H.R. Rothstein’s research while affiliated with City University of New York - Bernard M. Baruch College and other places

... According to our hypothesis that moderators of individual responses differ from those at population levels, we grouped the effect sizes of metrics by the biological organization level. For each level, random-effect models were adopted to estimate the mean response of the metrics to droughts across the observations, to account for the interdependence among observations within each study in addition to the studies' sampling variances (Borenstein et al., 2009). Moreover, because effect sizes of various power were ...

Reference:

Biological traits and biome features mediate responses of terrestrial bird demography to droughts

... Statistical heterogeneity was evaluated using Higgins I 2 statistics and the Cochrane Q (Chi 2 test) [18]. Meta-analysis was performed using Review Manager version 5.4 (Revman 5.4) and Comprehensive Meta-Analysis v3 (CMA V3) software [19,20]. The significant difference was revealed when the probability value (P) < 0.05. ...

Reference:

Factors predicting composite grafts survivability in patients with fingertip amputations; a systematic review and meta-analysis

... I 2 statistics of 25%, 50%, and 75% were interpreted as indicating low, moderate, and high heterogeneity, respectively [20]. Due to the anticipated differences among studies, a random-effects model was applied [21]. ...

Reference:

Efficacy of Zinc Supplementation in the Management of Primary Dysmenorrhea: A Systematic Review and Meta-Analysis

... By analyzing the effect sizes of predictors in different experiments, a weighted average effect size can emerge to produce a more generalizable effect size that may not have been apparent in any one study [62][63][64][65]. However, meta-analyses may be hindered by publication bias, as studies that do not present marked results tend not to be published [66]. A small-scale meta-analysis can be conducted using results from several experiments carried out by a research team. ...

Reference:

Who Learns from Reading Texts in General Chemistry?

... We further computed I 2 to estimate the ratio of true heterogeneity to total observed variation. Publication bias, or the tendency that studies with statistically significant results are more likely to find their way into the published literature, was examined by means of funnel plots, with Egger regression and trimand-fill analysis for estimation of the adjusted effect size and of missing studies ( Borenstein et al., 2009b). These procedures are based on the expectation that the publication bias will increase with a smaller sample size. ...

Reference:

Cerebellar non-invasive stimulation of social and emotional mentalizing: A meta-analysis

... For the meta-analysis, we used the effect size metric Hedges' g. To calculate Hedges' g ( Borenstein et al., 2009a), we used the same strategy and general procedures as in an earlier meta-analysis ( Barahona-Corrêa et al., 2018). We took the following descriptive data from each study: sample size, mean, and standard deviation (SD) of the outcome measure for the active stimulation and the sham/control site stimulation. ...

Reference:

Cerebellar non-invasive stimulation of social and emotional mentalizing: A meta-analysis

... The meta-analysis method is a method that can be used to estimate the effect size of the relationship. both variables (Borenstein, 2009). Therefore, it is necessary to do a meta-analysis method in this study. ...

Reference:

A Meta-Analysis of Pyschological Capital and Psychological Well-Being: Testing on Relationship between Self-Efficacy and Self-Acceptance

... To investigate the effects of potential moderators, we created subsamples of studies based on the moderator variables, and performed additional meta-analyses in these subsamples (Schmidt & Hunter, 2015b). While meta-regression is recommended by some for assessing the impact of moderators (Borenstein, Hedges, Higgins, & Rothstein, 2009), we opted against this approach due to its reduced statistical power in small samples (Schmidt & Hunter, 2015b) and because our moderators were categorical. ...

Reference:

Does Interest Fit Between Student and Study Program Lead to Better Outcomes? A Meta-Analysis of Vocational Interest Congruence as Predictor for Academic Success

... Meta-regression analyses were used to assess the association of different teaching methods and content of the intervention, with the effect size for each outcome. Since there was considerable heterogeneity in the reporting of biomedical outcomes, we also used full random effects analyses to combine studies within each subgroup (24). A random effects model is used to combine subgroups and yield the overall effect. ...

Reference:

Effectiveness of therapeutic patient education interventions for chronic diseases: A systematic review and meta-analyses of randomized controlled trials

... Heterogeneity, or variability among studies, is a common challenge in meta-analysis. Differences in study design, populations, interventions and outcomes can contribute to heterogeneity, complicating the interpretation of results [45]. Statistical methods, such as meta-regression and subgroup analyses, are used to address heterogeneity. ...

Reference:

Unveiling the Value of Meta-Analysis in Disease Prevention and Control: A Comprehensive Review