# newbie help needed - logsitic regression interpretation

#### nbellinger

##### New Member
Hello, complete newbie here - to both stats and SPSS.

I am completing a thesis and am running a binary logistic regression in SPSS with several independent variables.

What I need help with is interpreting the results. I know that the "sig" column represents my Rsquare, which is the probability. I don't need to know the stats in and out, just need to be able to intelligently interpret what is or is not statistically significant.

What I don't understand is basically everything else.
- In Block 0 - Beginning block, what does it mean when it says "variables not in the equation"?
- In Block 1: Method = Enter, under the model summary table, what does it mean when it gives a message stating "Estimating terminated at iteration 20 because maximum iterations have been reached. Final solution cannot be found"
- In Block 1: Method = Enter, under the variables in equation table, what do Wald and Exp(B) columns mean?

Thanks for any help!

#### CB

##### Super Moderator
Hiya!

What I need help with is interpreting the results. I know that the "sig" column represents my Rsquare, which is the probability.
I'm afraid not - there is no R square in binary logistic regression, it's not an OLS technique (although it will spit out some pseudo-R squares, such as Cox/Snell's and Nagelkerke). The "sig" column represent a p value for whatever coefficient or combination of coefficients you're looking at (I'll avoid getting into a discussion of what p values really mean! )

- In Block 0 - Beginning block, what does it mean when it says "variables not in the equation"?
Just that - in block zero, IV's haven't been entered yet, so are listed as not entered in the equation.

- In Block 1: Method = Enter, under the model summary table, what does it mean when it gives a message stating "Estimating terminated at iteration 20 because maximum iterations have been reached. Final solution cannot be found"
Likelihood figures etc are determined by an iterative computation procedure, for which a limit is set so as to prevent, well, crashing the mainframe. You can change these when setting up the regression - try again with a greater iteration limit (e.g. 50, as long as you're on a decent computer system).

- In Block 1: Method = Enter, under the variables in equation table, what do Wald and Exp(B) columns mean?
The Wald statistic is used to compute the statistical significance of a coefficient or combination of coefficients - you're probably more interested in the p value it's associated with. I'm not sure about the Exp(B) column I'm afraid.

Hope that's some help. It'd probably be a good idea to get your hands on a SPSS manual (s/a the SPSS survival guide) to help you through all this stuff - it's real confusing at first!