Browsing by Subject "nuorten dissosiaatio"

Objectives. The first aim of this study was to examine the factor structure of the Adolescent Dissociative Experiences Scale (A-DES) with more advanced methods than before, because the results of previous studies performed by traditional factor analysis had been inconclusive. The second aim of this study was to examine the measurement invariance of the A-DES across groupings by sex, age (younger than 16-year-olds versus at least 16-year-olds), number of friends, illicit drug use, transgender disposition and exposure to bullying at school. The third aim of this study was to create abbreviated versions of the A-DES with 5, 10 and 20 items and compare the information value of these versions to the original 30-item questionnaire. Methods. The sample (n = 4072) of this study was regional, population-based and consisted of adolescents attending lower and upper secondary schools. The average age of the participants was 16.1 years (with standard deviation of 1.5 years). Variables of the study were based on the A-DES questionnaire items, some items of the YSR-questionnaire and questions concerning illicit drug use. The factor structure of the A-DES was examined with item response theory (IRT) methods using Samejima’s Graded Response Model. Measurement invariance was examined with DIF analysis based on logistic regression analysis, likewise employing Samejima’s Graded Response Model. Abbreviated versions of the A-DES were created utilizing IRT item fit indices and the results of the DIF analysis. Results and conclusions. The factor structure of the A-DES turned out to be one-dimensional. Previous suggestions of three- and four-dimensional factor structure were not supported. The A DES was also highly measurement invariant over differences in sex, age, number of friends, illicit drug use, transgender disposition and exposure to bullying at school. The amount of dissociation among Finnish adolescents expressed by their A-DES responses was quite low compared to several other countries, which must be taken into account when using the Finnish-language questionnaire. Relative to the number of items, the information values of the abbreviated versions were comparable to the original A-DES.