robumeta: An R-package for robust variance estimation in meta-analysis

By Zachary F. Fisher, Elizabeth Tipton. 2015

352 citations recorded by the source corpus.

Meta-regression models are commonly used to synthesize and compare effect sizes. Unfortunately, traditional meta-regression methods are ill-equipped to handle the complex and often unknown correlations among non-independent effect sizes. Robust variance estimation (RVE) is a recently proposed meta-analytic method for dealing with dependent effect sizes. The robumeta package provides functions for performing robust variance meta-regression using both large and small sample RVE estimators under various weighting schemes. These methods are distribution free and provide valid point estimates, standard errors and hypothesis tests even when the degree and structure of dependence between effect sizes is unknown.

Subfields: Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/b6078c21aafa2e52a64197a3fcd8a0265a73557c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

A Meta-analysis of the Impact of Social and Emotional Learning Interventions on Teachers’ Burnout Symptoms (2021, methodology): “…effects from the same study are assumed to be independent and averaged (e.g., Rosenthal, 1991), RVE estimates the overall effect by taking into account the fact thatmost of the effect sizes pooled from the studies are correlated and based on the same study participants (Fisher & Tipton, 2015).”
A Meta-analysis of the Longitudinal Relationship Between Academic Self-Concept and Academic Achievement (2021, methodology): “The user-specified value must be between 0 and 1 when using the correlation effect model (Fisher & Tipton, 2015).”
Student Motivation and Associated Outcomes: A Meta-Analysis From Self-Determination Theory (2021, methodology): “Meta-Analytic Procedures All meta-analytic calculations were conducted using the R software, specifically using the robumeta package (Fisher & Tipton, 2015).”
When media become the mirror: a meta-analysis on media and body image (2020, methodology): “The overall and moderator analyses were perfumed in R software using packages “robumeta” (Fisher & Tipton, 2015).”
Effectiveness of spelling interventions for learners with dyslexia: A meta-analysis and systematic review (2020, methodology): “We used the packages compute.es (Del Re, 2013), robumeta (Fisher & Tipton, 2015), metafor (Viechtbauer, 2010), clubSandwich (Pustejovsky, 2016) and the Shiny application weightr that implements the weight function model published by Vevea and Woods (2005).”
Small-Sample Adjustments for Tests of Moderators and Model Fit Using Robust Variance Estimation in Meta-Regression (2015, methodology): “Software implementations of RVE are available in both R (robumeta package; Fisher & Tipton, 2014) and Stata (robumeta macro; Hedberg, 2011; Tanner-Smith & Tipton, 2014) and to a limited degree in SPSS (see Tanner-Smith & Tipton, 2014).”
The Relation Between Family Socioeconomic Status and Academic Achievement in China: A Meta-analysis (2019, methodology): “used the ROBUMETA package in R to examine the overall weighted correlation coefficients and potential moderators (Fisher and Tipton 2015).”
The Effects of Teacher Training on Teachers’ Family-Engagement Practices, Attitudes, and Knowledge: A Meta-analysis (2019, methodology): “Therefore, as studies included in the current analyses frequently provided more than one effect size estimate (i.e., a correlated estimation error), the “correlated effects” weighting scheme was used to determine approximate inverse variance weights (Fisher & Tipton, 2015, p. 4).”
Influencing Climate Change Attitudes in the United States: A Systematic Review and Meta-Analysis (2020, methodology): “Relying on the robumeta package (Fisher & Tipton, 2015) in R, we used RVE in the current meta-analysis because it allowed us to utilize the maximum amount of information from the included studies (e.g., using all attitude measures).”