Report of Cognitive Testing of Privacy- and Confidentiality-Related Statements in Respondent Materials for the 2010 Decennial: Results from Cognitive Interview Pretesting with Volunteer Respondents
Ashley Landreth, Eleanor Gerber, and Theresa DeMaio
KEY WORDS: 2010 Census, cover letter, privacy messages, confidentiality messages, statistical purposes, administrative records
ABSTRACT
This report describes the development of privacy and confidentiality messages by an interdivisional working group led by the Statistical Research Division. It presents results from two rounds of cognitive testing of the new messages developed by this team. These messages were slated for inclusion in the decennial mailing package and included the main confidentiality statement and supporting concepts. Once developed, the intent was to include these messages in the 2010 Census mail materials. Concepts were also included that either modify or describe how the data will be treated and/or used. This included producing statistics and improving production efficiency. It also included how the data may be augmented by records housed within other federal agencies. The research found that most respondents expressed a general belief that the Census Bureau was trying to “do the right thing” by keeping detailed data on individuals out of the public sphere, but felt that nothing was foolproof. Concern over potential, however inadvertent, data “leaks” was evident. Response to the statement “Your survey answers will only be used to produce statistics and for no other purpose” was positive. Respondents found it most clearly conveyed that the data would be used to compare groups or averages for particular variables. They adequately and fairly consistently interpreted “statistics” in this context. The statement “Your answers are confidential. That means the Census Bureau cannot give out information that identifies you or your household” was also positively received. The latter statement was interpreted as a definition of confidentiality and tended to clarify the meaning of this concept within the context of the survey letter.
When information about administrative records use was located on the back of the letter, it tended to be better received than a shorter statement, placed on the front of the letter. The former placement allowed respondents to process the main confidentiality message on the front before reading the detailed explanations on the back. In general, locating information on the pack of the letter was acceptable. It was not perceived as an attempt to hide “the fine print.”
This report includes additional findings and recommendations for revisions to the mailed materials.
CITATION:
