Search for: All records

Survey research is at a crossroads. For at least half a century, survey data have been essential to government agencies, policy-makers, businesses, and academics across different fields to inform a wide range of critical decisions with far-reaching consequences. Even in an era of “big data,” surveys remain fundamental to understanding and shaping the economy, politics and governance, and society. Yet challenges to conducting high quality surveys are substantial and increasing. Face-to-face interviewing remains the gold standard of survey research, but the rising costs of such interviews are prohibitive. New technologies, techniques, and data sources present opportunities to improve the efficiency and speed of survey data collection and/or reduce its costs but have shortcomings that may exceed their advantages. To examine and develop strategies to address the challenges facing survey research, the Duke Initiative on Survey Methodology hosted a conference January 14th and 15th, 2021, on the Future of Survey Research. This report summarizes the proceedings and highlights key recommendations that resulted. 

Abstract Often, government agencies and survey organizations know the population counts or percentages for some of the variables in a survey. These may be available from auxiliary sources, for example administrative databases or other high-quality surveys. We present and illustrate a model-based framework for leveraging such auxiliary marginal information when handling unit and item nonresponse. We show how one can use the margins to specify different missingness mechanisms for each type of nonresponse. We use the framework to impute missing values in voter turnout in a subset of data from the US Current Population Survey. In doing so, we examine the sensitivity of results to different assumptions about the unit and item nonresponse.