In March, the Committee on National Statistics (CNSTAT) of the National Academies of Sciences, Engineering, and Medicine held a “Workshop on Respondent Burden in the American Community Survey,” which brought together experts from the Census Bureau and the broader statistical community to discuss how to make the American Community Survey (ACS) a more pleasant, less intrusive experience for respondents.
Dimensions of Respondent Burden
The first session focused on “Understanding Respondent Burden.” Joe Solvio of the New York City Department of City Planning argued that if the Census Bureau does not find a way to change public perception of the ACS, it could severely damage the Bureau and limit the information they are able to collect. He proposed four changes that could make the survey less threatening: matrix sampling (giving respondents subsets of questions to reduce the survey’s length); changing the strategy (making it friendlier); communication and education; and “branding” the survey.
Deborah Stempowski, Chief of the American Community Survey Office at the Census Bureau, explained that the main concern expressed by respondents is that the survey questions seem intrusive; the average person filling out the survey doesn’t understand their purpose. In an effort to make the process friendlier and more transparent, the ACS office has hired a respondent advocate to directly address citizens’ concerns. They have also reduced Computer-Assisted Telephone Interviewing (CATI) contact attempts, begun including an explanatory brochure with the questionnaire packet, increased staff training, and drastically changed their mailing strategy.
Scott Fricker of the Bureau of Labor Statistics further expanded on what Census is doing to lessen respondent burden. Everyone asked to participate in the ACS feels this stress differently, and it is important to measure and accurately define what that really means. Part of the ACS survey now asks respondents to give their opinion on the survey and share how it made them feel. Understanding the different emotional reactions that people have to this process is important, and the information gathered can help the agency continue to make positive reforms.
Survey Materials and Respondent Burden
A session titled “Communicating with Respondents: Materials and the Sequencing of those Materials”
featured Don Dillman of Washington State University; Nancy Mathiowets of the University of Wisconsin-Milwaukee; Andy Peytchev of the University of Michigan; and Andrew Reamer of the George Washington University. The panelists discussed mailing strategy and adjustments that could change public perception of the survey. In the past, the ACS’s mailing strategy was sometimes seen as aggressive, with frequent mailings and a perceived tone of negativity that could cause respondents to lose trust in the legitimacy of the survey. Having a sound mailing strategy is important as mail is less intrusive than phone or in-person contact, and streamlining the process can not only save money, but improve respondents’ perceptions of the survey.
While it may seem small, changes made by the Bureau to envelope design have had a profound positive effect. The envelope has been streamlined, with less information provided on the front. The old envelope contained sometimes unnecessary information, for example, the fact that the Census Bureau is an equal opportunity employer. By streamlining the envelope and only providing pertinent information on the front, people have already begun to have a more positive reaction to the ACS. The Bureau’s new “Why We Ask” pamphlet, which is now being included with questionnaires, could still be described as being “in beta” and could also be further streamlined and improved.
The panel raised some concerns about continuing to use the term “respondent burden,” as it makes the survey easier to attack, and the term has been embraced by detractors when criticizing the survey. The Bureau hopes to establish responding to the ACS as a social norm, similar to jury duty—something that is common and necessary to keep our society functioning in a healthy manner.
Communicating the Value of the ACS to the Public
A session entitled “The American Community Survey: Communicating the Importance to the American Public,” included Sandra Bauman of Bauman Research and Consulting, Betty Lo of Nielsen Research, and George Terhanian of the NPD Group. They discussed how to better brand the ACS and effectively communicate its importance to the American public. While branding is mostly associated with the commercial world, tenets of branding could be effectively used to create interest, awareness, and trust in the ACS. The biggest hurdle the ACS faces is a lack of understanding by the public, so learning how to better market and communicate the purpose and benefits of the survey would help make it more accepted and effective. The panelists cited the Neilson Corporation as an example of a company that has learned how to get people to enthusiastically participate in surveys, thanks to their focus on marketing and driving brand awareness and familiarity. Learning how to market the ACS as a benefit and not a burden is critical to building trust and overcoming concerns about data privacy.
COSSA’s intern, Spencer Bailey from the University of Utah, contributed to this report.