The basic procedure in survey is that people are asked a number of questions on that aspect of behavior which the sociologist is interested in. A number of people carefully selected so that their representation of their population being studied are asked to answer exactly the same question so that the replies to different categories of respondents may be examined for differences.One type of survey relies on contacting the respondents by letter and asking them to complete the questionnaire themselves before returning it. These are called Mail questionnaires. Sometimes questionnaires are not completed by individuals separately but by people in a group under the direct supervision of the research worker. A variation of the procedure can be that a trained interviewer asks the questions and records the responses on a schedule from each respondent.
These alternate procedures have different advantages and disadvantages. Mail questionnaires are relatively cheap and can be used to contact respondents who are scattered over a wide area. But at the same time the proportion of people who return questionnaires sent through post is usually rather small. The questions asked in main questionnaires have also to be very carefully worded in order to avoid ambiguity since the respondents cannot ask to have questions clarified for them. Using groups to complete questionnaires means that the return rate is good and that information is assembled quickly and fairly. Administrating the interview schedules to the respondents individually is probably the most reliable method. Several trained interviewers may be employed to contact specific individuals. The questionnaires and schedules can consist of both close-ended and open-ended questions. Also a special attention needs to be paid to ensure that the questionnaires are filled in logical order.
Where aptitude questions are included great care must be exercised to ensure the proper words are used. In case of schedules emphasis and interactions may also be standardized between different individuals and from respondents to respondents. Finally proper sampling techniques must be used to ensure that the sample under study represents the universe of study. In order to enhance the reliability of data collected through questionnaires and schedules, these questionnaires and schedules must be pretested through pilot studies.