This document presents the findings of the Informed Respondent Approach (IRA) to data collection implemented in an experimental way in Nepal. The participants were female interviewers from the 1996 Nepal Family Health Survey (NFHS), which is part of the worldwide Demographic and Health Surveys (DHS) program.
Interviewers are potentially an excellent source of qualitative information. In a typical DHS survey an interviewer completes, on average, interviews with 150 individual women from different socioeconomic statuses and from various parts of the country. She sits in each of their homes, observes their living conditions and records their responses to various questions. More importantly, interviewers are exposed to respondents' body language, tone, conviction or reluctance, or understanding, in answering questions, all of which go unrecorded in a standard survey with formatted questions and precoded responses. Furthermore, respondents often qualify their responses to simple yes/no of closed-ended questions with conversation which does not get recorded in a precoded questionnaire. Given their wide array of exposure, using interviewers as proxy respondents, therefore, will provide useful additional insights into what the population of women (or men) interviewed think about the subjects raised in the survey.