A Comparison of Three Forms of the Semantic Differential

Rating scales are widely used by researchers to measure people's attitudes to a variety of stimuli, yet little time is spent examining respondents ' reactions to the form of the scales used for this purpose. Three forms of the semantic differential were tested to determine whether labelling the scale points affects the way that the scales are used, and to determine which form respondents prefer to use. The scale points were either unlabelled, labelled, or numbered. No significant differences were found in the ratings that were obtained with each form, but participants clearly preferred to use the labelled form. It is suggested that this is the form of the semantic differential to use when surveying a diverse audience; the other forms tested may be more suitable for specialist audiences