The Census Bureau estimates that it takes respondents an average of 40 minutes per household to answer the 72 questions included on the American Community Survey (ACS). To gain feedback about the respondent experience, the Census Bureau conducted a survey of 1,000 ACS interviewers and learned that respondents perceive some questions to be difficult, sensitive or burdensome to answer. To address this concern and develop strategies for reducing respondent burden, the Census Bureau partnered with a broader set of federal data users, including the OMB Interagency Committee for the ACS, to recommend potential changes to high-burden questions. To evaluate these modifications, the Census Bureau contracted with Westat to conduct cognitive testing on the proposed changes to the following survey questions:

• Year the respondent’s house was first built
• Computer and smartphone ownership
• Type of Internet access
• Telephone service
• Year of naturalization and year of entry to the United States
• Address for place of residence
• Address for place of work
• Number of weeks worked in the past 52 weeks

During June of 2016, Westat conducted 72 cognitive interviews with individuals meeting specific characteristics of interest for each topic tested. This report summarizes the findings and recommendations from Westat’s research, including the recruitment criteria, methodology, sample questions, and recommended question changes.