Journal of Educational Psychology, Vol 114(8), Nov 2022, 1723-1742; doi:10.1037/edu0000754
In this article, we leverage data from over 1,000 students participating in two different online courses to investigate whether better learning outcomes are associated with student decisions to practice instead of (re-)reading. Consistent with laboratory and classroom findings, we find that students’ decisions to practice are related to better learning outcomes. Moreover, we find that this benefit is particularly related to increasing the number of different practice activities completed and not repeating the same activity multiple times. Our findings are consistent with theories suggesting that practice testing improves learning by enhancing encoding promoted by practicing the same content with different problems and raise questions regarding the benefits of repeatedly practicing the same question. The work presented here also demonstrates one way we can leverage data from naturally occurring data sets and learning analytics approaches to inform theoretical developments and the understanding of cognitive phenomena. We argue that to fully understand the cognitive processes involved in learning we need to test our hypotheses in natural educational contexts, both using controlled experimentation and analyses of naturally occurring data. (PsycInfo Database Record (c) 2022 APA, all rights reserved)