I had a very large value for Exp(B) in SPSS binary logistic regression. What is wrong and what should I do?

The coefficients are shown on the picture

  • $\begingroup$ Which value of $B$ is "very large"? The largest one I can see is $0.703$ for "D2(1)", because it is more than twice its standard error, but there is nothing unusual or alarming about that. $\endgroup$ – whuber Apr 22 '15 at 16:29
  • 4
    $\begingroup$ Try search for "complete separation." It seems some of the predictors appear to be a constant or some zero cells were created. $\endgroup$ – Penguin_Knight Apr 22 '15 at 16:33
  • 1
    $\begingroup$ @Penguin_Knight I've created the hauck-donner-effect tag for this express purpose! $\endgroup$ – Sycorax Apr 22 '15 at 17:10

This usually indicates a problem with Quasi-Complete (categorical variables) or Complete Separation (continuous variables).

Quasi-complete: Where you have one category of a class variable x that has only one type of outcome (y=0 or y=11)

Complete Separation: some breaking point in a continuous variable where target has only one type of outcome on either side. For example, if $x>5$ then y=1 in all cases and if $x\leq 5$ then y=0 in all cases.

| cite | improve this answer | |
  • $\begingroup$ Complete separation is only when you have y=0 for x<c and y=1 for x>=c for some value of c (and 0 and 1 might be flipped) but just having a point where either above or below you only see a single type of response doesn't result in complete separation. $\endgroup$ – Dason Oct 3 '16 at 17:22
  • $\begingroup$ Right. I guess my answer was not exactly clear on that. updated. $\endgroup$ – ShainaR Oct 5 '16 at 17:46

Not the answer you're looking for? Browse other questions tagged or ask your own question.