The 1,000 women and 100 men who participated in the study received questionnaires based on the one used in the National Survey of Family Growth, and including items on abortion, sexual behavior, and illicit drug use. The experiment examined five variables: whether the questionnaire began with a series of medical questions or with questions on pregnancy; whether the interview was conducted by a nurse or field interviewer; whether the interview was done at the respondent's home or outside the home; whether the interviewer or respondent administered the questions; and whether the data were collected on paper or via computer. The variable with the most consistent effect was the method of administering the questions. Self-administration significantly increased the reported number of sexual partners, sexually transmitted diseases, and level of condom use. None of the variables affected reported drug use over the lifetime. Tables, references