Answers are grounded in this paper and cite their sources.
Research on attitudinal scales has shown that respondents without opinions who are requested to select an answer mostly choose middle scale values. This paper concerns design effects on central placement on the left-right scale across countries. We tested the effects in a split-half experiment in which the left-right scale was used either with or without an explicit “can’t choose” option. Open-ended questions on the meaning of left and right shed light on scale interpretation, in particular non-response to the open-ended question for those indicating a left-right position. Our results show that the likelihood of central placement depended on several aspects and that the best indicator was non-response to the probe questions, but some effects of scale design were not robust across countries.
No takes yet. Share an insight, caveat, or question.
Scholz et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: