What statistical models does this skill support?

It provides comprehensive guidance for both fixed-effect and random-effects models, including the Knapp-Hartung adjustment and the calculation of prediction intervals.

Does it help interpret heterogeneity?

Absolutely. It provides thresholds and interpretive context for I², τ², and H² statistics, helping you move beyond simple p-values to understand clinical significance.

Is this skill compliant with research standards?

Yes, the guidance is specifically aligned with the Cochrane Handbook for Systematic Reviews and the PRISMA reporting statement to ensure academic and clinical rigor.

Can this skill help with R programming for meta-analysis?

Yes, it includes standardized code snippets for popular R packages like 'meta' and 'metafor' to help you perform analysis, create forest plots, and run diagnostics.

How does it handle publication bias?

It offers detailed methodologies for visual assessment via funnel plots and provides instructions for statistical tests such as Egger's, Peters', and Begg's rank tests.

Pairwise Meta-Analysis Methodology

Name: Pairwise Meta-Analysis Methodology
Author: choxos

bychoxos

数据科学与机器学习

Provides rigorous methodological guidance for conducting and reviewing pairwise meta-analyses following Cochrane and PRISMA standards.

关于

This skill equips Claude with deep expertise in pairwise meta-analysis (MA) methodology, enabling it to assist researchers and data scientists in planning, executing, and interpreting evidence synthesis. It covers critical decision frameworks for fixed vs. random effects models, statistical assessments for heterogeneity (I², τ²), and rigorous techniques for detecting publication bias and conducting sensitivity analyses. Whether you are writing R code using the 'meta' or 'metafor' packages or interpreting complex forest plots, this skill ensures adherence to gold-standard academic reporting guidelines like PRISMA.

主要功能

Decision frameworks for choosing between fixed-effect and random-effects models
Implementation of sensitivity analyses including leave-one-out and GOSH analysis
Publication bias assessment via funnel plots and tests like Egger’s and Peters’
PRISMA-compliant reporting checklists and standardized R code snippets
Statistical interpretation of heterogeneity measures including Cochran's Q, I², and τ²
0 GitHub stars

使用场景

Reviewing and debugging R code for meta-analysis using the 'meta' or 'metafor' packages
Interpreting conflicting meta-analysis results and investigating sources of heterogeneity
Designing a meta-analysis protocol and selecting appropriate statistical models

关于

主要功能

Decision frameworks for choosing between fixed-effect and random-effects models
Implementation of sensitivity analyses including leave-one-out and GOSH analysis
Publication bias assessment via funnel plots and tests like Egger’s and Peters’
PRISMA-compliant reporting checklists and standardized R code snippets
Statistical interpretation of heterogeneity measures including Cochran's Q, I², and τ²
0 GitHub stars

使用场景

Reviewing and debugging R code for meta-analysis using the 'meta' or 'metafor' packages
Interpreting conflicting meta-analysis results and investigating sources of heterogeneity
Designing a meta-analysis protocol and selecting appropriate statistical models