社会调查质量研究：访员臆答与干预效果  

A Study on the Quality Control of Social Survey:Intervention in "curb-stolen" behaviors and its effects

作　　者：严洁[1] 邱泽奇[2,3] 任莉颖[4] 丁华[4] 孙妍[4] 

机构地区：[1]北京大学政府管理学院 [2]北京大学社会学系 [3]中国社会与发展研究中心 [4]北京大学中国社会科学调查中心

出　　处：《社会学研究》2012年第2期168-181,245,共14页Sociological Studies

基　　金：国家自然科学基金资助项目"并行数据和调查数据质量管理(71171004)"成果之一

摘　　要：本文利用中国家庭动态跟踪调查并行数据和问卷数据,采用事后准实验设计方法,对面访调查中访员臆答及对臆答进行干预的效果进行了分析,指出访员臆答是面访调查中不可忽视的直接影响调查数据质量的因素;访员的职业特征与臆答的发生比有显著关联,计算机辅助面访调查可以有效监测并干预访员臆答;干预可使臆答显著减少,减少的幅度因访员年龄不同而有显著差异,但不因访员性别、教育和职业因素而变化。Based on a post quasi-experimental design,this paper examines curb-stolen behaviors among field interviewers and the effects of behavior intervention using the data and paradata of CFPS(Chinese Family Panel Studies) in 2010.The study shows that curb-stolen behaviors have directly affected the quality of survey data and their occurrences are significantly related with interviewers' occupations.These behaviors could be successfully detected by the CAPI(Computed-Assisted Personal Interviewing) technique and effectively reduced by timely intervention.The effects of intervention,however,vary with interviewers' ages but are indifferent to their gender,education levels,and occupations.

关 键 词：臆答 问卷调查 质量监控 CAPI 

分 类 号：C915[经济管理]