Turkish validity and reliability of the depression scale in neurological disorder for young people with epilepsy

The aim of this study, which was carried out with a methodological design; To perform the Turkish Validity and Reliability of the Depression in Neurological Disorder Scale for Youth with Epilepsy, which was developed in English language and culture. The study was conducted with young patients aged 12-17 years with epilepsy who applied to Mersin University Faculty of Medicine Pediatric Neurology Outpatient Clinic between April 2021 and October 2021. IBM SPSS 22 and AMOS 22 programs were used to evaluate the data. To determine the validity of the scales; Content validity, criterion validity (internal and external criterion) and construct validity (exploratory and confirmatory) were conducted. The KGİ for content validity is 0.92. For criterion validity; internal criterion validity (lower-upper group comparison: r=.954, t=66,756, p=.000, and item-total correlations: .551-.865) and external criterion validity (DSM-5 level 2- Depression Scale ) were evaluated with (r=.894). In order to make the structure validity, descriptive factor analysis; "Kaiser Meyer Olkin" and "Barlett tests" (KMO:.931, Barlett's test x2:646.946, p<0.001), factor load: (.379-.806), eigenvalue analysis (6.608), explanation of total variance (55.069%), correlation matrix values (.615-.898) has been studied. In confirmatory factor analysis; factor loads of the items were x2/sd (1.262), NFI (.900), TLI (.972), CFI (.977), IFI (.978), RFI (.878), GFI (.894), AGFI (. 847), RMSEA (.052) and RMR (.047) fit values were examined. In order to determine the reliability of the scales; internal consistency reliability coefficients (cronbach alpha:.924), halving method (r:.884), item-total score correlations (.551-.865), Scoring consistency (Intraclass correlation:.939), Standard Error (SEM) ≤SS/2: 2.48≤4.5) (measurement accuracy of the scales), Base-ceiling effect analysis (Min.Fr=1.0%, Max.Fr=7.3) (scale homogeneity) and Hotelling's T2 test (F= 245.328, p=.000) ) has been made. It was found that the scale adapted to Turkish has a statistically valid and reliable structure.