How do you measure the public's awareness of an environmental hazard that has been known to be dangerous for a long time? This was one of the goals of the survey on lead-based paint which the Census Bureau was asked to design by the Department of Housing and Urban Development. This paper reports on the measurement strategies and pretesting methods we used to design such a questionnaire.
One difficulty with collecting "awareness" information is that respondents may give socially desirable answers to appear well-informed. Unless the questions regarding knowledge are structured to make them feel comfortable with saying "I don't know about that," some unknown portion will affirm knowledge they do not have.
Through cognitive interviews, we tested two methods of asking the same questions - one which asked if respondents agreed/disagreed/had no opinion about a topic and another which asked if they thought the topic was a problem or if they hadn't heard about it.
A revised set of questions was then tested in a large-scale field test. The results of the field test suggested that our strategy for collecting the information was acceptable. This seemed to be reiterated when the survey was conducted in December 1994 as a supplement to the Current Population Survey.