Washington Statistical Society on Meetup   Washington Statistical Society on LinkedIn

Title: Adaptive Survey Design in CAPI Data Collection: The 2013 Census Test

  • Date/Time: June 12, 2014
    12:30 pm - 1:30 pm
  • Speaker:Gina Walejko, Ph.D., U.S. Census Bureau
  • Chair: Jonathan Mendelson, WSS Data Collection Methods Chair
  • Location: Bureau of Labor Statistics Conference Center
    To be placed on the seminar attendance list at the Bureau of Labor Statistics, you need to e-mail your name, affiliation, and seminar name to wss_seminar@bls.gov (underscore after 'wss') by noon at least 2 days in advance of the seminar, or call 202-691-7524 and leave a message. Bring a photo ID to the seminar. BLS is located at 2 Massachusetts Avenue, NE. Use the Red Line to Union Station.
  • Sponsor: WSS Data Collection Methods
  • POC e-mail: gina.k.walejko@census.gov
  • WebEx event address for attendees: https://dol.webex.com/dol/onstage/g.php?d=643498885&t=a
  • For audio: Call-in toll-free number (Verizon): 1-866-747-9048 (US) Call-in number (Verizon): 1-517-233-2139 (US) Attendee access code: 938 454 2
  • Note: Particular computer configurations might not be compatible with WebEx.
  • Download the slides from the talk.


In recent years, several organizations have implemented "adaptive" or "responsive" survey designs that use auxiliary data, paradata, and/or response data to tailor or alter contact approaches before or during data collection. In late 2013, the Census Bureau tested an adaptive design approach in a computer-assisted personal interviewing (CAPI) census data collection. In this talk, I describe the 2013 Census Test study design and discuss quantitative results and qualitative takeaways related to the implementation of an adaptive design in a census and CAPI data collection. I focus on findings related to propensity models, interviewer compliance, and interviewer productivity.

The test used two logistical models to generate cases' daily response propensity scores. These models relied on frame information as well as paradata generated in the field. Findings from model output and the implementation of assigning propensity scores show the importance of constructing models that can deal with limited contact attempt information at the beginning of data collection.

In the test, interviewers were trained to follow procedures including timely transmission of case information, recording of all contact attempts, and working specified cases on certain days. Specifically, training taught interviewers to work cases with the highest response propensities, "high priority cases," each day. Findings illustrate the difficulty of gaining interviewer compliance.

The test measured interviewer productivity both during and after data collection. Findings suggest that adaptive procedures increased interviewer productivity.