Poster #23 - Accuracy and Stability of Self-Evaluations and -Regulation in Elementary School: A Short-Term Longitudinal Study

• In Event: 3-053 - Poster Session 10
  In Poster Session: PS 10 Section - Cognitive Processes

Integrative Statement

Children spend 30% of the time they are awake in school, where they learn new information (often from text materials), and acquire competences how to learn. Two such competences are self-evaluation and self-regulation. Self-evaluations affect self-regulation and performance. That is, to effectively learn and perform well, students need to be able to evaluate their own task progress and outcomes, in order to regulate their learning (e.g., allocating study time to tasks that are not yet well learned, or withholding responses when they think these are not correct). Although there is research on the development of self-evaluation and -regulation skills of children in elementary school, very limited research addresses this when learning from texts. The present study aims to fill that gap by testing elementary school children of two age groups twice within one school year. This research design allows to compare differences in accuracy of self-evaluation and -regulation between the two age groups, and to get insight about intra-individual stability over time.
Second-graders (N = 138) and 4th graders (N = 164) were tested two times within half a year. After reading six expository texts, children completed a test with 12 open-ended questions (e.g. “What happens to a bee’s stinger after stinging a human?”). After giving each answer, children immediately made a Confidence Judgment (CJ) on a 7-point scale ranging from “very unsure” to “very sure”. After answering all questions and giving CJs (self-evaluation), children went through their answers again: They could either maintain or withdraw their answers (self-regulation), and gain or lose points by accurate or inaccurate maintenance.
Findings confirm and extend previous research using recognition and memory tasks (rather than a comprehension task, as used in this study), and show that older elementary school children felt less certain after giving an incorrect answer and more often withdrew that answer than younger children. In contrast, 2nd grade children were more confident than 4th graders, and less likely to withdraw their incorrect answers, indicating less accurate self-evaluations and -regulation. Confidence for correct answers did not differ across age groups and measurement points, indicating that improvement in self-evaluation and -regulation mainly occurs when evaluating and regulating incorrect answers. Findings for stability between the two measurement points show that confidence as such (magnitude of CJs), and the tendency to maintain answers (% answers maintained) is stable for both age groups. In contrast, results for accuracy of self-evaluations (discrimination between CJs for correct and incorrect answers) and -regulation (discrimination between % of correct answers and incorrect answers maintained) show no stability at this age. It seems that children have a tendency to make similar confidence judgements and maintenance decisions over time. However, their ability to also accurately take their actual performance into account when evaluating and regulating learning does not yet seem stable. Knowing that development mainly occurs when evaluating and regulating incorrect answers, and which aspects are stable and which not, can give input when developing interventions aimed at improving children’s ability to self-evaluate and self-regulate learning.

Authors

• Martina Steiner, University of Bern, Switzerland
  Presenting Author

• Mariëtte van Loon, University of Bern, Switzerland
  Non-Presenting Author

• Natalie Bayard-Guggisberg, University of Bern, Switzerland
  Non-Presenting Author

• Claudia Roebers, Bern University
  Non-Presenting Author