|
Abstract:
|
The Division of Research and Methodology of the National Center for Health Statistics (NCHS) has been conducting a series of recruited web-based surveys, referred to as the Research and Development Survey (RANDS). While early rounds of RANDS were conducted exclusively via web administration, RANDS 4 (conducted in 2020) was one of the first RANDS surveys to incorporate telephone administration. This research evaluates how estimates of health-related outcomes in RANDS 4 differ between panelists responding by web and phone administration and what patterns, if any, can be identified between the two modes. We conduct descriptive analyses and tests for statistical differences to evaluate mode effects in RANDS 4, including comparisons of sociodemographic factors (e.g., age, sex, race) and common health outcomes (e.g., self-rate health status, asthma, diabetes) by mode. In addition, calibration is used to adjust the weights by raking and propensity score adjustment on the sociodemographic variables to study adjusted mode effects.
|
