The Impact of Readability Level on Questionnaire Measures Reliability

Abstract

Organisational research continues to rely almost exclusively on self reported attitudinal measures as a means for collecting empirical data. The primary instruments used for data collection are generally in questionnaire format (Salancik, 1979). Often this data is retrospective, as respondents are asked to remember past experiences or situations and to record their observations. As indicated by Fisher (1986), these recollections often involve attitudes and beliefs which are difficult to verify. Furthermore, the use of a questionnaire, although advantageous in terms of reliability, comparability, relative ease of analysis, and ability to obtain data from relatively large samples (Fisher, 1986), also may create several epistemological problems (Salancik, 1979). The validity of the data can depend on how well the subjects relate to the questions or statements used. As suggested by Stone (1978), “questionnaires cannot be used with illiterates or with individuals who have reading problems. To the extent that individuals cannot cope with the reading demands placed upon them by questionnaires, the researcher can expect one or more of the following problems; low return rates, missing data, and random responses” (1978:64). Furthermore, the richness of the response can also be impacted by the extent to which the questionnaire relates to the characteristics of the respondent (Alderfer and Brown, 1972). Questionnaires and scales used to obtain data in organisational research are generally reliable, cost effective, easily analysable, and the most frequently used form. Furthermore, they are often called into question in terms of response validity and that the more they reflect the nature of the respondents being assessed, the more likely they are to provide accurate data.