Analysis of covariance (ANCOVA)

#1
In my research in literature, I read that categorical covariates (eg. sex, race) can be adjusted in the analysis of covariance (ANCOVA), but the categorical variable has to first be recoded to contrasts. If there are n categories, you can create n-1 contrasts. How to create n-1 contrasts in the SPSS for covariates sex?

Thanks for the help
 
#2
Hi Cleliani,

what is the aim of your study? You have a continuous outcome and you want to investigate the influence of sex and race on this outcome? Usually the standard contrasts are fine: The software uses one baseline level for each predictor and compares all other levels of this variable with the baseline level. In the case of sex, no other options make sense. If you want to do other comparisons (e.g., comparing all races from one continent to all of another continent), you have to formulate this in terms of contrasts.

Cheers