How Meaning Is Given to Rating Scales: The Effects of Response Language on Category Activation

Abstract

Drawing from the categorical model of rating scale responses, a typology of rating scale formats based on scale response language is presented and the effects of the rating scale language on patterns of category activation and usage are examined using exemplar generation methodology. Bipolar rating scales which explicitly label two categories appear to activate both labeled categories. Respondents used each to correspond to approximately half of the provided rating scale intervals. In addition, an ad hoc category of exemplars irrelevant to either of the categories was created to provide meaning for the midpoint label. Unipolar scales, which explicitly only label one category, activated that explicitly labeled category. Respondents used this category to correspond to two-thirds of the scale intervals and a contrast category which was not explicitly labeled to correspond to the other third. Implications for rating scale selection and use are discussed.