AP class 5. s05.ap.uber - basic task analysis ~2~ question contrast

Stats with Gang
1

Hello,

i was just looking at this example AP class 5. s05.ap.uber - basic task analysis ~2~ on the website, and I wonder why to create the mean between 2 regressors we are using 0.5* each regressor (see below) ; and not 1* vis + 1* aud, Would the results be different with that?

-gltsym 'SYM: 0.5*vis +0.5*aud' 
-glt_label 2 mean.VA

Thanks

2

Howdy-

The stat will be the same, but the idea is that the mean of X and Y is (X+Y)/2, which is equivalent to X/2+Y/2 or 0.5*X + 0.5*Y, and the last representation there is easiest to implement in that fun R-style syntax for GLMs. With 3 items, the mean would be (X+Y+Z)/3, which would likely be translated to 0.333*X + 0.333*Y + 0.333*Z.

Does that clarify the syntax?

-pt

3

it absolutely does. Thank you very much!