The main purpose of this article is to demonstrate the advantages of statistical analysis of asymmetric relations between test items in the development of computer versions of psychodiagnostic tools. The main idea is that the statistical analysis of asymmetric relations between test items allows to describe the internal structure of the test in the form of incomplete oriented graph. The authors also attempt to identify some areas of possible applications of this analysis in psychodiagnostics and psychometrics: adaptive testing; assessment of discriminatory power of the test task; determining the reliability of the data. The authors base their arguments on analyzing results of 172 subjects tested by the Russian version of MMPI.

Keywords: Computer psychodiagnostics, adaptive testing, conjugation of variables, probability, statistical relationship, association, graphs