The Effects of Panel Conditioning on Data Quality

面板条件对数据质量的影响

• 批准号: 2050809
• 负责人: Ting Yan
• 金额: $ 67.5万
• 依托单位: Westat Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-15 至 2024-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2050809&HistoricalAwards=false
• 关键词: Effects; Panel; Conditioning; Data; Quality

This research project will advance scientific understanding of the mechanisms responsible for panel conditioning as well as the effects of panel conditioning on data quality. Panel conditioning refers to changes in reporting, attitudes, and behavior brought about by participating in a wave of a multi-wave panel survey. These changes are realized in responses at subsequent waves of the panel and introduce error in measures of change. Since panel surveys are primarily designed to measure change, panel conditioning is an important source of measurement error. Prior research largely has been based on observational data that could not clearly identify either the magnitude or the sources of panel conditioning. This project will address this limitation by administering a series of experimental studies to shed light on mechanisms underlying panel conditioning effects. The project will have broad impacts for social scientists who rely on data from panel surveys such as the Panel Survey of Income Dynamics, the American National Election Studies, and panel surveys administered by the Federal statistical system such as the Current Population Survey and the National Crime Victimization Survey. Results from this project will inform efforts to reduce panel conditioning effects and/or to adjust for them statistically. The project will promote teaching, training, and learning through courses offered in graduate programs, short courses, publications, and conference presentations.This project will conduct a series of experiments to address many of the open questions about panel conditioning. The experimental results will improve understanding about the causes of panel conditioning and provide insights as to when such effects should be expected to occur. Six experiments will be conducted, each examining a specific hypothesis about one type of conditioning with an eye to data quality. Experiment 1 will examine whether motivated misreporting or improved question understanding is the mechanism underlying decreasing reports of victimization over rounds of a survey. Experiment 2 will compare two formats (grouped versus interleafed) for organizing filter questions, examining how effective they are for reducing misreporting driven by a motivation to reduce burden. Experiment 3 will investigate whether attitudes become more consistent over time and what cognitive mechanisms might explain changes in consistency. Experiment 4 will examine how repeated questioning affect reports on potentially sensitive topics and what mechanisms underlie any observed changes in reporting. Experiment 5 will explore how different types of questions affect voting behavior and reports about voting, and whether asking respondents if they intend to vote increases the likelihood of voting. Experiment 6 will examine whether there are practice effects on reports about knowledge and personality with repeated interviews. This award is supported by the MMS Program and a consortium of Federal statistical agencies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

本研究项目将促进对面板调理机制的科学理解，以及面板调理对数据质量的影响。小组调节是指参与一波多波小组调查所带来的报告、态度和行为的变化。这些变化是在面板后续波的响应中实现的，并在测量变化时引入误差。由于小组调查的主要目的是测量变化，小组调节是测量误差的一个重要来源。先前的研究主要是基于观测数据，无法清楚地确定面板调节的大小或来源。本项目将通过管理一系列实验研究来阐明面板调节效应的机制，从而解决这一限制。该项目将对依赖小组调查数据的社会科学家产生广泛的影响，如收入动态小组调查、美国国家选举研究，以及由联邦统计系统管理的小组调查，如当前人口调查和国家犯罪受害调查。该项目的结果将为减少面板调节影响和/或对其进行统计调整提供信息。该项目将通过研究生课程、短期课程、出版物和会议演讲来促进教学、培训和学习。本项目将进行一系列的实验来解决许多关于面板调节的开放性问题。实验结果将提高对面板调节原因的理解，并提供有关何时应预期发生此类影响的见解。将进行六项实验，每项实验都考察一种条件作用的特定假设，并着眼于数据质量。实验1将检验动机性误报或问题理解的提高是否在调查过程中导致受害报告减少的机制。实验2将比较组织过滤问题的两种格式（分组与交叉），检查它们在减少由减轻负担的动机驱动的误报方面的有效性。实验3将调查态度是否随着时间的推移变得更加一致，以及什么样的认知机制可以解释这种一致性的变化。实验4将检验重复提问如何影响潜在敏感话题的报道，以及报道中任何观察到的变化背后的机制。实验5将探讨不同类型的问题如何影响投票行为和关于投票的报告，以及询问受访者是否打算投票是否会增加投票的可能性。实验6将通过重复访谈检验实践是否对知识和个性的报告有影响。该奖项由MMS项目和联邦统计机构联盟支持。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。