I need help determining the N for my overall likelihood test and odds ratios as run via genmod in sas.
I used a cumulative logit model, which produced a likelihood ratio test for each main effect of the independent variables and interactions based on the chi-square distribution. Significant effects were further explored using follow-up custom contrasts with Wald chi-square tests. Odds ratios (OR) were used to compare the multiplicative increase (or odds) of more accurate responses according to different levels of the independent variables.
My question is how do I know the N for the likelihood ratio tests or the follow-up custom contrasts? I know that the number of observations used was 3,878, but is that the number that I need. Thanks in advance for your help.