# Binary or Multinomial Logistic Regression?

I wish to analyze the following :

Predictor Variable (IV): Satisfaction of sexual needs as important (4 items scale and respond based on 4 point likert scale. Sum up to get the item score.)

Response Variable (DV): Condom usagae (2 options : never or sometime)

Questions:

1. Should I use binary logistic or multinomial logistic? (some people tell me to use multinomial logistic but a book said to only use it when the DV has more than two levels, and my DV only has 2 levels - never or sometime).

2. How can I use SPSS to analyse this? I need step by step help.

The SPSS dialog box for logistic regression has three boxes:

1. Dependent : I put in Condom Usage
2. Factor(s): (I am not sure should whether I put Satisfaction of sexual needs as important here?)
3. Covariate : (I am not sure whether I should put Satisfaction of sexual needs as important here?)

I am very sorry. Maybe my question sounds silly but i really need help, as i am beginner. I couldn't get any tutor within my town.

• From what I can understand, it's not clear what is the response variable and the explanatory variable. Can you please clarify this? – suncoolsu Nov 8 '10 at 9:25
• @suncoolsu I edited the question on Mengzhen's behalf to attempt to make it clearer – Jeromy Anglim Nov 9 '10 at 7:00

Binary or Multinomial: Perhaps the following rules will simplify the choice:

• If you have only two levels to your dependent variable then you use binary logistic regression.
• If you have three or more unordered levels to your dependent variable, then you'd look at multinomial logistic regression.

A few points:

• Satisfaction with sexual needs ranges from 4 to 16 (i.e., 13 distinct values). Such a variable is typically treated as a metric predictor (i.e., in the covariate box in SPSS).
• Possibly your dependent variable is causing some confusion because as you phrase it, it is not a standard dichotomy. It sounds like a frequency item that could range from never, to occasionally, to sometimes, to often, to always, etc. However, I'm guessing that either you have explicitly collapsed categories or you have required the respondent to implicitly collapse the categories down to a binary choice. As a side note, if you did have an ordered set of frequency categories, then you might want to use a model that incorporated that order.

SPSS: I posted some links to tutorials in SPSS and R for conducting binary logistic regression.

If you're collapsing the response and it had more in it's range, such as "frequently" and "always", then you should actually be doing ordinal regression. The ordinal package in R is quite nice for this.