Adaptation of child somatization inventory-8 (CSI-8) into Turkish and investigation of its relationship with demographic variables

The purpose of this study was to adapt the Children's Somatic Symptoms Inventory–8 (CSSI-8) into Turkish, examine its relationship with demographic variables, and assess its validity and reliability. The research employed a correlational survey model combined with a scale adaptation process within the framework of quantitative research methods. The sample was determined using the simple random sampling technique, one of the probabilistic sampling methods, based on a 95% confidence interval and 5% margin of error, in accordance with the sample size determination table for large populations. The study group consisted of 386 children aged 8–14 years who were enrolled in primary and secondary schools in the Karaman city center and had applied to Karaman Training and Research Hospital during the 2024–2025 academic year. Data were collected using a Personal Information Form and the Children's Somatization Inventory-8 (CSI-8) – Turkish version. The Turkish adaptation of the scale was completed through a linguistic and conceptual translation process, followed by expert review for finalization. To determine the scale's validity and reliability, item analysis, Cronbach's alpha internal consistency coefficient, item-total correlations, split-half reliability (Spearman-Brown and Guttman coefficients), test–retest reliability, and exploratory and confirmatory factor analyses were conducted. Additionally, t-tests and ANOVA were used to examine differences according to demographic variables. The Exploratory Factor Analysis (EFA) revealed a single-factor structure explaining 62% of the total variance. Confirmatory Factor Analysis (CFA) results indicated acceptable model fit indices (CMIN/df = 2.087; RMSEA = 0.053; CFI = 0.990), with factor loadings ranging from 0.69 to 0.80. The internal consistency coefficient was 0.914, the Spearman-Brown coefficient was 0.914, and the Guttman Split-Half coefficient was 0.912. The test–retest analysis showed a highly significant correlation between the first and second administrations (r = 0.907; p <.001). Item–total correlations ranged between 0.656 and 0.764, and the 27% upper-lower group comparisons further supported the discriminative power of the items (t =-44.57; p <.001). According to the findings obtained regarding demographic variables, statistically significant differences were identified in age, chronic illness/psychiatric diagnosis, family structure, and income level. Specifically, children aged 8 years had significantly lower somatization scores than those aged 10, 11, 12, 13, and 14 years, and those aged 9 years scored significantly lower than the 13- and 14-year-old groups. Furthermore, 12-year-olds demonstrated lower somatization scores compared to 13-year-olds. Children from extended families and those with chronic illnesses exhibited higher levels of somatization. In addition, children from high-income families reported higher somatization scores compared to those from low-income families. Conversely, no statistically significant differences were found in somatization levels based on gender, number of siblings, or parental education level. In conclusion, the findings indicate that the Turkish version of the CSSI-8 is a valid and reliable instrument for assessing somatic symptoms in children. Key Words: Children, somatization, validity, reliability.