Hello all! Working on my Grad study and using SPSS 20.
I have an ordinal DV being tested with multiple IV's (demographics etc)
A friend who uses SAS said I need to use a "Cumulative Logit Models for Ordinal Response."
1 - Isn't this a fancy way of saying an Ordinal Regression in SPSS?
I read that: "Ordinal logistic regression uses a cumulative logit link, setting up a series of binary comparisons between the given level or lower compared to higher levels. (Technically, the SPSS Analyze, Regression, Ordinal menu choice runs the PLUM procedure, which uses a logit link, but the logit link is used to fit a cumulative logit model. The SPSS Analyze, Generalized Linear Models, Generalized linear Models menu choice for ordinal regression assumes an ordered multinomial distribution with a cumulative logit link).
2 - And can you explain the distinction in parenthesis? Does it mean a ordered distribution of the DV & IV's? Which one is better suited?
Thank you sincerely.