Meta-Analysis in Excel: Effect Sizes, Fixed vs Random Effects & Forest Plots
Combine the results of several studies step by step in Excel: mean difference, Cohen's d and log odds ratios with their variances, inverse-variance weights, the fixed-effect estimate with CI and z-test, Q, I² and τ² (DerSimonian–Laird), the random-effects model and prediction interval, a forest plot, funnel plot and Egger's test, and leave-one-out sensitivity, in 8 graded labs.
The first lesson is free — no account needed.
About this course
A meta-analysis course that pools eight synthetic studies of a fictional peer-tutoring programme step by step in a spreadsheet, for students and researchers in psychology, education, medicine and the social sciences and analysts who read systematic reviews. You turn reported results into mean differences, Cohen's d and log odds ratios with their variances, weight the studies by inverse variance and calculate the fixed-effect estimate, add its confidence interval, z-test and p-value, test heterogeneity with Cochran's Q, I² and the DerSimonian–Laird τ², switch to the random-effects model with a prediction interval, draw a forest plot as an XY chart, check for publication bias with a funnel plot and Egger's regression test, and run a leave-one-out sensitivity analysis. Synthetic studies made up for learning, not real research results. The first lab is free.
Lessons
- Effect Sizes: Mean Difference, Cohen's d and the Log Odds RatioFree
- Inverse-Variance Weights & the Fixed-Effect Estimate Pro
- Confidence Interval & z-Test of the Pooled Effect Pro
- Heterogeneity: Cochran's Q, I² and τ² (DerSimonian–Laird) Pro
- The Random-Effects Model and the Prediction Interval Pro
- The Forest Plot: Every Study and the Pooled Effect in One Chart Pro
- Publication Bias: Funnel Plot & Egger's Regression Test Pro
- Sensitivity Analysis: Leave-One-Out Pro