Simulation-based power calculation procedure comparing classical and Bayesian MDEs across factorial designs¶

Element · All elements
Evidence · no claims cited

Description¶

The paper describes a simulation procedure for comparing the statistical power of classical and Bayesian factorial designs. The authors "repeatedly simulate factorial experiments with a variety of sample sizes and numbers of treatment arms to estimate the minimum detectable effect (MDE) for each combination," producing MDE comparisons across design complexity that quantify the sample-size savings from the Bayesian approach.

Design Implications¶

Context¶

Requirements¶

• Specification of sample sizes, numbers of treatment arms, and factor structure to simulate

Constraints¶

Target Learners¶

• Education researchers planning experiments

Target Learning Goals¶

• Estimating minimum detectable effects and required sample sizes for factorial designs

Claims¶

Related Elements¶

Examples¶

-

Key Sources¶

• Daniel Kassler, Ira Nichols-Barrer, Mariel Finucane. (2018). Beyond "Treatment versus Control": How Bayesian Analysis Makes Factorial Experiments Feasible in Education Research. Working Paper 61. Cambridge, MA: Mathematica Policy Research. https://www.mathematica.org/publications/beyond-treatment-versus-control-how-bayesian-analysis-makes-factorial-experiments-feasible