Deborah H. Griffin, U.S. Census Bureau Improving Operational Efficiencies and Survey Management of the American Community Survey in the United States Presented at the conference of the European Survey Research Association in Ljubljana, Slovenia July 2013 Deborah H. Griffin, U.S. Census Bureau
Outline Motivation for Change Census Bureau Initiatives Background on the American Community Survey Improving ACS Operational Efficiencies Improving Survey Management in the ACS
Motivation for Change Increasing costs and challenges of surveying the population Availability of paradata and other auxiliary data sources Need to reduce respondent burden
Motivation for Change Is there a solution that could leverage frame data, paradata, and auxiliary data to reduce costs and burden while maintaining quality?
Census Bureau Initiatives Contact History Instrument (CHI) Unified Tracking System Multimode Operational Control System Adaptive Design 2/16/2019
The American Community Survey Primary source of detailed demographic, social, economic, and housing data for large and small communities across the United States and Puerto Rico Continuous data collection broken into monthly samples that use four modes of data collection 2/16/2019
The American Community Survey Data Collection Sample Panel Calendar Month Mar 2013 April 2013 May 2013 Jan 2013 Feb 2013 Internet/Mail In-person Telephone In-person Telephone In-person Telephone 7 7
Improving ACS Operational Efficiencies Building a New Control System ACS Control System tracks and controls each monthly sample through all modes of data collection ACS will provide an important foundation for the creation of an enterprise system Critical transition
Incorporating Adaptive Design Methods Messaging Tailored messaging in initial contact letters and other mail materials 2010 Census targeted advertising campaign demonstrated the potential of tailored messaging in the ACS (Baumgardner, 2012)
Incorporating Adaptive Design Methods Messaging Source: Baumgardner, 2012
Incorporating Adaptive Design Methods Initial mode assignment Assigning each sample address to the initial mode that is most likely to result in a response Improving operational efficiencies by using auxiliary information to improve assignments and provide assistance
Incorporating Adaptive Design Methods Switching and Stopping Rules Use of full contact history paradata and other auxiliary information to predict response propensity Cost, quality, and burden trade-offs Reducing burden and costs in telephone follow up (Griffin and Hughes, 2013)
Incorporating Adaptive Design Methods Switching and Stopping Rules Source: Griffin and Hughes, 2013
Incorporating Adaptive Design Methods Switching and Stopping Rules Source: Griffin, forthcoming
Improving Survey Management Unified Tracking System Source: U.S. Census Bureau, Unified Tracking System reports, 2013
Improving Survey Management Baselines Documenting quality, burden and cost baselines Reviewing results to identify geographic and management areas with examples of operational inefficiencies or quality concerns Quality metrics must go beyond response rates
Conclusions ACS is changing the way it collects data to reduce survey costs ACS is prioritizing research to reduce respondent burden Research is on-going to develop best tools to balance costs, quality and burden 2/16/2019
Contact Information Deborah.H.Griffin@census.gov Any views expressed are those of the author and not necessarily those of the U.S. Census Bureau. 2/16/2019