I'm currently assisting with the analysis of a data set which, among other factors, wants to compare values obtained during correctly and incorrectly completed trials.
As a high accuracy rate results in some subjects having provided no incorrect answers I'm using SPSS' MIXED function as it seems to handle missing data (while RM ANOVA simply deletes the entire subject).
The problem is now that by averaging the dependent variable (Skin conductance response values) over correct and incorrect trials, we drastically change mean values. Can this issue be solved by weighting cases based on the number of trials averaged? If yes, is this done through the "Residual Weights" found in SPSS' MIXED procedure or through general case weighting?
Dear Rik
1. Several examples were given in :
. Landau, S. and Everitt, B. S.(2004). A Handbook of Statistical analyses using SPSS.
.Howitt, D. and Cramer, D.(2008). Introduction to SPSS.
2. The attached documents can help you in developing the analysis.
Regards,
Zuhair
- Badges
- Science topic
- Similar topics
- Correction