We hypothesized that differences between input techniques depend on whether an interviewer asks the questions and handles the computer or the interview is conducted as a self-administered interview. We therefore used this interview mode as a second factor and tested if there was a difference in the number of answers between self-administered interviews and interviews conducted by interviewers. We used interviewers who had high typing skills because we thought that the difference between paper-and-pencil interviews and keyboard interviews would disappear when people keying in the data have high typing skills