When data are collected personally at more dispersed locations, such as at households, the fieldwork problem becomes more difficult. The research director must develop job specifications for the specific project, decide what characteristics the field workers should have, and try to find interviewers with those characteristics. Since the sample design often calls for data to be collected at many different geographic locations, one or more field workers with the desired characteristics must be found at each of these places. Most field work organizations keep a file of such workers by geographical location. The sampling sites are matched against these files and qualified workers selected. Letters are sent to the selected interviewers telling them about the pending work and asking them to indicate by return mail their availability. Usually it is necessary to recruit some new workers for each job, since qualified interviewers may not be available at all locations, and some may not be available for the specific job. This recruiting is often done with the help of such local sources as newspapers and educational institutions.
After field workers are selected they have to be trained. Since most field work must be done within a limited time period and interviewers are widely dispersed, training is usually done by mail. Under such conditions the training program consists of written instructions that the interviewer is asked to study carefully to learn the purpose of the study, how to locate and approach respondents, to establish rapport, to ask questions, and to obtain and record accurate answers. On some projects the training is given by supervisors in person at one a few central locations.
Following the training program, the interviewers commence the field work. During this time interviewers need to be supervised, at least to the extent of ensuring they are proceeding on schedule and that their work is satisfactory. Since factors as sickness or bad weather may prevent some interviewers from completing their assignments on time, the director needs to keep tight control on the day-to-day field operations must be ready to replace individual interviewers quickly, if necessary.
After the fieldwork is completed and the completed forms are returned to the home office, a verification check is made to be certain that the interviews were actually made to ensure interviewers did not cheat. The work of each interviewer is then evaluated. The questionnaires or other data forms turned in are checked for completeness, compliance with instructions, and apparent ability of the worker to obtain useful data. Such information helps the research director to select the best field workers for future projects.
As the above indicates, it is difficult and therefore expensive to do good quality field work in door to door interviewing. Security and privacy issues further complicate the problem. Two career families have made evening interviewing necessary,, but there are sections of the city where interviewers are unwilling to go at night. In low income areas respondent cooperation is hard to secure because of privacy feelings. The result has been a steady decline in door-to-door interviewing and an increase in the use of the telephone as well as in interviewing individuals in shopping centers. In some cases the first questions will be to screen the respondents to identify individuals who fit a particular specification (e.g. recent purchaser of personal computer or owner of an automatic coffeemaker). These individuals may then be asked to visit the nearby “store” run by the research organization to answer further questions in more depth. This shopping center interview procedure permits a concentration of fieldwork and so permits better control and reduction of costs.