Propensity Score Weight Adjustment for Dual-Mode or Dual-Frame Longitudinal Surveys

• Activity Number: 477
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Section on Survey Research Methods
• Author(s): Christian Boudreau*+ and Mary E Thompson and Dagmar Mariaca Hajducek
• Companies: University of Waterloo and University of Waterloo and University of Waterloo
• Keywords: propensity score adjustment ; dual-frame ; dual-mode ; post-survey weighting ; longitudinal survey

Abstract

Dual-mode surveys are use to increase response rates and/or reduce costs. Similarly, dual-frame surveys can yield substantial cost savings when sampling rare populations, or when different survey modes are used in each frame and one mode is much cheaper than the other. In longitudinal surveys however, respondents with a given set of characteristics might experience both higher attrition and greater propensity to answer via one of the two modes; thus increasing undercoverage of these respondents as time goes on. Similarly, in dual-frame surveys, one frame might be incomplete, bias, suffer from undercoverage, or have higher nonresponse than the other. Our proposed post-survey weight adjustment method utilises propensity scores to adjust the weights of respondents from the mode/frame suffering from undercoverage/bias by making their probability distribution closer to that of respondents in the other mode/frame.