Careful supervision of the field work is necessary if accurate data to be collected. The supervision problems are much the same for telephone and for personal interviews, but these problems can be managed better and more easily telephone projects.
Control: Monitoring equipment permits a small number of specialized supervisors to listen to ongoing telephone interviews and to provide any needed corrective action as soon as the interview is over. The sample is identified by telephone numbers. The status of each number is maintained in the computer, which generates a daily report showing interviews completed, the number left to do, response rates, and work completed by each interviewers.
Quality control by the supervisor usually is based on verification that the sampling procedure is being followed insofar as telephone numbers addresses, names, and demographic characteristics can show this. Questionnaires are examined to see if all appropriate questions have been completed, if questions have been omitted when called for by respondent answers, if unsatisfactory or incomplete answers have been accepted without sufficient probing and if writing is legible.
Supervisors keep daily records on the number of calls, number of not-at-homes, numbers of refusals and number of completed interviews for each interviewer and the total for all interviewers under their control. In addition, they record hours worked. These permit them to determine the cost per completed interview, whether the job is moving on schedule and which interviewers if any are having trouble.
For personal interview studies geographically larger than one metropolitan area, local supervisors are necessary. In most instances they work for more than one research organization. It is important that they close contact with the central research office, that they be trained for work with the organization, that they receive periodic visits from representatives of the central organization to review their work and that they have written instructions and forms for each specific job.
Local supervisors have personal contact with the field force under their supervision. It is they who, typically receive the completed questionnaires daily and check them to insure that instructions are being followed and that the quality is good. They maintain the same types of daily records as to telephone supervisors but to the case of door-to-door interviewers there may be many supervisors spread throughout the country. This requires a second level of supervision to coordinate the local supervisors and to provide a regular progress report at a central point.
Detection of cheating: A special aspect of quality control is that of validation or detection of cheating. Since interviewers work independently it is not always easy to detect cheating or to distinguish cheating from honest misunderstandings as to procedure. All well run field surveys check a sample of each interviewer’s work to see if it represents actual interviewers with the individuals shown. The close supervision provided in centralized telephone studies makes cheating a very small problem, but checks are typically made anyway. In personal interview surveys local supervisors do this by phoning sample of respondents and asking if they were interviewed. When interviewers are spread over too wide an area for this to be feasible, postcards are mailed asking acknowledgment of the interview. This is usually done by saying the interview has been reported and asking if the interviewer was polite and efficient. An opportunity to report that no interview took place is provided. The number of such postcards returned is usually so low that reports of no interview can only be taken as suggestion for further investigation.