Abrami, P.C., Bernard, R.M., Borokhovski, E., Wade, A., Surkes, M.A., Tamim, R., & Zhang, D. (2008)

Critical thinking (CT), or the ability to engage in purposeful, self-regulatory judgment, is widely recognized as an important, even essential, skill. This article describes an on-going meta-analysis that summarizes the available empirical evidence on the impact of instruction on the development and enhancement of critical thinking skills and dispositions. We found 117 studies based on 20,698 participants, which yielded 161 effects with an average effect size (g+) of 0.341 and a standard deviation of 0.610. The distribution was highly heterogeneous (QT = 1,767.86, p < .001). There was, however, little variation due to research design, so we neither separated studies according to their methodological quality nor used any statistical adjustment for the corresponding effect sizes. Type of CT intervention and pedagogical grounding were substantially related to fluctuations in CT effects sizes, together accounting for 32% of the variance. These findings make it clear that improvement in students’ CT skills and dispositions cannot be a matter of implicit expectation. As important as the development of CT skills is considered to be, educators must take steps to make CT objectives explicit in courses and also to include them in both pre-service and in-service training and faculty development.

de Boer, H., Donker, A. S., & van der Werf, M. P. (2014)

This meta-analysis examined the influence of attributes related to the implementation of learning strategy instruction interventions on students’ academic performance, and also examined how the attributes related to the method of testing the intervention effects affected the actual effects measured. Using meta-regression, we analyzed the influence of the subject domain in which the intervention was implemented, the implementer, its duration and intensity, student cooperation, and research method aspects (including measurement instrument). Most attributes moderated the intervention effect. Using forward regression analysis, we only needed four attributes to obtain the best model, however. This analysis showed that the intervention effect was lower when a standardized test was used for evaluation instead of an unstandardized test. Interventions implemented by assistants or researchers were more effective than those implemented by teachers or using computers. Cooperation had a negative, and session duration a positive, contribution. Together, these attributes explained 63.2% of the variance in effect, which stresses the importance of emphasizing not only the instructional focus of an intervention but also its other attributes.

Chiu, C.W.T. (1998)

In this paper, meta-analysis is used to identify components that are associated with effective metacognitive training programs in reading research. Forty-three studies, with an average of 81 students per study, were synthesized. It was found that metacognitive training could be more effectively implemented by using small-group instruction, as opposed to large-group instruction or one-to-one instruction. Less intensive programs were more effective than intensive programs. Program intensity was defined as the average number of days in a week that instruction was provided to students. Students in higher grades were more receptive to the intervention. Measurement artifacts, namely teaching to the test and use of non-standardized tests and the quality of the studies synthesized played a significant role in the evaluation of the effectiveness of the metacognitive reading intervention.

Dignath, C., Buettner, G. & Langfeldt, H. (2008)

Recently, research has increasingly focused on fostering self-regulated learning amongst young children. To consider this trend, this article presents the results of a differentiated meta-analysis of 48 treatment comparisons resulting from 30 articles on enhancing self-regulated learning amongst primary school students. Based on recent models of self regulated learning, which consider motivational, as well as cognitive, and metacognitive aspects [Boekaerts, M. (1999). Self-regulated learning: Where we are today. International Journal of Educational research, 31(6), 445–457], the effects of self-regulated learning on academic achievement, on cognitive and metacognitive strategy application, as well as on motivation were analyzed. As the results show, self-regulated learning training programmes proved to be effective, even at primary school level. Subsequent analysis tested for the effects of several moderator variables, which consisted of study features and training characteristics. Regarding factors that concern the content of the treatment, the impact of the theoretical background that underlies the intervention was tested, as well as the type of cognitive, metacognitive, or motivational strategy which were instructed, and if group work was used as instruction method. Training context related factors, which were included in the analyses, consisted of students’ grade level, the length of the training, if teachers or researchers directed the intervention, as well as the school subject in which context the training took place. Following the results of these analyses, a list with the most effective training characteristics was provided.

Donker, A. S., De Boer, H., Kostons, D., Dignath van Ewijk, C. C., & Van der Werf, M. P. C. (2014)

In this meta-analysis the results of studies on learning strategy instruction focused on improving self-regulated learning were brought together to determine which specific analysis included 58 studies in primary and secondary education on interventions aimed at improving cognitive, metacognitive, and management strategy skills, as well as motivational aspects and metacognitive knowledge. A total of 95 interventions and 180 effect sizes demonstrated substantial effects in the domains of writing (Hedges’ g = 1.25), science (.73), mathematics (.66) and comprehensive reading (.36). These domains differed in terms of which strategies were the most effective in improving academic performance. However, metacognitive knowledge instruction appeared to be valuable in all of them. Furthermore, it was found that the effects were higher when self-developed tests were used than in the case of intervention-independent tests. Finally, no differential effects were observed for students with different ability levels. To conclude, the authors have listed some implications of their analysis for the educational practice and made some suggestions for further research.

Fauzan, N. (2003)

The purpose of the study was to investigate the effectiveness of metacognitive strategies on reading comprehension by means of (a) a meta-analysis and (b) an experiment designed following the meta-analysis implemented in Sarawak, Malaysia. Before the meta-analysis, the prevalent theories and issues in the reading literature such as metacognition, models of reading, measurements, motivation and previous meta-analysis were discussed to provide a better understanding of the research area in this study. A meta-analytic procedure conducted to review the primary research studies of metacognitive strategies used effect size as the measure of effectiveness. Searching for the articles and theses in the 1980s until 2001 yielded a record of 473 abstracts and articles from which there were twenty seven studies with a total number of eighty two effect sizes that could be quantitatively synthesized to compare the group performance of the experimental and control groups. The weighted effect size was 0.50 (95% Cl = 0.45 to 0.56) when dependent effect sizes were synthesized, and 0.55 (95% CI=0.48 to 0.63) when the extreme 'outliers' or deviated effect sizes were excluded and independent effect sizes were created. Overall, the effect size was moderate indicating a positive outcome of the metacognitive strategies. The effect sizes were not homogeneous and further analyses of the qualitative and quantitative features of the studies were made to develop possible reliable estimates.

Haller, E.P., Child, D.A. & Walberg, H.J. (1988)

To assess the effect of “metacognitive” instruction on reading comprehension, 20 studies, with a total student population of 1,553, were compiled and quantitatively synthesized. For 115 effect sizes, or contrasts of experimental and control groups' performance, the mean effect size was .71, which indicates a substantial effect. In this compilation of studies, metacognitive instruction was found particularly effective for junior high students (seventh and eighth grades). Among the metacognitive skills, awareness of textual inconsistency and the use of self-questioning as both a monitoring and a regulating strategy were most effective. Reinforcement was the most effective teaching strategy.

Higgins, S., Hall, E., Baumfield, V., & Moseley, D. (2005)

Executive Summary Methods: Relevant studies in the area of thinking skills were obtained by systematically searching a number of online databases of educational research literature, by identifying references in reviews and other relevant books and reports, and from contacts with expertise in this area. Twenty-six of the studies identified for this review were obtained from the database which resulted from the first thinking skills review (Higgins et al., 2004); a further three resulted from updating the original search and applying the more stringent criteria required for a quantitative synthesis. Studies were selected for the meta-analysis if they had sufficient quantitative data to calculate an effect size (relative to a control or comparison group of pupils) and if the number of research subjects was greater than 10. Effect sizes were calculated from the reported data and combined statistically using quantitative synthesis. Results: twenty-nine studies were identified which contained quantitative data on pupils’ attainment and attitudes suitable for meta-analysis. The studies come from a range of countries around the world with half set in the US or UK. The studies broadly cover the ages of compulsory schooling (5–16) and include studies set in both primary and secondary schools. A number of named thinking skills interventions are included, such as Feuerstein’s instrumental enrichment (FIE) and cognitive acceleration through science education (CASE) as well as studies which report a more general thinking skills approach (such as the development of metacognitive strategies). The quantitative synthesis indicates that thinking skills programmes and approaches are effective in improving the performance on tests of cognitive measures (such as Raven’s progressive matrices) with an overall effect size of 0.62. (This effect would move a class ranked at 50th place in a league table of 100 similar classes to 26th or a percentile gain of 24 points.) However, these approaches also have a considerable impact on curricular outcomes with the same effect size of 0.62. The overall effect size (including cognitive, curricular and affective measures) was 0.74. Conclusions: Overall, the quantitative synthesis indicates that, when thinking skills programmes and approaches are used in schools, they are effective in improving pupils’ performance on a range of tested outcomes (relative to those who did not receive thinking skills interventions). The magnitude of the gains found appears to be important when compared with the reported effect sizes of other educational interventions. This review found an overall mean effect of 0.62 for the main (cognitive) effect of each of the included studies, larger than the mean of Hattie’s vast database of meta-analyses at 0.4 (Hattie, 1999) but very similar to the overall figure reported by Marzano (1998, p 76) of 0.65 for interventions across the knowledge, cognitive, metacognitive and self-system domains. In particular, our study identified metacognitive interventions as having relatively greater impact, similar to Marzano’s study. Looking at a smaller part of our review, Feuerstein’s instrumental enrichment is one of the most extensively researched thinking skills programme. Our results broadly concur with those of Romney and Samuels (2001), whose meta-analysis found moderate overall effects and an effect size of 0.43 on reasoning ability (p 28). Our findings were of the same order, with an overall effect size of 0.58 (one main effect from each of seven studies included) and an effect size of 0.52 on tests of reasoning (one main effect from four studies). There is some indication that the impact of thinking skills programmes and approaches may vary according to subject. In our analysis there was relatively greater impact on tests of mathematics (0.89) and science (0.78), compared with reading (0.4).

Klauer, K.J. & Phye, G.D. (2008)

Researchers have examined inductive reasoning to identify different cognitive processes when participants deal with inductive problems. This article presents a prescriptive theory of inductive reasoning that identifies cognitive processing using a procedural strategy for making comparisons. It is hypothesized that training in the use of the procedural inductive reasoning strategy will improve cognitive functioning in terms of (a) increased fluid intelligence performance and (b) better academic learning of classroom subject matter. The review and meta-analysis summarizes the results of 74 training experiments with nearly 3,600 children. Both hypotheses are confirmed. Further, two moderating effects were observed: training effects on intelligence test performance increased over time, and positive problem solving transfer to academic learning is greater than transfer to intelligence test performance. The results cannot be explained by placebo or test-coaching effects. It is concluded that the proposed strategy is theoretically and educationally promising and that children of a broad age range and intellectual capacity benefit with such training.

Losinski, M., Cuenca-Carlino, Y., Zablocki, M., & Teagarden, J. (2014)

Two previous reviews have indicated that self-regulated strategy instruction (SRSD) is an evidence-based practice that can improve the writing skills of students with emotional and behavioral disorders. The purpose of this meta-analysis is to extend the findings and analytic methods of previous reviews by examining published studies regarding SRSD, analyzing the findings of both single-case and group designs using a common effect size metric (Hedges’ g), and applying methods to address publication bias. In addition, the present meta-analysis examined the difference in treatment effect due to differences in moderating variables. Sixteen of the 20 studies examined met inclusion criteria based on the Council for Exceptional Children's Standards for Evidence-Based Practices in Special Education. Results indicated that SRSD interventions had large effect sizes across three dependent variables (i.e., essay elements , quality , and word count), and treatment effects were significant for study design and race/ethnicity. Type of instruction, intervention agent , and gender did not significantly predict response to SRSD instruction. Also the results showed limited risk of bias in the tendency of journals to publish only positive findings. Based on these findings, implications for future research and teaching with SRSD are discussed.

Zheng, L. (2016)

This meta-analysis examined research on the effects of self-regulated learning scaffolds on academic performance in computer-based learning environments from 2004 to 2015. A total of 29 articles met inclusion criteria and were included in the final analysis with a total sample size of 2648 students. Moderator analyses were performed using a random effects model that focused on the three main areas of scaffold characteristics (including the mechanism, functions, delivery forms, mode, and number of scaffolds; how to promote self-regulated learning by scaffolds); demographics of the selected studies (including sample groups, sample size, learning domain, research settings, and types of computer-based learning environments); and research methodological features (including research methods, types of research design, types of organization for treatment, and duration of treatment). Findings revealed that self-regulated learning scaffolds in computer-based learning environments generally produced a significantly positive effect on academic performance (ES = 0.438). It is also suggested that both domain-general and domain-specific scaffolds can support the entire process of self-regulated learning since they demonstrated substantial effects on academic performance. Different impacts of various studies and their methodological features are presented and discussed.