MétaCan
Menu
Back to cohort
Record W4414246624 · doi:10.1111/2041-210x.70155

A pluralistic framework for measuring, interpreting and decomposing heterogeneity in meta‐analysis

2025· article· en· W4414246624 on OpenAlexafffund
Yefeng Yang, Daniel W. A. Noble, Rebecca Spake, Alistair M. Senior, Malgorzata Lagisz, Shinichi Nakagawa

Bibliographic record

VenueMethods in Ecology and Evolution · 2025
Typearticle
Languageen
FieldDecision Sciences
TopicMeta-analysis and systematic reviews
Canadian institutionsUniversity of Alberta
FundersAustralian Research CouncilCanada Excellence Research Chairs, Government of CanadaGovernment of Canada
KeywordsSpatial heterogeneityStudy heterogeneityVariation (astronomy)Variance (accounting)PopulationPopulation sizeTaxonomic rank

Abstract

fetched live from OpenAlex

Abstract Measuring heterogeneity, or inconsistency, among effect sizes is a crucial step for interpreting meta‐analytic evidence across diverse taxonomic groups and spatiotemporal contexts. However, ecologists and evolutionary biologists often interpret overall mean effects (mean population effects) as consistent across contexts, either explicitly or implicitly, without properly quantifying and interpreting heterogeneity. Here, we present a pluralistic approach that aims to quantify heterogeneity by introducing complementary metrics, each of which decomposes heterogeneity into within‐study, between‐study and between‐species (species and phylogenetic) variances. These metrics include the traditional I 2 (variance‐standardized metric), the newly derived coefficient of variation for heterogeneity ( CVH family; mean‐standardized metric), the second‐order coefficient of variation ( M family; variance–mean‐standardized metric) and their stratified variants. To demonstrate the benefits of the combined use of these measures, we synthesize heterogeneity estimates from 512 ecological and evolutionary meta‐analyses. We show that total heterogeneity (variance of true effects) is, on average, 10 times larger than statistical noise (sampling error variance), contributing to 91% of the observed variance (median I 2 = 91%). This amount of heterogeneity is nearly twice the size of the mean population effect (median CVH = 1.8 and M = 0.6), indicating substantial variation among studies within a meta‐analysis. Moreover, different effect size types yield different values of heterogeneity metrics because they are inherently influenced by statistical properties of their effect size estimators. As such, comparisons of heterogeneity across effect size types should be made with caution, albeit the proposed heterogeneity metrics are unit‐free. Our large‐scale synthesis also provides new benchmarks for the interpretation of heterogeneity and recommendations on how to quantify and report heterogeneity. New extensions for stratifying heterogeneity metrics will clarify our understanding of the generalisability, and at what level of meta‐analytic effects in ecology and evolution.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.120
metaresearch head score (Gemma)0.058
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesMetaresearch
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.458
Threshold uncertainty score0.949

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.1200.058
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0030.001
Bibliometrics0.0010.002
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.700
GPT teacher head0.614
Teacher spread0.085 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; both teacher heads agree on what is shown here.

Study designObservational
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations5
Published2025
Admission routes2
Has abstractyes

Explore more

Same venueMethods in Ecology and EvolutionSame topicMeta-analysis and systematic reviewsFrench-language works237,207