Hello everyone!
I am trying to set up my spss-v22 to calculate omega squared. Actually I have downloaded finished .sps but spss doesn't recognize SStotal and SSgroup commands (possible other problems too, but I am dealing one by one and i believe with this I could make it run).
I have searched online and spss manual all over, but no success.
Has anyone faced and solved similar problem?
Answer by JonPeck (4671) | Feb 16, 2015 at 08:54 PM
How were you expecting SS_A, SS_TOT, and SS_ERRO to be defined? There must have been some other code that did that. However, the SS's would be scalars, and the COMPUTE command is for casewise variables.
You could use the MATRIX procedure for all this with the appropriate inputs, but you need to figure out where they come from first. It could be that these were created using OMS wrapped around some statistical procedure. That could create an appropriate dataset from a table that appears in the Viewer.
Answer by NejcR (0) | Feb 16, 2015 at 08:46 PM
Default for SPSS 22 64 bit. I was afraid about that, but since someone wrote it that way, i thought i should be correct.
However, here is part of the code..
COMPUTE ETA_A=SS_A/SS_TOT.,
COMPUTE ETA_B=SS_B/SS_TOT.
COMPUTE ETA_AB=SS_AB/SS_TOT.
COMPUTE P_ETA_A=SS_A/(SS_A+SS_ERRO).
So...should i break down formula to the pieces to calculate SS in syntax from raw data first before going to omega?
Answer by NejcR (0) | Feb 16, 2015 at 09:06 PM
Thank you for your explanation!
I was trying to fill in all the missing data, but just can't find the trouble. However, it kept telling me these lines( and couple of others) are wrong (I didn't wrote it, as I said i have downloaded finished .sps file).
Well, I will keep trying and follow your suggestions!
Greetings,
Nejc
string and numeric types are not mixed, but error says they are 6 Answers
check whether a list of variables exist 11 Answers
SCRIPT command in Batch Production 0 Answers
Independent T-Test 1 Answer
Expanding logistic regression output 8 Answers