48,478 students’ data from 222 studies and 573 effects – Testing Effect (III)
48,478 students’ data from 222 studies and 573 effects – Testing Effect (III)

48,478 students’ data from 222 studies and 573 effects – Testing Effect (III)

The opening post dealt with the broad overview of the paper and the key weighty finding that testing (class quizzing) yields a variety of learning benefits. Part II covered the open questions two to six. Here we are at question seven.

Q7. Does the number of test repetitions matter?

The more occasions on which class content is quizzed, the larger the learning gains. Quiz once (g = 0.444), twice (g = 0.601), and three times or more (g = 0.642). Quizzes that permitted unlimited attempts also yielded significant enhancement (g = 0.762). The more occasions class content is tested, the larger the learning gains – however not explicitly explored, if noted in the paper, those Successive Relearning gains diminish each time.

Reflections: Successive Relearning can not be underestimated. Thinking about the recent addition of the Animal Farm deck to Classroom… “Four exposures good, one exposure bad.” Also, it is the Successive Relearning and re-exposure that adds to the indirect benefit of confidence. Not forgetting that even incorrect retrieval attempts enhance learning, Kornell et al. (2015).

There is a diminishing returns. With three recalls during initial learning, learners correctly recalled 80% of the content 2 days after a second relearning session and and 77% 1 week after a third relearning session, and 57% recall after 4 – 6 weeks. Rawson and Dunlosky (2011). What is not commented upon is that, in the classroom, as learners retain knowledgeable, their retrieval latency or thinking time reduces, and the successive relearning cycles become more efficient. Certainly, by the time learners more to covert of verbal responses, learners in my classes can respond to 10 prompts in 30 seconds, comfortably.

Yang et al (2021) also include a fantastic visual – see Figure 8.

Q8. Does test-enhanced learning work at all levels of education?

Test-enhanced learning generalizes to elementary school (g = 0.328), middle school (g = 0.597),
high school (g = 0.655), and university/college (g = 0.486).

Q9. Does testing enhance classroom learning to different extents for male and female students?

There is no reliable association between female gender ratios and the classroom testing effect, indicating that male and female students benefit from testing to a comparable extent.

Reflection: One other individual difference, after age and gender, I would like to explore is prior attainment and learner confidence definitely a variable I want to explore further.

Q10. Does test-enhanced learning generalize to a range of subjects?

Across 18 subject categories, testing consistently facilitates learning achievement. The paper actually effects from 31 different subjects. Little to explore further there. Whole school teaching and learning leads may see greater opportunity for impact given testings utility.

Q11. Does testing benefit different levels of knowledge?

Testing is not only beneficial for learning facts (g = 0.524), but also promotes conceptual learning (g = 0.644) and facilitates knowledge application in the service of problem solving (g = 0.453).

Reflection: Testing is often derailed for being “drill-and-kill,” yes. It is not a return to rote learning. We have four modes and over ten different activities for educators to explore. Clearly the design of the test or activity is critical here.

Indeed, Yang et al (2021) were surprised that this aspect of education, or question, has never been explored before.

8 more questions to go…

Opening postQuestions 2-6Question 7-11Questions 12-16Questions ?-?

Leave a Reply