How do you know you’re right? How student’s perceive their confidence in Chemistry

Title: ‘‘I guess it was more than just my general knowledge of chemistry’’: exploring students’ confidence judgments in two-tiered assessments

Authors: Casandra Koevoets-Beach, Karen Julian and Morgan Balabanoff

First author’s institution: University of Louisville

Cover Image Credit: Giulia Forsythe

            Metacognitive skills refer to the ability to think about and evaluate your own thinking. For introductory science students, it’s a useful skill because it helps build the ability to evaluate how well they understand new and often challenging material. However, exams typically test how well students know the material and don’t give students much practice evaluating their own understanding.

            One strategy for helping students build metacognitive skills is to  ask them how confident they feel about their answers to exam questions. A recent report in Chemical Education Research and Practice described the administration of a multiple choice test to 1638 students where they were asked multiple choice questions about material covered in a General Chemistry class. For each question, they were then asked to rate how confident they were in their answer on a scale of 1 to 5. The researchers then did interviews with 22 students and investigated how they evaluated their confidence.

            When discussing questions on which they were confident, but ultimately got the wrong answer, students often stated they had felt confident because they reached the answer quickly. Researchers found that often those students held prior conceptions about the concepts involved that they applied incorrectly. but quickly, leading to their high confidence. Alternatively, students who confidently arrived at the correct answer, often cited a mastery of the concepts underlying the questions.

            Students sometimes answered the questions correctly despite expressing low confidence in their answers. Some educators have suggested this could be a sign that students are simply guessing and happening upon the correct answer. However, interviews revealed that many low-confidence but correct students were assessing their confidence based on how successful they felt in the course relative to their peers. Alternatively, some felt unconfident because they were thinking about prior experiences or assessments where they had struggled. Some of these students also seemed to be uncertain because of the test format – they would describe narrowing it down to two multiple choice answers and then becoming stuck, but would be able to describe a concept verbally.

            Students who answered incorrectly and rated their confidence as low displayed metacognitive accuracy. However, these students who often had affective factors impacting their test-taking. They often expressed an inability to eliminate some multiple choice answers or a feeling it took them a long time to get to the answer. There is a concern that the confidence questions, by drawing students attention to these negative perceptions of their performance, may not have the intended effect for these students.

	Student Answers Correctly	Student Answers Incorrectly
Student feels confident in answer	Students often cite content mastery as the cause of their high confidence	Students often feel confident because they reached their answer quickly
Student does not feel confident in answer	Students often cite comparison to peers or prior experiences as reasons for their low confidence-ratings.	Students assessed their knowledge accurately, but often have negative self-judgements or feel less competent than peers that interfere with their learning

            The researchers suggest that educators or researchers who want to use confidence-assessing questions on assessments consider structuring the questions so they lead students to think about their own understanding of the material rather than their understanding relative to their peers or environmental factors. Previous work has found confidence-questions work well at improving student metacognition when used with in-class clicker questions. The authors recommend this kind of low-stakes formative assessment a potential use-case for confidence questions. In this setting, they allow the instructor to give students immediate feedback on their responses and allows for transparency about the purpose of these questions in helping students build metacognitive skills.