Adjusting for Effects of Survey Mode Difference Across a Longitudinal Mixed-Mode Study

• Activity Number: 445
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Heather Kitada* and Virginia Lesser and Lan Xue
• Companies: Oregon State University and Oregon State Univeristy and Oregon State University
• Keywords: Mixed-Mode ; Survey Error ; Longitudinal

Abstract

It is understood that survey modes have inherent biases that effect data quality. Mixed mode studies are often conducted to create a balance between these effects; however, when popular survey modes changed and new ones are created the modes used can differ. The effects of these differing modes confound the observations for a given year. For data that is collected over time in which the mode of data collection changes, it is necessary to tease the temporal effects of the variable of interest from the impact of mode. We will examine approaches to adjust data collected using old modes to what would be expected under a new mode of data collection.