Careful consideration should be given to survey staff selection. Interviewers are the interface between the survey organisation and the respondent. The impact an interviewer can have on a survey response rate or the item response rate should not be underestimated. A high response rate promotes confidence in survey results.

Supervisors and interviewers should be good communicators, friendly, flexible, literate, and reasonably numerate. The social distance between interviewer and respondent should be minimised. Interviewers should be the same sex, of similar age, ethnic group, social background and speak the same language or dialect as respondents.

In harder to reach populations, (eg. injecting drug users, street children, or sex workers), training members of that population group to be peer interviewers should be considered. Because of shared experience, peer interviewers can build rapport and reduce respondent’s reluctance to disclose sensitive or covert information.

Interviewers should have a full understanding of what the survey is about; what the respondents are being asked, why and for who, and what survey procedures are.

Training should emphasise:

• establishing contact with respondents and maintaining co-operation;
• role play and practicing the questionnaire with respondents (including the use of computers if needed);
• ethical considerations;
• a clear and thorough understanding of survey objectives;
• and how the data collected will serve survey objectives.

Training should also cover administrative tasks, such as returning work to supervisors, keeping track of questionnaires, and submitting expense forms. More staff should be trained than needed, so that the best may be selected for the job, but also for back up because interviewer turnover can be high.

Training manuals are a crucial component in training of survey staff, and should serve as a reference throughout the survey. They should explain the reason for the survey, detail the role of each category of staff involved in implementation, and provide a clear practical guide for all aspects of the survey.

Manuals should explain how to complete the questionnaires, how to convert verbal responses into pre-coded responses, define concepts and terms, explain how to contact respondents and conduct the interviews, as well as what to do in more unusual situations. For supervisors there should be clear guidance on how often study sites should be visited, whether there should be unscheduled visits and what to do if staff consistently underperform. The system of returning completed forms and passing them to data entry clerks should be clarified. For large scale surveys manuals should be developed separately for interviewers, supervisors, and data entry clerks.

Example: http://www.measuredhs.com/What-We-Do/Survey-Types/KIS.cfm

Click on right hand box, guidance manual for KIS. KIS-3 Interviewers Manual.