LITMUS Turkish sentence repetition test: The best items, effect of scoring and diagnostic accuracy.

Abstract

PURPOSE This study aimed to examine LITMUS Turkish Sentence Repetition Test's (LITMUS-TR) diagnostic accuracy, as well as the best scoring method and most distinguishing test items. We also sought to ascertain whether age has an impact on the sensitivity and specificity. METHOD Two hundred and fifty children with typical language development (TD) between the ages of 4 and 7, as well as 44 children with developmental language disorder (DLD), took part in the study. Data was collected using TODİL, LITMUS-TR, and the pediatric family interview form. LITMUS-TR was graded using four different methods. RESULTS The performance of children with DLD in each score type was lower and the number of errors higher than those with TD. All items have excellent or acceptable item difficulty and discrimination values for binary scoring and total number of errors. LITMUS-TR's most distinctive items were complex structures with dependencies, such as syntactic movement and embedding. LITMUS-TR had high diagnostic accuracy for the whole test (0.887) and each scoring method. A separate analysis of each age group showed sensitivity and specificity above 0.80. CONCLUSIONS When employed as a supportive objective measure, LITMUS-TR was proven to be an effective diagnostic tool for DLD, with age influencing the diagnostic accuracy outcomes.