The influence of early questions on learning from text

In this research we explored the use of short-answer questions to improve learning from chapter-like texts (3395 words). Experiment 1 investigated the influence of pre-questions on recall from a text passage when tested a week later; two question sets were counterbalanced within the experimental group. Participants with pre-questions scored higher both overall (d = 3.6, 95%CI [2.4, 4.8]) and on novel questions (d = 2.0 [1.6, 2.4]). In Experiment 2, questions were made available immediately after studying the text either alongside the text, open-book, or closed-book with the opportunity to check answers, or not at all with additional study time. Learning was tested after a week. Although the immediate test scores were substantially higher for open- than closed-book tests, week-delayed performance on the same items was much worse for open-book tests and was moderately improved for closed-book tests. For seen questions, closed-book tests led to better delayed recall than did open-book tests, d = 0.7 [0.02, 1.5]. For novel questions, observed differences were small; ds = .2 [-0.6, 0.9] for both comparisons.