Users
- Usually not mathematical statisticians
- Complexities should be kept out of sight
- Friendly interface
- Focus on scientific goals, not formulas
Problems/Scenarios
- ``What if?'' structure
- Adaptability needs
- Solve for n
- Get power for given n
- Find detectable effect size
- Meaningful ``effect size'' inputs
- Fitness for use in complex problems--e.g., multifactor ANOVA
NEXT: Excel Methods
PREVIOUS: Target power
TOP: GUIs for Power and Sample Size
Russ Lenth ---
10/23/1997