Also found in: Dictionary, Thesaurus, Medical, Legal, Financial, Acronyms, Wikipedia.
questionnairea form containing questions to be administered to a number of people mainly in order to obtain information and record opinions.
Social scientists use questionnaires to:
- examine the general characteristics of a population (e.g. age, sex, occupation, income, etc.);
- examine attitudes;
- establish the relationship between two variables (e.g. occupation and voting behaviour);
- test theories.
A number of problems exist in writing questions for questionnaires. Firstly, in wording the questions, care must be taken to try to ensure that the meaning which each respondent attaches to each question is the same. This means that when the questions are being written this should be done in relation to the target groups under study (see also PILOT STUDY). For example, in undertaking research with children the words used should be kept as simple as possible and long words avoided (e.g. ‘job’ rather than ‘occupation’, ‘mum’ rather than ‘mother’, etc.). A second problem is that of whether to use unstructured (open-ended) or structured (closed, or pre-coded) questions (see UNSTRUCTURED DATA and STRUCTURED CODING). The choice of which of these types of questions to use depends on the nature of the research topic and the means of administration. Where a relatively unexplored topic is being examined, open-ended questions might be preferred since the researcher may have little idea as to the range of possible replies. The same is often true where the questionnaire is being administered orally, since this enables the researcher to probe the replies given by respondents. Conversely, where postal or mail questionnaires are being used, precoded questions are generally preferred as this simplifies completion. A third problem concerns the sequence of the questions on the questionnaire. Generally, the questions should follow logically on from one another and they should be arranged in such a way that the order of the questions has as little effect as possible on how respondents answer subsequent questions. For example, when examining people's attitudes towards abortion a general question on whether abortion should be prohibited might be best placed before questions on possible reform of the abortion law. Finally, personal questions about age, sex, occupation and income, and possibly embarrassing questions should be placed at the end of the questionnaire.
Questionnaires can be administered in a number of ways:
- orally, by the researcher to the respondents in an interview situation;
- self-administered, where, for example, a teacher might give students questionnaires to be completed in the classroom;
- mail questionnaires, where they are sent to respondents through the post. The choice of how to administer a questionnaire depends on a number of factors, of which the nature of the research problems, the complexity of the questions, and cost are the most important. Where a relatively unexplored topic is being examined and many of the questions are open-ended, then interviews are preferred, and, conversely where a large number of people are being researched and the questionnaire has many precoded questions, mail questionnaires are preferred. Finally, it should be mentioned that although postal questionnaires are relatively cheap to administer they suffer from the disadvantage of a high nonresponse rate (see NONRESPONSE).