2 Flashcards
Factorial Experiment Design
more than one factor that could affect outcome.
What is the “nxn design” and combination: testing for distance traveled when 3 different golf balls are hit by 4 different golf clubs.
4x3 design, 12 combinations.
Difference between observational and experimental study?
Observational randomly selects subjects to observe. Experimental randomly assign subjects to treatments.
Treatment variability (MST and SST) is broken down into 3 things:
- main effect of Factor A
- main effect of Factor B
- Interaction between Factors A & B.
Factor Interaction
test whether the factors combine to affect the response.
Factor Main Effect
Test whether each factor affects the response separately.
The null hypothesis for the overall experiment
All treatments/conditions are equal.
Null hypothesis for interaction.
Factors A and B do not interact to affect the response.
The null hypothesis for the main effect for factor A.
The mean response is the same across all levels of Factor A.
The null hypothesis for the main effect for factor B.
The mean response is the same across all levels of Factor B.
What are levels of a factor?
the possible treatments of a factor in a factorial experiment.
Replications
rows per sample; =n/(a*b)
How to calculate MST given MS(A), MS(B), and MS(AB)?
sum all three.
What is a balanced design?
An experiment with replication which also has equal sample sizes among treatments.
Ex: 4x2 design and each combination gets 4 data points. The complete chart would have 32 data points. Sample size = # of replications.
