Double-zero-event studies (DZS) pose a challenge for accurately estimating the overall treatment effect in meta-analysis (MA). Current approaches, such as continuity correction or omission of DZS, are commonly employed, yet these ad hoc methods can yield biased conclusions. Although the standard bivariate generalized linear mixed model (BGLMM) can accommodate DZS, it fails to address the potential systemic differences between DZS and other studies. In this article, we propose a zero-inflated bivariate generalized linear mixed model (ZIBGLMM) to tackle this issue. This two-component finite mixture model includes zero inflation for a subpopulation with negligible or extremely low risk. We develop both frequentist and Bayesian versions of ZIBGLMM and examine its performance in estimating risk ratios against the BGLMM and conventional two-stage MA that excludes DZS. Through extensive simulation studies and real-world MA case studies, we demonstrate that ZIBGLMM outperforms the BGLMM and conventional two-stage MA that excludes DZS in estimating the true effect size with substantially less bias and comparable coverage probability.

Network meta-analysis (NMA), also known as mixed treatment comparison meta-analysis or multiple treatments meta-analysis, extends conventional pairwise meta-analysis by simultaneously synthesizing multiple interventions in a single integrated analysis. Despite the growing popularity of NMA within comparative effectiveness research, it comes with potential challenges. For example, within-study correlations among treatment comparisons are rarely reported in the published literature. Yet, these correlations are pivotal for valid statistical inference. As demonstrated in earlier studies, ignoring these correlations can inflate mean squared errors of the resulting point estimates and lead to inaccurate standard error estimates. This article introduces a composite likelihood-based approach that ensures accurate statistical inference without requiring knowledge of the within-study correlations. The proposed method is computationally robust and efficient, with substantially reduced computational time compared to the state-of-the-science methods implemented in R packages. The proposed method was evaluated through extensive simulations and applied to two important applications including an NMA comparing interventions for primary open-angle glaucoma, and another comparing treatments for chronic prostatitis and chronic pelvic pain syndrome.

Substantial productivity increases have been reported when incentives are framed as losses rather than gains. Loss-framed contracts have also been reported to be preferred by workers. The results from our meta-analysis and real-effort experiment challenge these claims. The meta-analysis’ summary effect size of loss framing is a 0.16 SD increase in productivity. Whereas the summary effect size in laboratory experiments is a 0.33 SD, the summary effect size from field experiments is 0.02 SD. We detect evidence of publication biases among laboratory experiments. In a new laboratory experiment that addresses prior design weaknesses, we estimate an effect size of 0.12 SD. This result, in combination with the meta-analysis, suggests that the difference between the effect size estimates in laboratory and field experiments does not stem from the limited external validity of laboratory experiments, but may instead stem from a mix of underpowered laboratory designs and publication biases. Moreover, in our experiment, most workers preferred the gain-framed contract and the increase in average productivity is only detectable in the subgroup of workers (~ 20%) who preferred the loss-framed contracts. Based on the results from our experiment and meta-analysis, we believe that behavioral scientists should better assess preferences for loss-framed contracts and the magnitude of their effects on productivity before advocating for greater use of such contracts among private and public sector actors.

This paper reports the findings of a meta-analysis of 37 papers with 75 results from ultimatum game experiments. We find that on average the proposer offers 40% of the pie to the responder. This share is smaller for larger pie sizes and larger when a strategy method is used or when subjects are inexperienced. On average 16% of the offers is rejected. The rejection rate is lower for larger pie sizes and for larger shares offered. Responders are less willing to accept an offer when the strategy method is employed. As the results come from different countries, meta-analysis provides an alternative way to investigate whether bargaining behavior in ultimatum games differs across countries. We find differences in behavior of responders (and not of proposers) across geographical regions. With one exception, these differences cannot be attributed to various cultural traits on which for instance the cultural classifications of Hofstede (1991) and Inglehart (2000) are based.

We collect individual participant data from 70 papers that use laboratory experiments to examine individual tax evasion behavior (or “Tax Evasion Games”), in order to use meta-analysis to estimate the impacts of different public policy, experimental design and individual level variables on tax evasion choices. Our results show that standard enforcement variables like audits (including audit rules) and fines perform differently on the extensive and intensive margins. We find that other fiscal variables like a flat tax system, tax rates, and tax amnesties have unambiguous negative impacts on tax compliance, and that specific features of the experimental setting, such as how subjects are directed to report income, or whether taxes are redistributed to the participants or to a real life public good, have significant impacts on tax compliance. Our results also indicate that the demographic characteristics of the subjects (e.g., gender, experimental income, occupation, risk attitude) affect compliance.

Objective: To use meta-analysis techniques to assess the impact of various factors on the extent of cooperation in standard linear public goods experiments using the voluntary contributions mechanism.

Data Sources: Potentially relevant experiments were identified through searches of EconLit, the Internet Documents in Economics Access Service (IDEAS), and a survey article.

Review Methods: A total of 349 potentially relevant studies were identified. Of these, 27 (representing a total of 711 groups of participants) met the inclusion criteria. Data were abstracted from these studies using a standardized protocol. Results were analyzed using weighted ordinary least squares. Average group efficiency was the dependent variable.

Results: The marginal per capita return, communication, constant group composition over the session (“partners”), positive framing, and the use of children as subjects had a positive and significant effect (p < 0.05) on the average level of contribution to the public good. Heterogeneous endowments to subjects, experienced participants, and soliciting subjects’ beliefs regarding other participants’ behaviour prior to the start of the session/period had a negative and significant effect. A number of other factors were not identified as significant.

Conclusion: The meta-analysis results parallel several key findings from previous literature reviews. In addition, they offer parameter estimates and an analysis of significance based on the totality of the available research evidence. More consistent reporting of the results of experiments would greatly improve the ability to conduct this type of research.

One important determinant of voluntary contributions to public goods is the value of the public good relative to that of the forgone private good. Isaac, Walker and Thomas (1984) formalized this relation in the Marginal Per Capita Return (MPCR) and demonstrated its influence on the provision of linear public goods. This paper develops a parallel concept, in the context of a threshold public good, the Step Return (SR). After providing a meta-analysis of the effect of SR in previous experiments, we compare contributions in threshold public goods games with low, medium and high SRs. Results show that subjects respond to the SR in this setting just as they respond to the MPCR in the linear public goods setting: higher SRs lead to more contributions.

In this paper, we report a replication of Engel’s (Exp. Econ. 14(4):583–610, 2011) meta-study of dictator game experiments. We find Engel’s meta-study of dictator game experiments to be robust, with one important exception: the coding of the take-option (List in J. Polit. Econ. 115(3):482–493, 2007; Bardsley in Exp. Econ. 11(2):122–133, 2008; Cappelen et al. in Econ. Lett. 118(2):280–283, 2013). While Engel reports this as having no statistically significant effect, in our replications, we find an economically and statistically significant negative effect on giving in line with the relevant literature.

A key parameter estimated by lab and field experiments in economics is the individual discount rate—and the results vary widely. We examine the extent to which this variance can be attributed to observable differences in methods, subject pools, and potential publication bias. To address the model uncertainty inherent to such an exercise we employ Bayesian and frequentist model averaging. We obtain evidence consistent with publication bias against unintuitive results. The corrected mean annual discount rate is 0.33. Our findings also suggest that discount rates are independent across domains: people tend to be less patient when health is at stake compared to money. Negative framing is associated with more patience. Finally, the results of lab and field experiments differ systematically, and it also matters whether the experiment relies on students or uses broader samples of the population.

Green tea, a plant rich in bioactive compounds, has been highlighted for its beneficial effects. In the present systematic review and meta-analysis of randomised controlled trials (RCTs), the impact of green tea on inflammatory and oxidative markers is investigated. Using pre-defined keywords, online databases (PubMed, Scopus, Web of Science Core Collection, and Google Scholar) were searched for relevant articles, published from inception up to February 2024. The outcomes included C-reactive protein (CRP), tumour necrosis factor-α (TNF-α), interleukin-6 (IL-6), interleukin-1 beta (IL-1β), total antioxidant capacity (TAC), malondialdehyde (MDA), superoxide dismutase (SOD), and glutathione peroxidase (GPX). Analyses of subgroups, linear, and non-linear associations were also carried out. Out of 1264 records initially retrieved, 38 RCTs were included. Supplementation with green tea improved the following indicators: IL-1β (weighted mean difference (WMD): −0.10 pg/mL; 95% CI: −0.15, -0.06), MDA (WMD: −0.40 mcmol/L; 95 % CI: −0.63, −0.18), TAC (WMD: 0.09 mmol/L; 95% CI: 0.05, 0.13), SOD (WMD: 17.21 u/L; 95% CI: 3.24, 31.19), and GPX (WMD: 3.90 u/L; 95% CI: 1.85, 5.95); but failed to improve others, including CRP (WMD: 0.01 mg/L; 95% CI: −0.14, 0.15), IL-6 (WMD: −0.34 pg/mL; 95% CI:−0.94, 0.26), and TNF-α (WMD: −0.07 pg/mL; 95% CI: −0.42, 0.28). Supplementation with green tea can improve the body’s oxidative status. However, the results showed no significant effect of green tea on inflammatory markers, except for IL-1β. Further studies are needed to determine the effectiveness of green tea, particularly on inflammatory status.

The current meta-analytic structural equation modeling (MASEM) techniques cannot properly deal with cases where there are multiple effect sizes available for the same relationship from the same study. Existing applications either treat these effect sizes as independent, randomly select one effect size amongst many, or create an average effect size. None of these approaches deal with the inherent dependency in effect sizes, and either leads to biased estimates or loss of information and power. An alternative technique is to use univariate three-level modeling in the two-stage approach to model these dependencies. These different strategies for dealing with dependent effect sizes in the context of MASEM have not been previously compared in a simulation study. This study aims to compare the performance of these strategies across different conditions; varying the number of studies, the number of dependent effect sizes within studies, the correlation between the dependent effect sizes, the magnitude of the path coefficient, and the between-studies variance. We examine the relative bias in parameter estimates and standard errors, coverage proportions of confidence intervals, as well as mean standard error and power as measures of efficiency. The results suggest that there is not one method that performs well across all these criteria, pointing to the need for better methods.

Random effects meta-analysis model is an important tool for integrating results from multiple independent studies. However, the standard model is based on the assumption of normal distributions for both random effects and within-study errors, making it susceptible to outlying studies. Although robust modeling using the t distribution is an appealing idea, the existing work, that explores the use of the t distribution only for random effects, involves complicated numerical integration and numerical optimization. In this article, a novel robust meta-analysis model using the t distribution is proposed (tMeta). The novelty is that the marginal distribution of the effect size in tMeta follows the t distribution, enabling that tMeta can simultaneously accommodate and detect outlying studies in a simple and adaptive manner. A simple and fast EM-type algorithm is developed for maximum likelihood estimation. Due to the mathematical tractability of the t distribution, tMeta frees from numerical integration and allows for efficient optimization. Experiments on real data demonstrate that tMeta is compared favorably with related competitors in situations involving mild outliers. Moreover, in the presence of gross outliers, while related competitors may fail, tMeta continues to perform consistently and robustly.

The prevalence of co-morbid anxiety and depression varies greatly between research studies, making it difficult to understand and estimate the magnitude of this problem. This systematic review and meta-analysis aim to provide up-to-date information on the global prevalence of co-morbid anxiety and depression in pregnant and postpartum women and to further investigate the sources of heterogeneity. Systematic searches of eight electronic databases were conducted for original studies published from inception to December 10, 2024. We selected studies that directly reported prevalence data on co-morbid anxiety and depression during the perinatal periods. We extracted data from published study reports and calculated the pooled prevalence of symptoms of co-morbid anxiety and depression. There are 122 articles involving 560,736 women from 43 different countries included in this review. The global prevalence of co-morbid anxiety and depression during the perinatal period was about 9% (95%CI 8%–10%), with approximately 9% (95%CI 8%–11%) in pregnant women and 8% (95%CI 7%–10%) in postpartum women. Prevalence varied significantly by the assessment time points, study country, study design, and the assessment tool used for anxiety and depression, while prevalence was not dependent on publication year, country income level, and COVID-19 context. No publication bias was observed for this prevalence rate. These findings suggest that approximately 1 in 10 women experience co-morbid anxiety and depression during pregnancy and postpartum. Targeted action is needed to reduce this burden.

Researchers would often like to leverage data from a collection of sources (e.g., meta-analyses of randomized trials, multi-center trials, pooled analyses of observational cohorts) to estimate causal effects in a target population of interest. However, because different data sources typically represent different underlying populations, traditional meta-analytic methods may not produce causally interpretable estimates that apply to any reasonable target population. In this article, we present the CausalMetaR R package, which implements robust and efficient methods to estimate causal effects in a given internal or external target population using multi-source data. The package includes estimators of average and subgroup treatment effects for the entire target population. To produce efficient and robust estimates of causal effects, the package implements doubly robust and non-parametric efficient estimators and supports using flexible data-adaptive (e.g., machine learning techniques) methods and cross-fitting techniques to estimate the nuisance models (e.g., the treatment model, the outcome model). We briefly review the methods, describe the key features of the package, and demonstrate how to use the package through an example. The package aims to facilitate causal analyses in the context of meta-analysis.

In small meta-analyses (e.g., up to 20 studies), the best-performing frequentist methods can yield very wide confidence intervals for the meta-analytic mean, as well as biased and imprecise estimates of the heterogeneity. We investigate the frequentist performance of alternative Bayesian methods that use the invariant Jeffreys prior. This prior has the usual Bayesian motivation, but also has a purely frequentist motivation: the resulting posterior modes correspond to the established Firth bias correction of the maximum likelihood estimator. We consider two forms of the Jeffreys prior for random-effects meta-analysis: the previously established “Jeffreys1” prior treats the heterogeneity as a nuisance parameter, whereas the “Jeffreys2” prior treats both the mean and the heterogeneity as estimands of interest. In a large simulation study, we assess the performance of both Jeffreys priors, considering different types of Bayesian estimates and intervals. We assess point and interval estimation for both the mean and the heterogeneity parameters, comparing to the best-performing frequentist methods. For small meta-analyses of binary outcomes, the Jeffreys2 prior may offer advantages over standard frequentist methods for point and interval estimation of the mean parameter. In these cases, Jeffreys2 can substantially improve efficiency while more often showing nominal frequentist coverage. However, for small meta-analyses of continuous outcomes, standard frequentist methods seem to remain the best choices. The best-performing method for estimating the heterogeneity varied according to the heterogeneity itself. Röver & Friede’s R package bayesmeta implements both Jeffreys priors. We also generalize the Jeffreys2 prior to the case of meta-regression.

Results of meta-analyses are of interest not only to researchers but often to policy-makers and other decision-makers (e.g., in education and medicine), and visualizations play an important role in communicating data and statistical evidence to the broader public. Therefore, the potential audience of meta-analytic visualizations is broad. However, the most common meta-analytic visualization – the forest plot – uses non-optimal design principles that do not align with data visualization best practices and relies on statistical knowledge and conventions not likely to be familiar to a broad audience. Previously, the Meta-Analytic Rain Cloud (MARC) plot has been shown to be an effective alternative to a forest plot when communicating the results of a small meta-analysis to education practitioners. However, the original MARC plot design was not well-suited for meta-analyses with large numbers of effect sizes as is common across the social sciences. This paper presents an extension of the MARC plot, intended for effective communication of moderate to large meta-analyses (k = 10, 20, 50, 100 studies). We discuss the design principles of the MARC plot, grounded in the data visualization and cognitive science literature. We then present the methods and results of a randomized survey experiment to evaluate the revised MARC plot in comparison to the original MARC plot, the forest plot, and a bar plot. We find that the revised MARC plot is more effective for communicating moderate to large meta-analyses to non-research audiences, offering a 0.30, 0.34, and 1.07 standard deviation improvement in chart users’ scores compared to the original MARC plot, forest plot, and bar plot, respectively.

Cannabis use is highly prevalent in people with schizophrenia and is related to adverse clinical outcomes, including relapse and hospitalization. However, the relationship between cannabis and suicide remains inconclusive. This study aimed to systematically review and meta-analyze the relationship between cannabis use and suicide-related outcomes in people with schizophrenia. A comprehensive search of Medline, Embase, and PsycINFO for cross-sectional, case-control, and longitudinal studies was conducted using search terms from database inception to November 2024 inclusive. Computation of odds ratios (ORs) and hazard ratios (HRs) was performed using random effects models with DerSimonian-Laird estimation. All studies were appraised for quality. We also evaluated heterogeneity, publication bias and performed sub-group analyses and meta-regression. Twenty-nine studies comprising 36 samples met eligibility criteria. Cannabis use was not associated with odds of suicide death or suicidal ideation but was associated with risks of suicide death (HR = 1.21, 95% CI = 1.04 – 1.40) and odds of attempted suicide (OR = 1.40, 95% CI = 1.16 – 1.68). While between-sample heterogeneity was moderate in analyses of attempted suicide (I2 = 39.6%, p = 0.03), there was no publication bias. Summary effects remained significant in most sub-groups, but just failed to reach significance in longitudinal studies of attempted suicide (OR = 1.40, 95% CI = 0.97 – 1.68) and studies investigating first episode samples (OR = 1.24, 95% CI = 0.99 – 1.55). Cannabis use is significantly associated with some, but not all, suicide-related outcomes in people with schizophrenia. More work is needed to examine potential mechanisms of significant relationships.

The number of published systematic reviews has increased over the last years, with a non-negligible proportion displaying methodological concerns. We aimed to develop and evaluate a tool to assess the reported methodological quality of medical systematic reviews. The developed tool (ReMarQ) consists of 26 dichotomous items. We applied an item response theory model to assess the difficulty and discrimination of the items and decision tree models to identify those items more capable of identifying systematic reviews with higher reported methodological quality. ReMarQ was applied to a representative sample of medical systematic reviews (excluding those published in the Cochrane Database of Systematic Reviews) to describe their methodological quality and identify associated factors. We assessed 400 systematic reviews published between 2010 and 2020, of which 196 (49.0%) included meta-analysis. The most discriminative items were (i) conducting a risk of bias assessment, (ii) having a published protocol and (iii) reporting methods for solving disagreements. More recent systematic reviews (adjusted yearly RR=1.03; 95%CI=1.02 −1.04, p<0.001) and those with meta-analysis (adjusted RR=1.34; 95%CI=1.25 −1.43, p<0.001) were associated with higher reported methodological quality. Such an association was not observed with the journal impact factor. The items most frequently fulfilled were (i) reporting search dates, (ii) reporting bibliographic sources and (iii) searching multiple electronic bibliographic databases. ReMarQ, consisting of dichotomous items and whose application does not require subject content expertise, may be important (i) in supporting an efficient quality assessment of systematic reviews and (ii) as the basis of automated processes to support that assessment.