Which Is the Better Investment for Nonresponse Adjustment: Purchasing Commercial Auxiliary Data or Collecting Interviewer Observations?
Which Is the Better Investment for Nonresponse Adjustment: Purchasing Commercial Auxiliary Data or Collecting Interviewer Observations?
复制标题
对于无应答调整，购买商业辅助数据或收集访谈员观察结果哪个投资更好？
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
F. Kreuter
中科院分区：
文献类型：
--
作者：
J. Sinibaldi;Mark Trappmann;F. Kreuter
调查方法学家正在寻找协变量用于无反应调整模型，最终希望找到与感兴趣的结果和反应倾向高度相关的变量。这些协变量可以来自提供受访者和非受访者信息的辅助数据。两种这类辅助数据是采访者的观察(ParaData的一种形式)和关于小地区或家庭的商业数据。旨在用于无反应调整的面试者观察可以被专门设计成与感兴趣的结果变量相匹配，而商业数据提供了可能与多个结果相关的广泛的小区域描述符。这一分析检查了这两个数据源，以确定哪一个对特定调查的感兴趣结果更具预测性，从而满足良好调整变量的标准之一。这项分析的有趣结果是家庭收入的自我报告和从劳动力市场参与度调查中获得的失业救济金。研究结果表明，在这个时间点上，与商业数据相比，采访者的观察更能预测这些结果，特别是在调查目标人群中。因此，观测值与真实值共享更多(准确)的信息，使它们更适合在这个维度上进行调整。这一结果将为希望改善无反应调整的研究人员和希望更好地利用调查预算的调查管理人员的工作提供参考。
Survey methodologists are searching for covariates to use in nonresponse adjustment models, ultimately hoping to find variables that are highly correlated with both the outcomes of interest and the propensity to respond. These covariates can come from auxiliary data that provide information on both respondents and nonrespondents. Two such types of auxiliary data are interviewer observations (a form of paradata) and commercially available data on small areas or households. Interviewer observations intended for use in nonresponse adjustment can be specifically designed to match the outcome variables of interest, while commercial data provide a broad set of small area descriptors that may be correlated with multiple outcomes. This analysis examines these two data sources to determine which is more predictive of the outcomes of interest for a particular survey, thereby fulfilling one of the criteria for a good adjustment variable. The outcomes of interest in this analysis are self-reports of household income and receipt of unemployment benefits from a survey of labor market participation. The findings suggest that at this point in time, compared to commercial data, interviewer observations are better at predicting these outcomes, particularly in the subpopulation that the survey targets. Therefore, the observations share more (accurate) information with the true value, making them better for adjustment on this dimension. The results will inform the work of both researchers wishing to improve their nonresponse adjustments and survey managers looking to make better use of their survey budget.