Retrieval Practice and Spacing in Naturalistic Settings: Undergraduates’ Achievement Goals, Digital Quiz Use, and Performance in Biology

• In Event: Toward an Understanding of How Learning Strategies Can Be Effective for Learners Across Contexts

Abstract

Background
Retrieval practice is a powerful learning strategy shown to facilitate efficient construction of declarative and even conceptual levels of understanding (e.g., Roediger & Karpicke, 2006). The strategy becomes even more powerful when enacted on a timely schedule of spaced repetition (Storm et al., 2010). Retrieval can be prompted and scheduled in the laboratory, but few studies have examined unprompted, naturalistic retrieval practice behaviors in authentic learning tasks (e.g., Authors, Date; Rodriguez et al., 2021). Individuals’ achievement goals also influence how people engage in strategic learning and affect academic performance (Elliot, 2005). Relationships between learners’ achievement goals, retrieval practice behaviors, and achievement are theorized but remain unstudied in naturalistic learning environments.
We surveyed students’ achievement goals and used learning analytics to observe the frequency and timing of undergraduates’ naturalistic retrieval practice behaviors using self-quizzing resources in a biology course to examine students’ goals and retrieval and spacing, and their effects on course performance.

Method
Data were collected from 174 undergraduates (Mage = 25.73 years [SD = 7.62]; 81% female; 36.8% Hispanic, 27% White, 20.1% Asian, 4.6% Black, 1.7% Native Hawaiian/Pacific Islander, 9.8% Unknown) enrolled in a prerequisite health science lecture and lab course pairing at a Southwestern U.S. community college.
Instructors provided an optional enrichment course site on the Canvas learning management system that supplemented the lab and instructors’ lecture sites. Digital events reflecting students’ use of the site and four practice quizzes were collected and used to engineer features reflecting retrieval practice and spaced practice variables. These are hypothesized to predict performance on lab quizzes and a first lecture exam. Using timestamped attempts at quizzes, we engineered variables including individuals’ visits to the enrichment site, total quiz events, timing of each quiz use, and instances of spaced retrieval practice schedules (e.g., increasing or decreasing spacing gaps; Table 1). We administered the Achievement Goals Questionnaire Revised (Elliot & Murayama, 2008) to assess mastery approach goals, performance approach goals, performance avoidance goals within a digital questionnaire in the first week of class.
We examined correlations among features of practice behavior and academic performance, fitted models using these variables to predict quiz and exam performance, and investigated the relationship between achievement goals and retrieval practice behaviors.

Results
The amount of quiz attempts and some spacing metrics correlated with performance (Table 1). However, retrieval practice was infrequent and spacing even less frequent (Table 2). Regression findings (Table 3) confirm benefits of retrieval practice in course settings, but no effects of spacing were observed. Performance approach goals negatively predicted retrieval practice; mastery approach and performance approach goals did not (Table 4).
Seldom observed in naturalistic settings, retrieval and spacing were shown to benefit all learners and learners’ achievement goals predicted their use. Undergraduate science courses are important places to learn through rehearsal, but learners’ performance approach goals often drive behavior. Instructors can enable and encourage strategic rehearsal by designing mastery structured courses with digital quizzing resources to encourage desirable learning goals and engagement (Meece et al., 2006).

Authors

• Leiming Ding, University of North Carolina - Chapel Hill

• Matthew L. Bernacki, University of North Carolina - Chapel Hill

• Erin Windsor, College of Southern Nevada

• Nancy Webb, College of Southern Nevada

• Robert D Plumley, University of North Carolina - Chapel Hill

• Michael Abdul Ghani Berro, University of North Carolina - Chapel Hill

• Jeff A. Greene, University of North Carolina - Chapel Hill

• Jonathan C. Hilpert, University of Nevada - Las Vegas