Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review.

Abstract

Over the last century hundreds of studies have demonstrated that testing is an effective intervention to enhance long-term retention of studied knowledge and facilitate mastery of new information, compared with restudying and many other learning strategies (e.g., concept mapping), a phenomenon termed the testing effect. How robust is this effect in applied settings beyond the laboratory? The current review integrated 48,478 students’ data, extracted from 222 independent studies, to investigate the magnitude, boundary conditions, and psychological underpinnings of test-enhanced learning in the classroom. The results show that overall testing (quizzing) raises student academic achievement to a medium extent (g = 0.499). The magnitude of the effect is modulated by a variety of factors, including learning strategy in the control condition, test format consistency, material matching, provision of corrective feedback, number of test repetitions, test administration location and timepoint, treatment duration, and experimental design. The documented findings support 3 theories to account for the classroom testing effect: additional exposure, transfer-appropriate processing, and motivation. In addition to their implications for theory development, these results have practical significance for enhancing teaching practice and guiding education policy and highlight important directions for future research.

Generalizability, transferability, and the practice-to-practice gap

“…Moreover, when field research observes null findings, the failure is often attributed to constraints of implementation rather than limitations of theory. As an example, classroom research on retrieval practice interventions is promising (Dunlosky et al, 2013) but not consistently observed (e.g., Gurung & Burns, 2019; see also Moreira et al, 2019;Yang et al, 2021). Such mixed evidence hardly justifies the bold recommendations that it works for "all grade levels, all subject areas, and all students" (Agarwal et al, 2020, p. 6).…”

Standardized Test Proficiency in Public Montessori Schools

“…Some research shows negative effects of mixed-age classrooms (Checchi & De Paola, 2018), which could suggest Montessori schools would do less well on performance measures. Further, frequent testing is associated with better learning (Yang, Luo, Vadillo, Yu, & Shanks, 2021), and Montessori minimizes testing. For all these reasons, one might expect children at Montessori schools to perform worse on standardized tests.…”

Learning With a Double-Edged Sword? Beneficial and Detrimental Effects of Learning Tests—Taking a First Look at Linkages Among Tests, Later Learning Outcomes, Stress Perceptions, and Intelligence

“…The following work raises the question if normally beneficial learning tests actually serve as double-edged swords, thus, if they can result in both beneficial as well as detrimental effects: More specifically, the present work was conducted to simultaneously focus on the often observed positive long-term learning effects of tests as difficult and demanding learning strategies (see, e.g., Adesope et al, 2017 ; Yang et al, 2021 ) but also on potential negative (side) effects caused by such learning tests, namely, increased stress or anxiety perceptions (see, e.g., Hinze and Rapp, 2014 ; Wenzel and Reinhard, 2021 ). Such increased stress perceptions should have further detrimental effects on learning in general as well as on the beneficial effects of tests on long-term learning in specific (see, e.g., Seipp, 1991 ; Hinze and Rapp, 2014 ).…”

“…The term desirable difficulties thereby acts as an umbrella term for different intentionally hindered learning strategies, which lead to beneficial effects for later long-term learning outcomes: These include, for instance, disfluency (using harder-to-read fonts; Diemand-Yauman et al, 2011 ) and generation (generating materials and solutions instead of passive consumption; Bertsch et al, 2007 ). One especially robust desirable difficulty is the application of tests (also: testing, testing effect, retrieval practice, test-enhanced learning, and learning/practice tests ): Taking (learning) tests on previously studied materials increases long-term learning compared to easier and more passive re-reading tasks or compared to note-taking as a stronger control task—even concerning a multitude of difficult, complex, and curricular subjects in realistic learning contexts (e.g., McDaniel et al, 2007 ; Dunlosky et al, 2013 ; Rowland, 2014 ; Karpicke and Aue, 2015 ; Adesope et al, 2017 ; Batsell et al, 2017 ; Rummer et al, 2017 ; Yang et al, 2021 ). These beneficial effects of tests were, among others, found for different types of learning materials (e.g., factual information, vocabulary, conceptual information, longer scientific textbook paragraphs, traditional (live) lectures/lessons, and recorded e-lectures/video-presentations) and for different types of test questions (e.g., multiple-choice questions, short-answer questions, fill-in-the-blank questions, comprehension-based questions, application-based questions, transfer questions, and inferences; e.g., Roediger and Karpicke, 2006 ; McDaniel et al, 2011 , 2013 ; Dunlosky et al, 2013 ; Rowland, 2014 ; Khanna, 2015 ; Jing et al, 2016 ; Adesope et al, 2017 ; Iwamoto et al, 2017 ; Heitmann et al, 2018 ; Feraco et al, 2020 ; Yang et al, 2021 ).…”

“…One especially robust desirable difficulty is the application of tests (also: testing, testing effect, retrieval practice, test-enhanced learning, and learning/practice tests ): Taking (learning) tests on previously studied materials increases long-term learning compared to easier and more passive re-reading tasks or compared to note-taking as a stronger control task—even concerning a multitude of difficult, complex, and curricular subjects in realistic learning contexts (e.g., McDaniel et al, 2007 ; Dunlosky et al, 2013 ; Rowland, 2014 ; Karpicke and Aue, 2015 ; Adesope et al, 2017 ; Batsell et al, 2017 ; Rummer et al, 2017 ; Yang et al, 2021 ). These beneficial effects of tests were, among others, found for different types of learning materials (e.g., factual information, vocabulary, conceptual information, longer scientific textbook paragraphs, traditional (live) lectures/lessons, and recorded e-lectures/video-presentations) and for different types of test questions (e.g., multiple-choice questions, short-answer questions, fill-in-the-blank questions, comprehension-based questions, application-based questions, transfer questions, and inferences; e.g., Roediger and Karpicke, 2006 ; McDaniel et al, 2011 , 2013 ; Dunlosky et al, 2013 ; Rowland, 2014 ; Khanna, 2015 ; Jing et al, 2016 ; Adesope et al, 2017 ; Iwamoto et al, 2017 ; Heitmann et al, 2018 ; Feraco et al, 2020 ; Yang et al, 2021 ). Moreover, tests were beneficial in varying (face-to-face or online) settings (e.g., laboratories, universities, classrooms, and at home/outside of class) and for students of different age groups (e.g., elementary school students, high school students, and university students; e.g., McDaniel et al, 2007 , 2011 ; Roediger et al, 2011 ; Rowland, 2014 ; Adesope et al, 2017 ; Yang et al, 2021 ).…”