This paper investigates a person-based method of data collection in which a series of demographic questions is asked for each person in the household. Interviewers are instructed to read the questions as worded, using standardized interviewing techniques. However, in field observations it became apparent that the more people that were in the household, the less likely the interviewer was to continue reading the questions as scripted. We are interested in whether or not this practice leads to more respondent behavior problems. To address this, we behavior-coded a sample of interviews using a demographic questionnaire. We analyzed the interviewer/respondent interactions for the first person in the household and compared that to the interactions for the same questions about additional people in the household. The hypotheses were that when the questions were asked the first time, they would be asked in a more standardized way than when they were asked subsequent times and that deviating from the standardized wording would lead to more respondent behavior problems. We report differences in both interviewer and respondent behavior that stemmed from using a more conversational method of interviewing for later persons in the household.