“Error” estimation of classification made by means of a neural network for the analysis of socio-psychological characteristics of students at risk for drug addiction

Abstract

Background. Presented study shows the potential of using nonlinear analysis algorithms as an opportunity for a comprehensive study of the socio-psychological characteristics of students who are and are not assigned to the risk groups for drug addiction when processing the large data sets.

Objectives. The study had its purpose to compare linear and nonlinear algorithms for assigning students to the risk group for narcotic drug use and to determine by longitudinal comparative analysis which of the algorithms has better prognostic potential.

Study Participants. The study involved 27,790 high school students (Mage = 16.2; SD = 2.03), from this sample, 11,786 students participated in the longitudinal study in 2020 and 2021.

Methods. Organizational method: comprehensive. Empirical method: psychodiagnostic — socio-psychological testing (SPT). The study was carried out using the Unified Methodology of Socio-Psychological Testing (EM SPT-19). Data processing methods: quantitative analysis — neural network method, comparative analysis, contingency analysis, discriminant analysis. Interpretive method: structural. Software: “Statistica” statistical analysis software package, “Loginom” analytical low-code platform.

Results. The study allowed to select errors of positive and negative attribution to the risk group made by using the neural network model. The analysis revealed that erroneous assignment to the risk group is associated with homogeneity of the matched groups on all scales of risk factors, while such homogeneity is not identified by the protection factors. Comparative analysis of data from students positively and negatively assigned to the risk group found significant differences only in anxiety and frustration scales. It is interesting fact that anxiety and frustration also turn out to be the scales that, according to discriminant analysis, do not have a distinguishing ability when predicting the assignment of a student to a risk group.

Conclusions. Classification implemented by means of the neural network model has an advantage over the linear algorithm, associated with consideration of intra-scale relationships and greater stability over time in the identified risk/norm groups. The correctness of the neural network solution is confirmed by the results of discriminant analysis. The study demonstrated that social desirability of answers, which is the most common reason for recognizing respondents’ answers as unreliable, does not play a significant role in classification using a neural network. The problematized categories of “anxiety” and “frustration” require further analysis concerning their role in forming a sample of students at risk.

References

Aptikieva, L.R. (2019). Differences between “at-risk” adolescents and typical adolescents: psychological and pedagogical aspect. Vestnik Orenburgskogo gosudarstvennogo universiteta = Bulletin of the Orenburg State University, 3(221), 6–14. (In Russ.). https://doi.org/10.25198/1814-6457-221-6

Artishcheva, L.V., Abitov, I.R., Nasibullov, K.I. (2022). Validity of neural network tools for measuring the state of students’ involvement in the educational process in the digital environment. In Psychology of mental states: collection of materials of the XV International Scientific and Practical Conference. (pp. 16–21). Kazan: Kazan University Publishing House Publ. (In Russ.)

Akhmed, S.Kh., Skorodumov, S.V. (2020). Using neural network approaches in diagnosing diseases. Modelirovaniye i analiz dannykh = Modeling and Data Analysis, (2), 49–61. (In Russ.)

Berebin, M.A., Pashkov, S.V. (2006). Experience in the use of artificial neural networks for the purposes of differential diagnosis and prognosis of mental adaptation disorders. Vestnik Yuzhno-Ural’skogo gosudarstvennogo universiteta. Seriya: Komp’yuternyye tekhnologii, upravleniye, radioelektronika = Bulletin of the South Ural State University. Series: Computer Technologies, Control, Radio Electronics, (14), 41–45. (In Russ.)

Borisova, E.V., Khachkiev, V.V. (2018). Application of artificial neural networks for solving problems of psychodiagnostics. In Science, education, society: trends and development prospects: materials of the XII International. scientific-practical conf. (pp. 124–125). Cheboksary: Center for Scientific Cooperation “Interactive Plus” Publ. (In Russ.)

Bulycheva, E.V. (2023). The role of adolescents’ relationships with the social environment as a protective factor against risky behavior. National Journal of Psychology, 1(49), 88–101. https://doi.org/10.11621/npj.2023.0108

Vasilyeva, E.V. (2017). Diagnosis of risk factors for computer addiction in adolescents. Gaudeamus, 16(1), 69–75. (In Russ.)

Gazya, G.V., Eskov, V.V. (2022). Artificial neural networks in assessing age-related changes. Vestnik novykh meditsinskikh tekhnologiy = Bulletin of New Medical Technologies, (1), 101–105. (In Russ.)

Gusev, A.L., Okunev, A.A. (2017). Construction of a “successful” predictive neural network. In Artificial intelligence in solving current social and economic problems of the 21st century: a collection of articles based on materials from the Second All-Russian. scientific-practical conf. Perm. (In Russ.)

Dakhin, A.N., Semenov, N.G., Yaroslavtseva, N.V., Ermolaev, S.Yu. (2020). Pedagogical technologies and neural networks. Shkol’nyye tekhnologii = School Technologies, (2), 28–33. (In Russ.)

Dzhakobia, A.Z., Kutashov, V.A., Khabarova, T.Yu. (2017). Psychological characteristics of persons prone to chemical dependence. Tsentral’nyy nauchnyy vestnik = Central Scientific Bulletin, 2(2), 19. (In Russ.)

Dyakova, V.V. (2022). Features of the analysis of sociological research data using loginom: one-dimensional distribution. IN SITU, (6), 8–11. (In Russ.)

Kurbanova, Z.S., Ismailova, N.P. (2023). Neural networks in the context of digitalization of education and science. Mir nauki, kul’tury, obrazovaniya = World of Science, Culture, Education, 3(100), 309–311. (In Russ.)

Lichko, A.E. (2010). Psychopathy and character accentuations in adolescents. St. Petersburg: Peter Publ. (In Russ.)

Lyakhov, A.L., Aleshin, S.P. (2010). Artificial neural network as a measuring tool for model adequacy with an adaptive accuracy class. Matematicheskiye mashiny i sistemy = Mathematical Machines and Systems, (1), 61–66. (In Russ.)

Nadezhdin, A.V., Kolgashkin, A.Yu., Tetenova, E.Yu., Phyodorof, M.V., Lenkov, P.G. (2011). Interactive test for rapid diagnosis of drug addiction. Narkologiya =Narcology, 111(3), 58–63. (In Russ.)

Nasledov, A.D. (2011). Structural modeling of causal hypotheses: a study of pedagogical stereotypes in assessing junior schoolchildren. Vestnik Sankt-Peterburgskogo universiteta. Sotsiologiya = Bulletin of St. Petersburg University. Sociology, (1), 305–313. (In Russ.)

Petrov, A.N., Ivanova, G.F., Slavutskaya, E.V. (2018). Training a neural network as a tool for system analysis of multidimensional psychodiagnostic data. Vestnik Chuvashskogo universiteta = Bulletin of the Chuvash University, (1), 162–168. (In Russ.)

Podzorova, M.I., Ptitsyna, I.V., Bakhtiyarova, O.N. (2022). Neural network as one of the promising areas of artificial intelligence. Modern European Researches, (1), 169–176. (In Russ.)

Slavutskaya, E.V., Abrukov, V.S., Slavutsky, L.A. (2019). Simple neural network algorithms for assessing latent connections of psychological characteristics of younger adolescents. Eksperimental’naya psikhologiya = Experimental Psychology, (2), 131–144. (In Russ.)

Slavutskaya, E.V., Vostretsova, N.S. (2019). A neural network as a tool for analyzing the intellectual and emotional sphere of preschool children. Kazanskiy pedagogicheskiy zhurnal = Kazan Pedagogical Journal, (4), 61–65. (In Russ.)

Sologub, G.B., Pukhov, V.A. (2023). Problems of classification of natural language texts using classical machine learning methods. Modelirovaniye i analiz dannykh = Modeling and Data Analysis, (2), 64–76. (In Russ.)

Sorokin, A.S., Bondarev, V.Yu., Krotova, E.L. (2016). Artificial neural network as a means and method of statistical data processing. Vestnik UrFO. Bezopasnost’ v informatsionnoy sfere = Bulletin of the Urals Federal District Security in the Information Sphere, (2), 19–22. (In Russ.)

Tarasova, S.Yu. (2023). Psychological risks of adolescents studying under conditions of increased academic load. Bulletin of Moscow University. Series 14. Psychology, 46(1), 280–302. (In Russ.). https://doi.org/10.11621/vsp.2023.01.12

Telepova, N.N. (2015). Diagnosis of addictive behavior: integrated test. Vestnik Moskovskogo gorodskogo pedagogicheskogo universiteta. Seriya: Pedagogika i psikhologiya = Bulletin of the Moscow City Pedagogical University. Series: Pedagogy and Psychology, 1(31), 47–58. (In Russ.)

Tkachenko, A.L., Burilichev, B.V. (2021). Analytical solutions of the low-code platform Loginom for working with Big data. Vestnik Kaluzhskogo universiteta = Bulletin of Kaluga University, (3), 46–49. (In Russ.)

Fedorova, P.N., Ivanova, G.F., Slavutskaya, E.V. (2018). Neural network: selective and statistical assessment of intrasystem connections of heterogeneous psychodiagnostic data. Vestnik Chuvashskogo universiteta = Bulletin of the Chuvash University, (3), 235–242. (In Russ.)

Shumkov, E.A., Pozdnyakov, E.A. (2020). Processing tests using neural networks. Elektronnyy setevoy politematicheskiy zhurnal “Nauchnyye trudy KubGTU” = Electronic network polythematic journal “Scientific works of KubSTU”, (5), 22–32. (In Russ.)

Yakovlev, A.N., Pashkevich, N.V., Pajitnykh, D.V., Tkachev, A.A., Brodyansky, V.M., Chuprova, N.A., Vitchinkina, V.I. (2017). Search for genetic and psychological markers of drug use risk among adolescents with addictive behavior within the framework of primary prevention: preliminary results of a pilot study in Lipetsk. Sibirskiy vestnik psikhiatrii i narkologii = Siberian Bulletin of Psychiatry and Narcology, (2), 5–11. (In Russ.)

Bityutskaya, E.V., Gasanov, E.E., Khazova, K.V., Patrashkin, N.A. (2024). Classifying the Perception of Difficult Life Tasks: Machine Learning and/or Modeling of Logical Processes. Psychology in Russia: State of the Art, 17(1), 00–00. https://doi.org/10.11621/pir.2024.0205

Epskamp, S., Rhemtulla, M., Borsboom, D. (2017). Generalized network psychometrics: Combining network and latent variable models. Psychometrika, (82), 904–927. https://doi.org/10.1007/s11336-017-9557-x.

Hudson, C.J. (1990). Anxiety disorders and substance abuse. Journal of Chemical Dependency Treatment, 3(2), 119–138.

Kranzler, H.R., Liebowitz, N.R. (1988). Anxiety and depression in substance abuse: Clinical implications. Medical Clinics of North America, 72(4), 867–885.