Analyze Factorial Design

Two-Level Split-Plot Designs
Estimated Effects and Coefficients Table - S and R-Sq Values

  

Use the S and R2 statistics to determine how much of the variation in the response data is explained by the model.

·    S is the estimated standard deviation of subplot error in the model.

·    S (WP) is the estimated standard deviation of error among whole plots.

·    R2 (SP) is the proportion of variation among subplots (within whole plots) accounted for by the subplot model.

·    R2 (WP) is the proportion of whole plot variation explained by the hard-to-change model.

Example Output

Model Summary

 

      S  R-sq(SP)     S(WP)  R-sq(WP)

2.55865    69.72%  0.822598    86.02%

Interpretation

For the water resistance data, S is 2.55865, R2 (SP) is 69.72%, S (WP) is 0.822598, , and  R2 (SP) is 86.02%.