Software Implementation of "Decision Support System for Psychological and Pedagogical Testing"

852

Abstract

We present a software implementation of decision support system for psychological and pedagogical testing. The proposed testing technology is based on use of Markov nets as a model describing the dynamics of tests completion, parameters of which may be identified from experimental data. The system is focused on testing readiness for school in preschool children. The main advantages of the system are possibility of probabilistic assessment of subject's affiliation to one of the diagnostic groups and computation of the differentiating power for tests not being presented. Based on these estimates, an expert conducting the test will be able to choose tests order close to the optimal, which will allow to differentiate the most likely diagnosis in the shortest time. Estimation of the diagnosis probability and prognosis is done using the Bayesian approach.

General Information

Keywords: Markov model, psychological testing, decision support system

Journal rubric: Psychological Diagnostics

Article type: scientific article

For citation: Dumin P.N. Software Implementation of "Decision Support System for Psychological and Pedagogical Testing". Psikhologicheskaya nauka i obrazovanie = Psychological Science and Education, 2012. Vol. 17, no. 5, pp. 50–57. (In Russ., аbstr. in Engl.)

References

  1. Bardina R. I i dr. Diagnostika umstvennogo razvitija detej starshego doshkol'nogo vozrasta (ot 5 do 7 let). M., 1996.
  2. Kramer G. Matematicheskie metody statistiki. M., 1976.
  3. Kuravskij L. S., Baranov S. N., Kornienko P. A. Obuchaemye mnogofaktornye seti Markova i ih primenenie
    dlja issledovanija psihologicheskih harakteristik // Nejrokomp'jutery: razrabotka i primenenie.
    2005. № 12.
  4. Kuravskij L. S., Margolis A. A., Jur'ev G. A., Marmaljuk P. A. Koncepcija sistemy podderzhki prinjatija reshenij dlja psihologicheskogo testirovanija // Psihologicheskaja nauka i obrazovanie. 2012. № 1.

Information About the Authors

Pavel N. Dumin, PhD in Physics and Matematics, head of laboratory of quantitative psychology, faculty of information technologies, Moscow State University of Psychology and Education (MSUPE), Moscow, Russia, ORCID: https://orcid.org/0000-0001-9122-252X, e-mail: duminpn@gmail.com

Metrics

Views

Total: 1153
Previous month: 10
Current month: 8

Downloads

Total: 852
Previous month: 2
Current month: 2