1
$\begingroup$

Prefacing this by admitting that I don't really know what I'm doing, so please use gentle and untechnical terms when explaining. Thank you.

I used half my data to create a prediction using my B values in the table I got from performing Logistic Regression. I got an equation looking like this:

constant+(var1*-1.23)+(var2*-3.45)+(var3*1.23)+(var4*3.45)+(var5*6.78)

I plugged this into 'compute variable) on SPSS and got a bunch of new variables:

PREDIC_ which is the result of the equation PGROUP_ which is whether the equation is more or less than my cut off FILTER which seems to be saying 'Selected' based on whether the PGROUP is Yes and 'Not Selected' based on it being No (I suspect that this might be what is tripping me up) Predicted which is just a number and I don't know what it relates to.

So I go ahead and split my file, select my second half of my cases, and attempt to run cross tabs and I get no results and a warning that my crosstabulation of Predicted group * DV ix empty.

I know I am looking for the reduction or increase in specificity, sensitivity, and total accuracy, but I can't seem to make SPSS do the output and I've looked for tutorials but can't find one that makes sense to me using SPSS, Binary Logistic Regression, and that cross-validated with only the remaining half of the cases.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.