I have a question about the estimated marginal means in MANOVA SPSS v.27. We used MANOVA with four dependent variables (e.g. days of poor mental health, stress) and six fixed factors (i.e., year, gender, race, sexual identity, international students, disability). All fixed factors were coded 1, -1. My understanding is that the estimated marginal means would differ from the sample means because they are adjusted for other variables in the model. However, in our output, the estimated marginal means for each level of each factor seem very different from the sample means. For instance, for days of poor mental health, the overall mean is 7.24 (SD = 8.25), but the estimated marginal means for males and females are 9.29 (SE = .39) and 11.70 (SE = .36), respectively, and the estimated marginal means of each level of other factors are also higher than the overall mean. We are having trouble figuring out how these estimated marginal means are calculated. Thank you very much for your help.