Utprovning av SWITCH, ett svenskt förståelighetstest för barn

Abstract

SWITCH- Swedish Intelligibility test for children, is a test that measures intelligibility of children's speech through repetition of single words. The test is computer generated and consists of 1000 word lists randomly selected for each assessment. The purpose of this study was to examine the equivalence, reliability and validity in the word lists. Word lists were tested on ten children with typical language development (N group) and ten children with speech deviation (T group). The children were audio recorded when they repeated two lists. Twenty speech therapy students assessed the intelligibility. Percentage of consonants correct (PCC) were calculated. The results of each child´s two word lists were compared to calculate the equivalence. Inter- and intraraterreliability were examined. Validity was examined by examining correlations between PCC and intelligibility, and by comparing the intelligibility of the N-group and Tgroup. The results showed that the word lists in SWITCH were equivalent and that the method was reliable and valid.