The problems of wording survey questions: sociological-linguistic analysis

Abstract

The critical part of preparation for conducting a mass survey is the proper wording of questions that allows revealing and assessing the preferences, experience and motives of respondents’ behavior. Despite the importance of adequate sampling of respondents and use of effective processing techniques of the acquired information, the result of research would be inaccurate without properly worded and comprehensible questions. The choice of words and putting them together is crucial for understanding of the question and its interpretation by a respondent; even small differences in wording may significantly affect the results. This article is an attempt of carrying out a sociological-linguistic analysis of the problems associated with wording of questions for mass surveys on the relevant political topics. The analysis of survey questions used in public opinion polling determined a number of important factors that influence the veracity and accuracy of research results. The detected problems are multifaceted and reflect different aspects of working survey questions. The analysis of structural aspect demonstrated the significance of proper choice of the type of question, either open-ended or close-ended. The suitability of one or another type is first and foremost defined by the content of question, objectives and tasks of research, and characteristics of respondents. Examination of the spoken form of questions (linguistic aspect) emphasized the importance of selection of linguistic units for comprehensible wording. The paramount role is played by lexical and syntactic linguistic units. The results of semantic analysis of the wording of survey questions is also important. The key problems within the framework of this aspect are associated with imposition of certain answers using imbalanced or suggestive questions, as well as various negative effects that appear as a result of particular placement of questions in a survey or creation a necessary context. All of the aforementioned problems significantly affect the quality of the conducted surveys.