Determination principles of the criterial levels of the sociological test


The article evaluates the concept of the “norm” in its relation to sociological testing. Since there are no methodological works regarding norms in sociological testing, a starting point of the analysis is a study of standardization in psychological testing. For this matter the author considered the following components of standardization studies: achievement of representativeness, statistical normalization and finding of investigated property’s levels. Based on this analysis, paper presents general model of standardization studies of sociological tests, and proposes a substitution of the term “norm” for a more appropriate one – a “criterial level”.

It is understood as a numerical value of the sociological test, which denotes a qualitatively different level of a studied characteristic. The properties of the criterial levels are specificity, quantitativeness and discriminativeness. To demonstrate the theoretical principals of the article, the work presents results of criterial levels exploring of the brief-test SCL-9-NR. The corresponding analysis is based on data from five studies: 1) online survey in Lviv (2015, N = 200); 2) online survey in Kyiv (2015, N = 200); 3) online survey in Kyiv (2016, N = 205); 4) survey of patients with psychiatric disorders in Chernihiv and Kharkiv (2016-2017, N = 217); 5) sociological monitoring “Ukrainian society” (2016, N = 1802). The exploring of the criterial levels of SCL-9-NR involved three stages: analysis of the SCL-9-NR distribution properties (Shapiro-Wilk test); analysis of psychological distress levels in various groups of respondents and determination of the criterial levels on this basis (median analysis); assessment of the discriminatory potential of SCL-9-NR (classification using decision trees).

According to the results, it is determined that an increased level of distress with the use of SCL-9-NR starts at 13 points, and the high level – at 17 points.

  1. Avanesov, V. (1982). Tests in sociological research. Moscow: Nauka [in Russian]

  2. Dembitskyi, S. (2016). Sociological tests: the essence and validation. Sotsiologiya: teoriya, metody, marketing – Sociology: theory, methods, marketing, 3, 140–155 [in Russian]

  3. Dembitskyi, S. (2017). Measurement scales VS integral indices. URL: [in Russian]

  4. Kaplan, R., Saccuzzo, D. (2009). Psychological Testing: Principles, Applications, and Issues. Belmont: Wadsworth

  5. Anastazi, A., Urbina, S. (2007). Psychological testing. Saint Petersburg: Piter [in Russian]

  6. Klain, P. (1994). A Handbook of Test Construction: Introduction to Psychometric Design. Kyiv: PAN LTD [in Russian]

  7. Dembitskyi, S. (2016). Theoretical validation in sociological research: methodology and methods. Moscow: URSS [in Russian]

  8. Campbell, D. (1980). Models of experiments in social psychology and applied research. Moscow: Progress [in Russian]

  9. Shmelev, A. (1996). Fundamentals of Psychodiagnostics. Rostov-on-Don: Feniks [in Russian]

  10. Dembitskyi, S. (2016). SCL-9-NR: a brief test for measurement of psychological distress in survey research. Sotsiologiya: teoriya, metody, marketing – Sociology: theory, methods, marketing, 1, 52–64 [in Russian]

  11. Dembitskyi, S., Yashchenko, V., Sidorov, N., Aleksandrova, S. (2016). Validation criterion of brief test SCL-9-NR. Sotsiologiya: teoriya, metody, marketing – Sociology: theory, methods, marketing, 2, 102-116 [in Russian]

  12. Dembitskyi, S. (2017). Histograms of SCL-9-NR index distributions. URL: [in Russian]

  13. James, G., Witten, D., Hastie, T., Tibshirani, R. (2016). An Introduction to Statistical Learning with Applications in R. Moscow: DMK Press [in Russian]

Full text