Refutation texts, rhetorical tools designed to reduce misconceptions, have garnered attention across four decades and many studies. Yet, the ability of a refutation text to change a learner’s mind on a topic needs to be qualified and modulated. In this critical review, we bring attention to sources of constraints often overlooked by refutation text researchers. Methodological issues we identified centered on problems of using a single topic (or very few) within a study, the role of testing in conceptual change, and the durability of change beyond an immediate posttest. Theoretical issues included the interpretation of attentional measures, what knowledge domains lend themselves to refutation, what makes a text refutational, and unexplored assumptions about how conceptual change occurs. We sought to clarify how refutation texts may function as an antidote to misconceptions and how future research on refutation texts can better inform understanding of this phenomenon.

Researchers across many fields have called for greater attention to heterogeneity of treatment effects—shifting focus from the average effect to variation in effects between different treatments, studies, or subgroups. True heterogeneity is important, but many reports of heterogeneity have proved to be false, nonreplicable, or overestimated. In this review, we catalog ways that past researchers fooled themselves about heterogeneity and recommend ways that we, as researchers, can stop fooling ourselves about heterogeneity in the future. We make 18 specific recommendations. The most common themes are to (a) seek heterogeneity only when the mechanism offers clear motivation and the data offer adequate power, (b) shy away from seeking “no-but” heterogeneity when there is no main effect, (c) separate the noise of estimation error from the signal of true heterogeneity, (d) shrink variation in estimates toward zero, (e) increase p values and widen confidence intervals when conducting multiple tests, (f) estimate interactions rather than subgroup effects, and (g) check whether findings of heterogeneity are sensitive to changes in model or measurement. We also resolve long-standing debates about centering interactions in linear models and estimating interactions in nonlinear models, such as logistic, ordinal, and interval regression. If researchers follow these recommendations, the search for heterogeneity should yield more trustworthy results in the future.

Financial literacy is an important life skill, yet the impact of financial education has often been found to be modest. We conducted a field experiment to assess the effectiveness of a postinstruction intervention using a smartphone app that incorporated cognitive science principles aimed at improving learning. College students who completed a required credit review workshop during their sophomore year used the smartphone app in one of three practice conditions: control (no practice), massed, or spaced retrieval practice with elaborative feedback. On a final assessment about 5 months later, students who engaged in spaced retrieval practice were superior to those in the control and massed practice conditions in terms of knowledge. Given the ubiquity of smartphones today, the results highlight the potential of harnessing easily accessible technology as learning tools to augment the retention and transfer of knowledge.