I used mean of data(66,6) and Sd(22,6). Using M+-Sd, I found some values (44 and 89,2). I coded them as 26-44=0, 45-130=1.I used Medcalc to find out cut off point and 44 point (in fact M-Sd score) was found as cut off point. But both sensivity and specificty 1.00, that is excellent. Can you explain this situation?
Dear Pamik,
However, you did not explain your data perfectly, you need to get the cut off point that lead you to distinguish between positive and negative cases. For example you went to know what is the cut off point for a certain marker for a certain disease you must perform the Receiver Operation Characteristic curve (ROC curve). MedCalc software is the very efficient to do this analysis.
Thanks Firas your answer. I have knowledges which you told me. But you are right. I didn't explain perfectly. Which cut off point methods do you advice to me at normal disribution? Using mean-+Sd or roc curve.
You should explain your aim and question explicitly. It is not understood.
In order to perform a ROC analysis, you should have a continuous score and the results of gold standard. Gold standard results are usually entered as 0 and 1s. Continuous score is another candidate for the gold standard for which we would get a cut-off point using ROC curve.
I think your situation is not related to this. You want to do something different.
Dear Mustafa,
Firstly, you don't need to mean or STD. You need cutoff point when the marker was scores or continuous. In such cases you will have several cutoff points and several sensitivities and specificities and then you should use Youdin index to identify the cutoff point. All these stages will done by the software. If you need an example I will send it to you.
First, Thanks Firas your all answers. I can look your example you told me.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis