The Effect Of Number Of Siblings And Birth Order On The Development Of Junior High School Students

With the continuous development of China's economy and society,the country has issued a series of outline documents on health education,advocating health education,while strengthening scientific and cultural education,it has gradually attached importance to children's mental health.The junior high school stage is not only a critical period of a child's academic studies,but also a critical period of its physical and psychological development.A healthy and positive psychology helps to form a healthy and good character and correct values.The difference in the number and birth order of children in the family will affect the child's performance and psychology,which will affect the child's own development.This article uses data from China Education Panel Survey(CEPS)for the 2014-2015 school year to analyze the impact of the number of siblings and birth order on middle school students' performance and mental health.According to the research content,the student questionnaire,parent questionnaire and school leadership questionnaire were processed to extract relevant variables.In this paper,two methods of spearman correlation test and univariate feature selection are used to screen for covariates,and covariates related to outcome variables are screened.Because the number of siblings and birth order are not randomly generated,they are affected by various confounding factors such as family economic conditions,parental education,and local fertility policies.In order to eliminate the estimation bias caused by confounding factors,this article uses the GBM propensity score weighted method and the regression adjustment method to estimate the processing effect.The propensity score method is divided into matching method,weighting method,layering method,and covariate adjustment method.Among them,the matching method is the most widely used,followed by the weighting method.This article will use the GBM(Generalized Boosted Model)to estimate propensity scores,because the GBM model is a non-parametric model and does not need to assume a functional relationship between the variables.There are non-linearities,interaction effects,etc.between processing variables and confounding factors.The determined functional relationship has a significant advantage over parametric models such as Logistic.In addition,the GBM model also has advantages when dealing with high-dimensional features.The propensity score is used to calculate the weights for weighting,and the regression adjustment method is used to estimate the processing effect.The estimated processing effect value has double robustness.In addition to using the propensity score weighting method,this article also uses OLS regression and propensity score matching methods to estimate the processing effect.Among them,the propensity score matching method uses the nearest neighbor one-to-one matching method to match the results of these two methods with the propensity score.The results of the method are compared to obtain more reliable conclusions.In order to obtain accurate conclusions,in addition to testing the balance of the covariates and the support region,this paper also conducted a robustness test by excluding some covariates.After the robustness test,the estimated value is considered to be robust,and the conclusion has not changed.It also confirms the dual robustness of the GBM propensity score weighting method combined with the regression adjustment method.Through empirical analysis,it is found that the number of siblings and birth order have an impact on their academic performance and mental health.Among them,no sibling group and two or more sibling groups have significant differences in academic performance and mental health.There were significant differences in mental health between the first-born group and the last-born group.According to individual household registration,agricultural household registration and non-agricultural household registration are distinguished,and the groups under different household registrations are significantly different.For agricultural household registration students,there is no significant difference in academic performance between groups with different numbers of siblings,and the final birth group The mental health of the group was significantly higher than that of the first-born group;for non-agricultural students,the academic performance of the group without siblings was significantly higher than that of groups with two or more siblings.There are no significant differences.The article concludes with suggestions for schools and parents based on the results of empirical research.It is hoped that parents and schools attach importance to their children's mental health.Schools in rural areas should provide corresponding courses or lectures on mental health counseling.