Fostering Self-Regulated Learning in Online Environments: Positive Effects of a Web-Based Training With Peer Feedback on Learning Behavior.

Although training in self-regulated learning (SRL) is effective in improving performance, human trainers can reach only a few people at a time. We developed a web-based training for potentially unlimited numbers of participants based on the process model of SRL by Schmitz and Wiese (2006). A prior study (Bellhäuser et al., 2016) observed positive effects on self-reported SRL and self-efficacy. In the present randomized controlled trial, we investigated an improved version of the web-based training, augmented by the application of peer feedback groups. Prospective university students in an online mathematics preparation course were assigned randomly to one of four experimental conditions: Group D (diary), group TD (training + diary), group TDP (training + diary + peer feedback group), and group C (control). Complete data was obtained for 136 participants (78.8% male; = 19.8 years). The learning diary was intended to trigger goal setting, planning, and self-motivation in the morning and reflection in the evening. The web-based training consisted of three lessons (approximately 90 min each) with videos, presentations, self-tests, and exercises. In the peer feedback condition, participants were randomly assigned to groups of five persons each and used a bulletin board to discuss pre-defined topics related to the content of the web-based training. Outcome measures included a test of declarative SRL knowledge, an SRL questionnaire, a general self-efficacy scale, log file data, and a mathematics test. Results showed positive effects for the web-based training, particularly when combined with peer feedback on both SRL knowledge and SRL questionnaires, self-efficacy, and on objective time-investment, but not on the mathematics test. The learning diary did not exhibit positive effects. We conclude that additional peer-feedback seems to be a useful supplement to web-based trainings with comparably low organizational costs.