Does confidence in a wrong answer imply a misconception?

Abstract

Our study investigates whether confidence correlates with consistency in reasoning, specifically about radioactive decay. In prior work, we developed and tested a survey designed to measure consistency of student reasoning about radioactive decay by comparing responses to three prompts that are isomorphic, meaning that, despite having different surface features, they can all be answered appropriately with the understanding that radioactive decay occurs at random. In this paper, we compare (i) student patterns on these isomorphic prompts with (ii) confidence ratings that students provided together with their responses. Our research question is "to what extent does student confidence correlate with consistency in reasoning about radioactive decay?"We have found that there is no significant correlation, suggesting that more confident students are not more likely to be consistent. One reason why this finding is relevant is that the misconceptions model attributes consistency to student ideas (as opposed to the pieces model, which describes student ideas as potentially being context dependent). Our findings suggest that it is premature to describe a student idea as a misconception, even if the student is confident in that idea.