The effect of question format on measured HIV/AIDS knowledge: detention center teens, high school students, and adults.

This study examines the effect of question format on HIV/AIDS knowledge assessed in teens in a detention center, public high school students, and adults. Multiple-choice items were taken from a Red Cross questionnaire and were transformed into open-ended and true/false/don't know formats. Each respondent received an open-ended and a structured version of the test (consisting of multiple-choice and true/false/don't know items). Format effects varied by group and order of presentation: High school students and adults performed better on the open-ended questions if they had answered the structured versions first-suggesting that the structured questions provided these respondents with unintended cues. Detention center youths did not benefit from having answered the structured items, and scored especially low on the open-ended questions. However, they did almost as well as the other groups with the true/false/don't know format. Implications are discussed for measuring HIV/AIDS knowledge and evaluating educational programs for different target audiences.