Selection Bias and Treatment Methods: A response to the questions raised by Yang Ge and Wang Guangzhou

本文是针对杨舸和王广州的《户内人口匹配数据的误用与改进》所提出的商榷意见的回应。基于杨文对人口普查和抽样调查数据匹配所产生的选择性偏差的讨论分析以及数据再检验结果,作者对于调整数据偏差方法及其效果进行了详细分析和数据论证,其结论是:杨文提出的商榷意见没有得到数据分析的支持,同时,杨文所采用的调整偏差的方法也不能有效地估计和解决作者相关研究的选择性偏差问题。

This article is the author’s response to the questions raised by Yang Ge and Wang Guangzhou about selection bias of data used by the author’s previous research． Based on discussion and data re-testing made by Yang and Wang about selection bias of reconstructed data from population sample database，the author examines the effect of selection bias and evaluates the effectiveness of treatment methods． It concludes that effect of selection bias has not been verified by data re-testing． The treatment methods used by Yang and Wang could not estimate exactly the effect of selection bias and also not reduce its impact effectively．

欢迎阅读《社会学研究》！您是该文第1660位读者！