Characteristics of a Good Experimental Design
- Provides unbiased estimates of the factor effects and associated
uncertainties
- Enables the experimenter to detect important differences
- Includes the plan for analysis and reporting of the results
- Gives results that are easy to interpret
- Permits conclusions that have wide validity
- Shows the direction of better results
- Is as simple as possible
Source: The Statistical Approach to Design of Experiments
by Ronald D. Snee and Lynne B. Hare, Chapter 5 (pp. 71-91) in
Perspectives on Contemporary Statistics, David C. Hoaglin and
David S. Moore, editors, MAA Notes and Reports Series, 1992.
Last Modified: