Power-Enhanced Funnel Plots for Meta-Analysis

Kossmeier, Michael; Tran, Ulrich S.; Voracek, Martin

doi:10.1027/2151-2604/a000392

Cited by 41 publications

(42 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For instance, heterogeneity among effect sizes can create asymmetries ; (e) a residual funnel plot from a meta-regression with one moderator; and (f) a radial plot showing the overall effect by its slope's steepness and heterogeneity with the degree of scattering of the data points (for more details, see the main text). We used the r packages metafor (panels a-c and e; Viechtbauer, 2010), metaviz (panel d; Kossmeier et al, 2020) and meta (panel f; Schwarzer et al, 2015) for visualizations of many kinds (Figure 3b). Incidentally, the other potential sources of asymmetry are data irregularities (e.g.…”

Section: Funnel Plotsmentioning

confidence: 99%

Methods for testing publication bias in ecological and evolutionary meta‐analyses

et al. 2021

View full text Add to dashboard Cite

1. Publication bias threatens the validity of quantitative evidence from meta-analyses as it results in some findings being overrepresented in meta-analytic datasets because they are published more frequently or sooner (e.g., 'positive' results). Unfortunately, methods to test for the presence of publication bias, or assess its impact on meta-analytic results, are unsuitable for datasets with high heterogeneity and non-independence, as is common in ecology and evolutionary biology.2. We first review both classic and emerging publication bias tests (e.g., funnel plots, Egger's regression, cumulative meta-analysis, fail-safe N, trim-and-fill tests, p-curve and selection models), showing that some tests cannot handle heterogeneity, and, more importantly, none of the methods can deal with non-independence. For each method we estimate current usage in ecology and evolutionary biology, based on a representative sample of 102 meta-analyses published in the last ten years.3. Then, we propose a new method using multilevel meta-regression, which can model both heterogeneity and non-independence, by extending existing regression-based methods (i.e.Egger's regression). We describe how our multilevel meta-regression can test not only publication bias, but also time-lag bias, and how it can be supplemented by residual funnel plots.4. Overall, we provide ecologists and evolutionary biologists with practical recommendations on which methods are appropriate to employ given independent and non-independent effect sizes.No method is ideal, and more simulation studies are required to understand how Type 1 and 2 error rates are impacted by complex data structures. Still, limitations of these methods do not justify ignoring publication bias in ecological and evolutionary meta-analyses.

show abstract

Section: Funnel Plotsmentioning

confidence: 99%

Methods for testing publication bias in ecological and evolutionary meta‐analyses

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In addition, another approach to determine the evidential value of studies included in the analysis is to check the statistical power of individual studies. Therefore, we performed a sunset funnel plot [ 48 ], which is a funnel plot variant that visualizes the statistical power of each study included in the meta-analysis based on the summary effect size.…”

Section: Methodsmentioning

confidence: 99%

Evaluating the Effectiveness of Gamification on Physical Activity: Systematic Review and Meta-analysis of Randomized Controlled Trials

Mazéas¹,

Duclos²,

Pereira³

et al. 2022

J Med Internet Res

View full text Add to dashboard Cite

Background Gamification refers to the use of game elements in nongame contexts. The use of gamification to change behaviors and promote physical activity (PA) is a promising avenue for tackling the global physical inactivity pandemic and the current prevalence of chronic diseases. However, there is no evidence of the effectiveness of gamified interventions with the existence of mixed results in the literature. Objective The aim of this systematic review and meta-analysis is to evaluate the effectiveness of gamified interventions and their health care potential by testing the generalizability and sustainability of their influence on PA and sedentary behavior. Methods A total of 5 electronic databases (PubMed, Embase, Scopus, Web of Science, and the Cochrane Central Register of Controlled Trials) were searched for randomized controlled trials published in English from 2010 to 2020. Eligibility criteria were based on the components of the participants, interventions, comparators, and outcomes framework. Studies were included when they used gamified interventions in daily life with an active or inactive control group and when they assessed a PA or sedentary behavior outcome. We conducted meta-analyses using a random-effects model approach. Sensitivity analyses, influence analyses, and publication bias analyses were performed to examine the robustness of our results. Results The main meta-analysis performed on 16 studies and 2407 participants revealed a small to medium summary effect of gamified interventions on PA behavior (Hedges g=0.42, 95% CI 0.14-0.69). No statistical difference among different subgroups (adults vs adolescents and healthy participants vs adults with chronic diseases) and no interaction effects with moderators such as age, gender, or BMI were found, suggesting good generalizability of gamified interventions to different user populations. The effect was statistically significant when gamified interventions were compared with inactive control groups, such as waiting lists (Hedges g=0.58, 95% CI 0.08-1.07), and active control groups that included a nongamified PA intervention (Hedges g=0.23, 95% CI 0.05-0.41). This suggests that gamified interventions are not only efficient in changing behavior but also more effective compared with other behavioral interventions. The long-term effect (measured with follow-up averaging 14 weeks after the end of the intervention) was weaker, with a very small to small effect (Hedges g=0.15, 95% CI 0.07-0.23). Conclusions This meta-analysis confirms that gamified interventions are promising for promoting PA in various populations. Additional analyses revealed that this effect persists after the follow-up period, suggesting that it is not just a novelty effect caused by the playful nature of gamification, and that gamified products appear effective compared with equivalent nongamified PA interventions. Future rigorous trials are required to confirm these findings.

show abstract

“…We used R (Version 4.0.2; R Core Team, 2020) to conduct all of our analyses and create all of our figures using the following packages: “metafor” (Version 2.4; Viechtbauer, 2010), “tidyverse” (Version 1.3.0; Wickham et al, 2019), “readxl” (Version 1.3.1; Wickham & Bryan, 2019), “PRISMAstatement” (Version 1.1.1; Wasey, 2019), “metaviz” (Version 0.3.1; Kossmeier et al, 2020), “metameta” (Version 0.1.1; Quintana, 2020a), and “ggthemes” (Version 4.2.0; Arnold, 2019). We opted to use a random-effects model that estimated heterogeneity using restricted maximum likelihood.…”

Section: Methodsmentioning

confidence: 99%

Suicidal Thoughts, Behaviors, and Event-Related Potentials: A Systematic Review and Meta-Analysis

Gallyer

Dougherty

Burani

et al. 2020

Preprint

View full text Add to dashboard Cite

Suicidal thoughts and behaviors (STBs) are thought to be the result, at least in part, of abnormalities in various neural systems. Event-related potentials (ERPs) are a useful method for studying neural activity, and can be leveraged to study neural deficits related to STBs, but it is unknown how effective ERPs are at differentiating various STB groups. The present metaanalysis examined how well ERPs can differentiate those with and without suicidal ideation (SI), suicide attempts (SA), and those determined to be at different levels of suicide risk, or suicidality. Our meta-analysis included 192 effect sizes from 24 studies. We used a random effects meta-analysis and found a small-to-moderate relationship between ERPs and STBs.However, we also found evidence that the literature may be suffering from small sample publication bias. Our results suggest that ERPs may compliment other approaches, such as fMRI, in the study of the neurobiology of individuals with STBs. However, the clinical utility of ERPs to predict STBs remains to be tested, and larger samples are needed to confirm the results of the present meta-analysis.

show abstract

Power-Enhanced Funnel Plots for Meta-Analysis

Cited by 41 publications

References 30 publications

Methods for testing publication bias in ecological and evolutionary meta‐analyses

Methods for testing publication bias in ecological and evolutionary meta‐analyses

Evaluating the Effectiveness of Gamification on Physical Activity: Systematic Review and Meta-analysis of Randomized Controlled Trials

Suicidal Thoughts, Behaviors, and Event-Related Potentials: A Systematic Review and Meta-Analysis

Contact Info

Product

Resources

About