I recently was asked whether to report means from descriptive statistics or from the Estimated Marginal Means with SPSS GLM.
The Estimated Marginal Means in SPSS GLM tell you the mean response for each factor, adjusted for any other variables in the model. They are found in the Options button. (These are the same as the LSMeans in SAS GLM).
If all factors (aka categorical predictors) were manipulated, these factors should be independent.
Or at least they will be if you randomly assigned subjects to conditions well.
In this situation only, the estimated marginal means will be the same as the straight means you got from descriptive statistics.
If however, you have a covariate in the model that was measured, not manipulated, things are a little different. The estimated marginal means will now be adjusted for the covariate.
This, of course, is the reason for including the covariate in the model–you want to see if your factor still has an effect, beyond the effect of the covariate. You are interested in the adjusted effects in both the overall F and in the means.
In SPSS, the Estimated Marginal Means adjust for the covariate by reporting the means of Y for each level of the factor at the mean value of the covariate. You can change this default using syntax, but not through the menus.
For example, in this syntax, the EMMEANS statement will report the marginal means of Y at each level of the categorical variable X at the mean of the Covariate V.
UNIANOVA Y BY X WITH V
If instead, you wanted to evaluate the effect of X at a specific value of V, say 50, you can just change the EMMEANS statement to:
Another good reason to use syntax.
Editor’s Update: If you want to learn more about Estimated Marginal Means, how to implement and interpret them, as well as the other options in SPSS GLM, check out our workshop on Running Regressions and ANCOVAs in SPSS GLM. It’s now available in a home study version.