, yet the term “student leader” is mostly used, and similar with the cadre in China. To some extent, the term of student cadre in the Chinese context has a political meaning. It refers in particular to the student group with the cadre position. With a dual identity of “student” and “cadre”, student cadres take part in school public administration, ideological and political education and student affairs. The more capable, the more likely he is to become a student cadre. Thus, this kind of students is probably the elites and this concept has a screening value. Also, being a student cadre can be considered as an educational process, a kind of accumulation of human and social capital which theoretically helps them get a better job in the labor market and enjoys a higher wage premium. Nevertheless, there is a lack of empirical analysis of student cadres as core variable, and a lack of deep analysis of its influence on employment.

Based on the data of the China College Student Panel Survey (CCSPS), this study deeply analyzed the effect of student cadre on wages. Firstly, from the results of basic regression, the student cadre has a significant and positive effect on wages. When adding all the control factors (family, university, and enterprise), we find that the student cadre will increase the starting wage by about 6.98%. The results of Heckman model are similar, which means student cadre does increase the salary of graduates. Also, inverse probability weighting method is employed to check the robustness of the results of OLS estimation. We interpret the effect of student cadre by using IPWRA and AIPW. The treatment effects are 6.49% and 6.88% respectively, which are quite similar to the results of OLS estimate. Therefore, the results of Heckman’s correction and inverse probability weighting method both indicate that the OLS estimation of our study is robust.

Being student cadres is a development process and serving as a student cadre can directly improve the employability. This educational process is based on a certain amount of time. We need to explore his/her professional knowledge, work ability and interpersonal relationships before and after serving as a student cadre. Changes in these three aspects can be accurately interpreted as the effect of the cadres on the ability increase. Professional knowledge uses academic ranking. Work ability and interpersonal relationships are estimated by the scores of the three items respectively in the CCSPS. In order to prove the effect of development, the study uses student cadres as intervention measures. Difference in differences method is used to analyze the influence of the student cadre on professional knowledge, work ability and interpersonal relationships. Being student cadres can increase the work abilities and develop interpersonal relationships by 0.1-0.8 standard deviations, which reveal that the experience of being a student cadre can develop their ability. Then, the variables of professional knowledge, work ability and interpersonal relationships are added in the multilinear regression model to analyze the wage premium of student cadre. The effect of student cadres has been significantly reduced from 0.0698 to 0.0419, with a decline of 40%. Particularly, in the model of interpersonal relationships, student cadres have no significant effect on salary. The scores of these three abilities all have significant and positive effect on wages. Finally, it is fully concluded that the influence of student cadre comes from professional knowledge, work ability and interpersonal relationship. Besides screening signal, serving as a student cadre is an educational process.

Based on previous studies about the economic returns of student cadre and about college graduates’ starting wages, this research controls for the individual employee’s factors, the family factors of parents’ income, the education factors of major type, and the employer’s factors of ownership structure. Here, the ordinary least squares method is used to estimate the coefficients: 〖LnWage〗_i=β_0+β_1 〖Cadre〗_i+β_k X_ki+ε_i LnWagei is the logarithm of monthly income of starting salary for student i. The variable cadre, which stands for whether a student is student cadre. Many variables Xi, includes individual factors besides CEE scores, family factors, educational factors, and employer’s factors. OLS estimator using observational data could have selection bias problem. Assignment to treatment group or non-treatment group is not random in observational study. Data in our study is observational data. This could bias the OLS estimation of the treatment effect. Inverse probability weighting method is often used to solve the problem. This method first attains the probability of assignment into treatment group and estimates separate outcome regression model for subjects in each treatment level with weighting. The treatment effect is the difference of estimated weighted outcomes of different treatment level. We use Heckman’s correction, augmented inverse probability weighted (AIPW) estimation and inverse probability weighted regression adjustment estimation (IPWRA) to estimate the average treatment effect of student cadre. The experience of student cadres is a development process and serving as a student cadre can directly improve the employability. This educational process is based on a certain amount of time. We need to explore the professional knowledge, work ability and interpersonal relationships before and after serving as a student cadre. In order to prove the effect of development, the study uses student cadres as intervention measures. Difference in differences (DID) method is used to analyze the influence of the student cadre: 〖Ability〗_it=β_0+β_1 〖Cadre〗_i+β_2 Time+β_3 〖Cadre〗_i*Time+β_k X_ki+ε_i The dependent variable is the scores (professional knowledge, work ability or interpersonal relationships) of student i in t period (benchmark or graduation). 〖Cadre〗_i is the intervention measure. 〖Cadre〗_i=1 means that student i is not a student cadre in benchmark but has become a student when graduation. 〖Cadre〗_i=0 means that student i is neither a student cadre in benchmark nor when graduation. 〖Cadre〗_i*Time is the treatment effect.

Different from other countries, the student cadre in the Chinese context has a dual identity of “student” and “cadre”. Student cadres take part in school public administration, ideological and political education and student affairs. Based on the comprehensive database of the China College Student Panel Survey(CCSPS) held by the Research and Data Center of Renmin University of China, this study deeply analyzed the complicated mechanism of student cadre on wages. Firstly, from the results of basic regression, the student cadre has a significant and positive effect on wages When adding all the control factors, specifically, the student cadre will increase the starting wage by about 6.98%. Being a student cadre is a screening signal for ability which can be much easily identified in the labor market. Student cadres are also superior to non-student cadres in certain aspects including CEE scores, family background, political status. Additionally, we interpret the effect of student cadre by using IPWRA and AIPW. The treatment effects are 6.49% and 6.88% respectively, which are quite similar to the results of OLS estimate. The results of Heckman’s correction and inverse probability weighting method both indicate that the OLS estimation of our study is robust. Being student cadres also have actually improved employability. With Difference in differences method, we found the mechanism: becoming student cadres can increase the work abilities and develop interpersonal relationships by 0.1-0.8 standard deviations. Serving as a student cadre can significantly develop their abilities in professional knowledge, work ability and interpersonal relationships. Besides a screening signal, being a student cadre is indeed a kind of accumulation of human and social capital which helps them get a better job in the labor market. That is the mechanism of student cadre on wage premium.

Astin, H. S., Astin, H., Boatsman, K., Bonous-Hammarth, M., Chambers, T., & Goldberg, S. (1996). A social change model of leadership development: Guidebook. Los Angeles: Higher Education Research Institute, University of California. Dugan, J. P., Komives, S. R., & Segar, T. C. (2008). College student capacity for socially responsible leadership: Understanding norms and influences of race, gender, and sexual orientation. NASPA journal, 45(4), 475-500. Dunham, R. B., & Pierce, J. L. (1989). Management: Scott, Foresman. Egger, T. M. (2009). Factors influencing College of Agriculture students' participation in leadership development certificate programs: A tri-state study: Purdue University. Heckman, J. J. (1977). Sample selection bias as a specification error (with an application to the estimation of labor supply functions): National Bureau of Economic Research Cambridge, Mass., USA. Heckman, J. J., Ichimura, H., & Todd, P. (1998). Matching as an econometric evaluation estimator. The review of economic studies, 65(2), 261-294. Khandker, S. R., Koolwal, G. B., & Samad, H. A. (2009). Handbook on impact evaluation: quantitative methods and practices: World Bank Publications. Komives, S. R., Dugan, J. P., & Owen, J. E. (2011). The handbook for student leadership development: John Wiley & Sons. Komives, S. R., Lucas, N., & McMahon, T. R. (2009). Exploring leadership: For college students who want to make a difference: John Wiley & Sons. Lyons, P. (1993). Leadership Education Across the Curriculum. Posner, B. Z. (2014). The Impact of Gender, Ethnicity, School Setting, and Experience on Student Leadership: Does It Really Matter? Robins, J. M., & Rotnitzky, A. (1995). Semiparametric efficiency in multivariate regression models with missing data. Journal of the American Statistical Association, 90(429), 122-129. Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41-55. Wooldridge, J. M. (2007). Inverse probability weighted estimation for general missing data problems. Journal of Econometrics, 141(2), 1281-1301. Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data: MIT press.