Examining Variation in Survey Costs Across Surveys

• Activity Number: 185 - The Missing Puzzle Piece: Estimating Survey Data Collection Costs
• Type: Invited
• Date/Time: Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Kristen Olson* and Jennifer Dykema and John Stevenson and Lindsey Witt-Swanson
• Companies: University of Nebraska-Lincoln and University of Wisconsin Survey Center and University of Wisconsin Survey Center and Bureau of Sociological Research
• Keywords: survey costs; mixed mode surveys; self-administered surveys

Abstract

Survey methodologists are often tasked with designing data collection protocols that maximize response and minimize costs, including recommendations about whether self-administered surveys should be conducted with a single or mixed data collection mode. Although mail-web mixed-mode surveys increase response rates relative to web-only surveys and yield response rates similar to mail-only surveys, whether reductions in fixed and variable costs are actually obtained by mixing modes is less well understood. Cost structures likely vary across design features including target populations, sampling frames, concurrent versus sequential mixed-mode designs, incentive structures, sample sizes, response rates to the initial mode and subsequent contact attempts, and number and type of contacts. However, the relationship between design features and survey costs is understudied. We examine survey costs across multiple mail and web single- and mixed-mode surveys conducted by two academic survey centers between January 1, 2018 and December 31, 2019. We examine how costs for single- and mixed-mode surveys vary across design features and for what design features are mixed-mode surveys cost effective.