We compared direct (a survey) and indirect (mock petition drive) attitude measures regarding opening a group home for people with mental retardation in the neighborhood. We systematically assigned participants to one of three groups: positively worded petition statement, negative statement, or no petition. All participants answered survey questions about attitudes toward persons with disabilities. Order of the measures was counterbalanced in the first two groups. The point-biserial correlation between the petition response and survey was a moderate .40, suggesting that the survey was not a very reliable predictor of how people would react when asked whether they agreed or objected to a group home being opened in their neighborhood. Survey responses were influenced further by the wording of the petition drive statement.