比较政治学研究选择性偏差及其规避探索

Selection Bias in Comparative Political Studies and its Avoiding Strategies

研究方法对任一学科发展均具有基础性影响。伴随比较政治学对其他学科研究方法的吸收借鉴和自身发展,在学者应用过程中,无论是理论构建环节的案例选择,还是因果推论中的变量或参数设定均存在低估"关联效应"或形成伪因果关系的风险,内置于研究方法的"选择性偏差"最终干扰研究结果的信度和效度。本文通过分析比较政治学不同研究方法使用中选择性偏差产生原因和具体来源,提出选择性偏差"被动型规避策略"和"主动型规避策略"两种不同的解决路径。具体而言:既基于研究实践经验,着重就代表性案例确定标准与选取策略,支持性案例(positive cases)与否决性案例(negative cases)检验,理论定位与案例选择关系,定量研究抽样设计与抽样过程,二手数据使用与筛选标准,模型类型与建模目标匹配等"被动型规避策略"进行讨论,亦从研究设计层面探讨主动型选择性偏差规避。实现为学者未来比较政治学研究提供基础性方法工具清单,最终促进学科研究方法进步和提升学者比较研究的方法论自觉。

Recently,comparative political methodology has great achievement by drawing on research methods from other subjects. However,there exists some problems in the application of methodology,such as case selection bias,variable selection bias and false causality. This phenomenon,which is called selection bias,interferes with the reliability and validity of the findings. By analyzing the reasons and sources of selection bias in qualitative and quantitative comparative political studies,this study proposes two different solutions to selection bias including passive and active avoiding strategies.From the perspective of passive avoiding strategy,qualitative studies need to focus on case selection standard and keep the balance between positive and negative cases,while quantitative studies need to highlight model matching accuracy,efficiency and adaptability; from the perspective of active avoiding strategy,how to optimize the research design is attached most importance. The aim of this study is to provide a list of methodology tools for avoiding selection bias in comparative political studies,and promote the improvement of methodology in the subject.