The collection of self-reported survey data for people of all ages is central for monitoring the health and wellbeing of the population. For some experiences, people's self-reports are the only way to acquire the information, e.g., pain, fatigue, and life satisfaction. Given the importance of self-reports, they need to be measured as precisely as possible. There are many considerations for obtaining accurate and comparable self-reports. We examine several factors that are likely to vary by age and influence respondents'interpretations of survey questions and their use of response scales. Such factors could lead to erroneous conclusions about aging and health. A series of five studies, utilizing a range of research techniques from laboratory-based methods to repeated, real-time data collection to Internet panel surveys, is proposed. They will examine age-specific effects (1) in frame-of-reference (the comparison standards people use in making ratings), (2) in the impact of prior questions on responses (context effects), (3) in the effect of recall period on responses (where longer recall periods may be associated with more bias), and (4) in the use of response scales (e.g., tendency to avoid extreme responses). Based on the results of prior studies, the last study will explore ways to improve self- report measurement. Overall, the aim of this application is to identify factors that could impede our ability to understand age-related differences in health and wellbeing, and to develop ways to mitigate those problems.
Given the importance of self-reports in medical research and for evaluating governmental policies, it is vital that these data are of the highest quality. Erroneous conclusions about aging and health are possible if people of different ages interpret questions and use response scales differently. By addressing these issues, the proposed studies will contribute to more accurate self-report measurement and to improved understanding of the aging experience.
Showing the most recent 10 out of 11 publications
