Assessing the Impact of Adopting a Mixed-Mode Design: Examining Respondent Skip Logic Errors in the 2015 Residential Energy Consumption Survey

• Activity Number: 128 - Impact of Data Collection Modes and Data Sources
• Type: Contributed
• Date/Time: Monday, July 31, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Kevin Adams* and Peter Frechtel and Amang Sukasih
• Companies: RTI International and RTI International and RTI International
• Keywords: mixed-mode ; skip logic errors ; skip patterns ; survey design ; Residential Energy Consumption Survey ; RECS

Abstract

The 2015 Residential Energy Consumption Survey (RECS) is a household level survey used to estimate energy consumption behaviors of U.S. residents. Initially, the 2015 RECS was administered solely using Computer-Assisted Personal Interviewing (CAPI). However, mid-way through the collection process, it became evident that relying on CAPI as the only collection method was not prudent, so a mixed-mode approach was adopted, adding Computer Assisted Web Interview (CAWI) and Paper and Pencil Interviewing (PAPI) alongside CAPI. This research discusses the impact a last-minute switch to a mixed-mode design may have had on the 2015 RECS. Specifically, we investigated to what existent complex skip patterns increased skip logic error rates. This paper also considers how the layout of the PAPI instrument may have played a role. The findings and opinions of this paper are solely that of the authors and do not necessarily reflect the beliefs of the U.S. Energy Information Administration.