2.42.1 Power and Sample Size (DOE)

Introduction

The Power and Sample Size analysis in DOE helps determine how many experimental runs are needed to detect a meaningful effect with a specified probability, or how much statistical power a proposed experimental design can achieve.

The analysis is useful when planning an experiment to ensure that the design has sufficient runs to detect important factor effects or interactions.

You can solve any one of four quantities given the other three.

• Number of replicates - how many times to run the full design
• Number of center points - adds runs to detect curvature
• Effect size - the smallest effect you care about detecting (in the same units as your response)
• Power - probability of detecting the effect if it truly exists (typically target 0.8 or 0.9)

Supported design types

• Factorial designs

Run the Analysis

• Select Statistics: Quality Improvement: Design of Experiment from Origin menu
• Select the Power and Sample Size icon

  

Topics covered in this section: The Power and Sample Size (DOE) Dialog Box Tutorial of Power and Sample Size in DOE