Difficult wordings in VAAs. Predicting no-opinion and middle-response answers from question characteristics

Survey respondents have been shown to use both the middle-category response option and the no-opinion option to indicate that they do not understand a question. This mixing of the meaning of these response options is a validity problem, which is especially problematic for Voting Advice Applications (VAAs). In this type of political survey, users receive voting advice based on their answers to a list of political attitude questions. Kamoen and Holleman (2017) therefore predicted the occurrence of middle-response and no-opinion answers in VAA brand Kieskompas based on a set of question characteristics known to increase comprehension difficulties. The current study replicates that study for a different VAA brand (Stemwijzer) with a different type of response scale (a three-point scale instead of a five-point scale). In line with earlier findings, the choice for a middle-category answer or a no-opinion answer does not seem random: question characteristics associated with semantic meaning problems often result in no-opinion answers, whereas characteristics triggering pragmatic meaning problems are more frequently related to middle-category responses. This suggests that although both response options are used to express comprehension problems, there is – at the same time - a difference in their perceived meaning.