Introduction:
The purpose of statistical analysis in research is to identify accurate and reliable conclusions where the researcher has a great deal of sources and information. Usually, one can point to a few different methods that allow the task to be fulfilled, but each time the question arises: which one to choose?

Material and methods:
The study was conducted using a database that included 3246 patients in the Second Department of Cardiology, Silesian Medical Centre in Katowice-Ochojec in 2003–2008. We were A model in which the STROKE dependent variable was considered was subjected to statistical analysis, and the results of the analysis suggested selecting the following variables: gender, transfusion, PTCA, IVA, IVM, SVA, aneurysm and hematocrit.

Results:
The essential factors affecting the occurrence of stroke, according to logistic regression are: aneurysm, transfusion of blood components, prior treatment with PTCA and according to the classification tree: aneurysm and level of hematocrit.

Conclusions:
The results achieved by both the two statistical models complemented each other, and by combining them one is able to obtain reliable information to use as a base for the decision-making process.