The brief formal interview in which the working of the questions and the order in which they are asked is fixed is called structured interview while the freer discursive interview is called unstructured interview. The object of using structured interview is to standardize the interview as much as possible and thus to reduce the effect that the interviewer's personal approach or biases may have upon the result and even when structured interviews are used, proper training can do a lot to ensure further the reliability and validity of research. The personality of the interviewer and the social characteristics that the respondents attribute to him can be having influence on the result. The effort of interviewer's bias can be estimated by comparing one interviewer's result with other. The problem of interviewer's bias in an unstructured interview is much greater. Here the interviewer is left to his common devices as far as the way he approaches a respondent is concerned. There is no fixed list of questions to work through. Instead the interviewer may work from a guide that will remind him of the topics he wishes to cover.
The training of the interviewer is crucial here not simply training in the social skills of keeping the conversation going on a topic that the respondent may not be very interested in but also in acquiring sensitivity to those things his respondents tells him which are specially relevant to the theoretical topics he is pursuing. This means that unstructured interviews can be carried out by people trained in sociological theory. They are then able to size upon stray comments made by the respondents which can be developed and lead on to important theoretical insight.