I am looking for any papers or analysis on the use of a carousel vs. a matrix layout for a likert-style study. I have the broad impression a carousel is used for images or certain studies where each item is independent, and a question like how much the subject thinks/agrees on some topic. I am interested in understanding the use of this format for ratings of words (e.g. how well do you understand this topic 1-5).
Intuitively (perhaps wrongly), I suspect the carousel would be better than a matrix layout for avoiding influence from surrounding words when evaluating the answer choice. It also seems easier (less cognitive load) than looking at a massive list of words (>50). It would be nice to have a source for these though.
On the other hand, in the carousel view, some pilot I did received feedback that people felt their initial answers were off because they hadn't 'calibrated' themselves to understand the scope of what words there would be. This sounds like a learning effect, but also one that is learning to be more accurate in the answers?
I have not been able to find much on this topic, but it could also be the keywords I am using for search. I also searched simply for 'carousel' on this SE, and seemingly no results.
e.g. I have found
Evidence-Based Survey Design: ... series by SY Chyung. and
Likert-based vs. semantic differential-based scorings of positive psychological constructs: A psychometric comparison of two versions of a scale measuring resilience