Learning From Errors in the Math Class: Associations With Students' Affective-Motivational Self-Regulation, Attributions, and Adaptive Learning Actions

• In Event: Being Wrong (Sometimes): How Students’ Classroom Assessment Mistakes and Learning Errors Enhance Learning and Motivation

Abstract

Contemporary research provides evidence that (feedback on) errors and formative classroom assessment contribute to individual learning and achievement (e.g., Mathan & Koedinger, 2005; McMillan, 2004; Oser & Spychiger, 2005). However, in order to think about and to modify the underlying misconceptions or knowledge gaps, effective learning from one’s errors requires adaptive affective-motivational adaptive self-regulation processes (Author, 2016). In addition, adaptive learning actions and metacognitive activities specifically adjusted to the error have been shown to be associated with students’ effort and engagement (Author, 2013). Research findings indicate that academic self-concept, mastery goals, and internal-unstable attributions facilitate adaptive reactions to errors (e.g., Author, 2017), but there is less evidence for the impact of adaptive error reactions on learning and achievement—specifically with a focus on differential effects of affective-motivational and behavioral reactions. It was assumed that, following error feedback, affective-motivational adaptivity is positively associated with conducive changes of task-enjoyment and success-expectations, whereas action adaptivity should be related to a reduced number of errors in similar tasks.

The current study was designed to test these hypotheses in an ecologically valid setting in the domain of mathematics. Fifty-six seventh grade students (54% female, mean age = 13 years) from three classes participated in the study which was integrated in their regular, everyday math classes. Students enjoyment and expectations for success were assessed on-task (1 = not at all true to 6 = very true) before individual seatwork. Immediately after this practice phase, students were provided with written error feedback followed by an assessment of their affective-motivational adaptivity and action adaptivity of error reactions, causal dimensions of error attributions, task-enjoyment, and expectations for success for a subsequent testing phase. The number of errors in the practice tasks (range = [2; 14]) and in the testing phase with similar tasks (range = [0; 13]) was recorded. In the testing phase, students made less errors than before (t(55) = 3.428, p = .001). In addition to students’ performance in the practice phase (β = .61, p < .001), adaptive learning actions negatively predicted the number of errors in the testing phase (β = -.21, p = .04). Also in line with our hypotheses, students’ affective-motivational adaptivity predicted subsequent task-enjoyment and success-expectations— both analyses controlled for previous task-enjoyment and expectations for success. Bivariate correlations underpin our assumption that students with more adaptive affective-motivational reactions to errors report less maladaptive changes (differential values) in task-enjoyment and success-expectations, and less stable but more controllable attributions.

The results emphasize the importance of affective-motivational self-regulation and adaptive learning actions following errors and their differential effects on learning from errors, as well as the benefits of attributing one’s errors to unstable, controllable causes.

Authors

• Maria Tulis, University of Salzburg

• Robert Grassinger, University of Augsburg

• Martin Daumiller, University of Augsburg