domains / families / research / psychbull-meta-analysis-methods

psychbull-meta-analysis-methods

separate

Use when computing effect sizes and fitting the meta-analytic model for a Psychological Bulletin manuscript — effect-size metrics, random-effects vs. fixed-effect choice, dependent effect sizes (RVE / multilevel), and heterogeneity (Q, I², τ², prediction interval). Guides the core synthesis; moderators and bias diagnostics live in psychbull-moderators-and-bias.

When to use

• Computing effect sizes from coded statistics
• Choosing fixed-effect vs. random-effects models
• Handling multiple effect sizes per study
• Quantifying and interpreting heterogeneity
• Converting disparate metrics to a common scale
• Fitting random-effects or mixed-effects models
• Applying robust variance estimation (RVE)
• Reporting Q, I², τ², and prediction intervals

When not to use

• Moderator analysis
• Publication-bias diagnostics
• Primary-study reanalysis
• Causal-inference toolchain execution

Skill page

Source