I am doing similar analyses, and have same question. Could you tell me the method if you solved this problem?  ... View more

Use option MEDIAN in proc npar1way to test the difference in medians. ... View more

Without the data, it is difficult to know how the response might be affected by the 'female' and 'agegroup' variables. However, in general, what you describe can when you add or exclude a categorical variable to a model. It is known as Simpson's Paradox and it means that the within-group relationships between variables are different from the between-group relationships. There are some pictures in the Wikipedia article that show why it occurs.  You might try graphing your data in a similar fashion to see if it reveals whether Simpson's paradox is responsible for what you are seeing. ... View more

Yes, it is because it has two levels.  The number of coefficients will correspond with the number of levels and the ordering of the levels.   There are a number of good examples and explanations of estimate/contrast statements found here: http://support.sas.com/kb/24/447.html  ... View more

Yes, it is because it has two levels.  The number of coefficients will correspond with the number of levels and the ordering of the levels.   There are a number of good examples and explanations of estimate/contrast statements found here: http://support.sas.com/kb/24/447.html  ... View more

Since all your independent variables are CLASS variables, then your data indeed like a contingency table. Check PROC FREQ + TEAND option. there is an example of it in PROC FREQ's documentation. ... View more

See if this is helpful:   https://communities.sas.com/t5/SAS-Statistical-Procedures/Sample-size-calculation-for-three-groups/m-p/171262#M8906   ... View more