do I have to have a p value statistically significant to make ROC curve? p value is 0.068, can I make a ROC curve?
Yes, you can go ahead and plot your ROC curve IMO.
A p-value for what, by the way?
Thanks a lot, I made a chi square test to see if there is a statistically significant difference between my test variable and state variable, but ROC curve shows performance of the test, I think the first thing I did was wrong. Thanks a lot for answer. buy the way, in SPSS under ROC curve in a table a p value is given, is it necessary to be significant?from your explanation I understand it is not.
Hello Cemre Alan Just for the reader's attention and to see what the probability value obtained in the output of the results indicates, you can read the attached file.
I have highlighted the relevant sections. End of page 6.
Hello Cemre Alan. I think the table you are describing is the Area Under the Curve table, like the one shown at the bottom of page 4 here:
- https://case.truman.edu/files/2015/06/SPSS-Logistic-Regression.pdf
As footnote B below that table shows, the p-value is for a test of the null hypothesis that (in the population), AUC = 0.5.
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Nucleic Acids
- RNA
- Small RNA
- microRNA