Internet Interviewing and the HRS

我们要求资助一个为期五年的项目，对健康与退休研究（HRS）的一部分受访者进行两次互联网访谈，并建立一个由1,000名非HRS受访者组成的单独互联网小组，他们将在四年中每年通过互联网接受两次访谈，另外还有500名非HRS受访者组成的对照组，他们将在四年中每年接受一次电话访谈。数据的收集和分析将告知HRS网络访谈的潜力，并可作为HRS的测试样本。我们提出的研究包括方法论问题和实质性问题，首先是经济学、认知心理学和流行病学。鉴于目前互联网接入在人口中的分布情况（年龄较大的群体的接入情况有所下降），预计在可预见的未来，HRS将在使用其他模式的同时使用互联网访谈。这激发了本研究的混合模式设计。具体来说，我们的目标是：1。提供数据，以便对互联网受访者与电话受访者进行比较研究，以了解选择偏差；2. 与其他收集调查数据的方式相比，更系统地了解网络访谈的特性，特别是在中老年受访者中；3. 开展问卷和问题设计方面的各种实验，挖掘网络访谈的潜力。我们将特别关注认知功能的作用以及互联网作为面试环境的适宜性。4. 改进对消费或财富等各种重要客观变量的测量；5. 为可能主观的复杂变量（如概率、期望、健康史、风险态度和时间偏好）开发新的测量方法；6. 使这些活动产生的数据可供研究界使用。从2004年开始，将研究结果与部分和逐步实施网络访谈相结合。

We request funding for a five year proj4ect to conduct two Internet interviews with a subset of the respondents to the Health and Retirement Study (HRS), and to set up a separate Internet panel of 1,000 non-HRS respondents, who will be interviewed via Internet twice a year for four years, plus a control group of 500 non-HRS respondents, who will e interviewed once a year by telephone for four years. The data collection and analysis will inform the HRS about the potential of Internet interviewing and may serve as a tested sample for the HRS. Our proposed study covers both methodological issues and substantive issues, foremost in economics, cognitive psychology and epidemiology. In view of the current distribution of Internet access among the population (with access dropping of among older age groups), it is anticipated that in the foreseeable future the HRS will use Internet interviewing alongside other modes. This motivates the mixed mode design of the proposed study. Specifically, we aim: 1. To provide data that will permit a comparative study of the population of Internet interviewees with the population of telephone interviewees with the goal of understanding selection bias; 2. To provide more systematic insight in the properties of Internet interviewing in comparison with other modes of collecting survey data, in particular among middle aged and elderly respondents; 3. To initiate various experiments in questionnaire and question design, exploiting the potential of Internet interviewing. Special attention will be paid to the role of cognitive functioning and the suitability of the Internet as an interviewing environment. 4. To improve the measurement of various important objective variables such as consumption or wealth; 5. To develop new measurement methods for complex possibly subjective variables such as probabilities, expectations, health histories, risk attitudes and time preferences; 6. To make the data generated by these activities available to the research community. To gear the outcomes of the research towards partial and gradual implementation of Internet interviewing in the HRS as of 2004.