Overview of Experimental Design

Study note for Design of Experiments (DoE)
Show Details

Experimental Design or Design of Experiments (DoE)

1. Key concepts:
   
   • Effect size
   • Fixed effects/Random effects
     
   • Power
   • Sample size
   • Population
   • Bias/Error
     
   • Randomization
   • Experimental Unit
   • Blocking
   • Stratification
     
   • Replication
     
   • Blinding/Masking
   • Factors
   • Levels
   • Confounding factor
     
   • Treatment/Control, Placebo
   • Parallel design
   • Balanced design
     
2. Design:
   
   • Factorial design
   • Row-Column design
   • Crossover design
     
3. Models:
4. Steps for a good design
   
   1. Define the problem
   2. Determinate the primary/secondary endpoints or dependent variables
   3. Determinate the potential independent variables
   4. Determinate the levels and combination of each independent categorical variables
   5. Determinate the sample size based on the assumed effects, model, and power
   6. Design optimization
   7. Model specification
   8. Data collection
   9. Data clean-up
   10. Data verfication
       
5. Experimental Design in clinical trial
   
   • Phase-I design
   • Phase-II design
   • Phase-III design
   • Phase-IV design
     
6.