VERBAL BEHAVIORS IN COMPUTERIZED LIFECOURSE SURVEYS

描述（由申请人提供）：旨在为生命过程幸福的决定因素建模的社会科学调查往往依赖于回顾性报告的质量，以准确估计关键社会、经济和健康指标的影响。最近开发的事件历史日历（EHC）方法已被证明与传统的问题列表（Q-list）方法（使用纸笔和计算机辅助访谈模式）相比，可以提高回顾性报告的质量。然而，产生这种数据质量改进的确切访谈特征仍未明确。该项目意义重大，因为它将通过使用言语行为编码技术来确定与增强的生命历程回顾性回忆相关的特定访谈者-被调查者互动行为，从而解决这一知识差距。具体而言，本项目将1)确定由计算机辅助版本的EHC和Q-list方法产生的突出检索和交流模式。这将通过口头行为编码和586个录音访谈的会话分析来评估，这些录音访谈是在一项研究中获得的，该研究比较了EHC和Q-list访谈技术在获得终身反应框架的高质量回顾性报告方面的有效性。2)使用来自前几波小组调查的验证记录，本项目将确定检索和交流模式，这些模式与每个各自方法的受访者一生回顾报告中更高的完整性和准确性相关联。3)该项目将确定检索和交流模式在促进回忆方面最有效的参考时期（较近的事件与较远的事件）的长度。4)本项目将通过确定最佳记忆线索和会话特征，将其纳入利用回顾性回忆措施的健康相关调查，从而为调查方法和认知科学文献做出贡献。最终，这些知识可以改善国家卫生政策。

DESCRIPTION (provided by applicant): Social science surveys aimed at modeling the determinants of life course well-being often depend on the quality of retrospective reports to accurately estimate the impact of key social, economic, and health indicators. Recently developed Event History Calendar (EHC) methodologies have been shown to improve the quality of retrospective reports in comparison to traditional question-list (Q-list) approaches using both paper and pencil and computer assisted interviewing modes. However, the precise interviewing features that produce this improvement in data quality remain unspecified. This project is significant because it will address this gap in knowledge by using verbal behavior coding techniques to identify the specific interviewer-respondent interaction behaviors associated with enhanced life course retrospective recall. Specifically, this project will 1) identify the prominent retrieval and communicative patterns engendered by computer-assisted versions of the EHC and Q-list methodologies. This will be assessed through verbal behavior coding and conversational analyses of 586 audio taped interviews that were obtained in a study that compared the efficacy of EHC and Q-list interviewing techniques in terms of their ability to obtain quality retrospective reports over a lifetime response frame. 2) Using validation records from previous waves of a panel survey, this project will determine the retrieval and communicative patterns that are associated with greater completeness and accuracy in respondents' lifetime retrospective reports for each of the respective methodologies. 3) This project will determine the length of the reference periods (more recent vs. distant events) for which retrieval and communicative patterns are most effective in facilitating recall. 4) This project will contribute to the survey methodology and cognitive science literature by identifying the optimal memory cueing and conversational features to be included in health-related surveys that utilize retrospective recall measures. Ultimately, such knowledge can improve national health policy.