| |
An instrument used for the data collection segment of a survey analysis, comprising a carefully structured and ordered set of questions designed to obtain the needed information without either ambiguity or bias. Every respondent answers the same questions, asked in the same way and in the same sequence, which contrasts with the more open-ended formats used in interviews and other qualitative methods for obtaining information from individuals. The questionnaire maybe administered by a trained person, either in a face-to-face meeting or by telephone, or it may be self-administered.
Questionnaires are devised to obtain a variety of data. The simplest are the factual, ascertaining information such as age, place of birth etc. Others are attitudinal, for which questions are designed to probe people\'s values, attitudes and opinions. (These must be carefully pre-tested and piloted to ensure their validity.) Such data may be obtained through open-ended questions, with the \'free\' responses recorded and later categorized, but more common are specially designed scaling instruments for measuring different types of attitude (personality, political ideology, etc.): some are generally applicable but many are specific to a particular study, reflecting the local cultural situation and context. (Likert scales, repertory grid analysis, and semantic differential are scaling instruments widely used in the social sciences.)
Production of questionnaires is sometimes assumed to be a straightforward task and undertaken in a rather cavalier manner. For collecting all but the simplest of factual information, however, great care is needed to ensure the absence of ambiguity (recalling that the questions may be asked of people from a wide range of backgrounds), so that all respondents will interpret them in the same way. (If a respondent has to ask what a question means, then the questioner\'s answer will potentially introduce bias, especially if not all respondents ask.) With a self-administered questionnaire, the problems of varying interpretations of an unclear question are very substantial and may invalidate any analysis of the data obtained. (RJJ)
Suggested Reading Dixon, C.J. and Leach, B. 1976: Questionnaires and interviews in geographical research. Concepts and Techniques in Modern Geography 18. Norwich: Geo Books. |
|
