A Keilmann1, G Moein, H Schöler. 1. Schwerpunkt Kommunikationsstörungen der HNO-Klinik, Universitätsmedizin der Johannes Gutenberg-Universität, Langenbeckstr. 1, 55131, Mainz, Deutschland. annerose.keilmann@unimedizin-mainz.de
Abstract
BACKGROUND: The German language development test for 3- to 5-year-olds (SETK 3-5) as well as its short form, the language screening for pre-school children (SSV), are widely used tests for language impairment in German mother-tongue children. However, data published on validation are sparse. We investigated to what extent those children who demonstrated language impairment in clinical assessment were also detected by the SETK 3-5 and SSV tests. MATERIAL AND METHODS: A group of 201 children aged 4-5 years was tested using the SETK 3-5, in addition to which assessments in phonology, active vocabulary, grammar, receptive language abilities, and the recall of nonsense syllables and sentences were conducted. Correlation analyses were performed and the clinical assessment based on the aforementioned tests was compared to the results of the SETK 3-5 and the SSV. RESULTS: Raw values of the test results were significantly correlated on different levels of correlation. Those of comparable subtests were high. The SETK 3-5 had good specificity but sensitivity of only 71.9%. CONCLUSION: Children with language impairment are not always detected using the SETK 3-5 test. Standardization with a sufficient number of children should be carried out.
BACKGROUND: The German language development test for 3- to 5-year-olds (SETK 3-5) as well as its short form, the language screening for pre-school children (SSV), are widely used tests for language impairment in German mother-tongue children. However, data published on validation are sparse. We investigated to what extent those children who demonstrated language impairment in clinical assessment were also detected by the SETK 3-5 and SSV tests. MATERIAL AND METHODS: A group of 201 children aged 4-5 years was tested using the SETK 3-5, in addition to which assessments in phonology, active vocabulary, grammar, receptive language abilities, and the recall of nonsense syllables and sentences were conducted. Correlation analyses were performed and the clinical assessment based on the aforementioned tests was compared to the results of the SETK 3-5 and the SSV. RESULTS: Raw values of the test results were significantly correlated on different levels of correlation. Those of comparable subtests were high. The SETK 3-5 had good specificity but sensitivity of only 71.9%. CONCLUSION:Children with language impairment are not always detected using the SETK 3-5 test. Standardization with a sufficient number of children should be carried out.