T Risse1, C Kiese-Himmel. 1. Abt. Phoniatrie/Pädaudiologie, Universitätsmedizin Göttingen, Robert-Koch-Str. 40, 37075, Göttingen, Deutschland.
Abstract
BACKGROUND: The Mottier test to examine phonological processing is a widespread, frequently used instrument in clinical diagnostics. However, data on its validity and reliability are scarce. The reference scores for non-schoolchildren in 1981 (collected with small samples) have become outdated. METHOD: Three hundred and eight children (154 boys, 154 girls) from the German federal states of Lower Saxony and North Rhine-Westphalia were examined with the Mottier test (4-year-olds: n=86; 5-year-olds: n=126; 6-year-olds: n=96). RESULTS: The average item difficulty P for all children was 45.5 (range 5.8-90.6). On average, the items were easier to perform with increasing age (4-year-olds: P=37.3; 5-year-olds: P=46.5; 6-year-olds: P=52.7). The average item discrimination indices r(it) were 0.49 for 6-year-olds and 0.91 for 4-year-olds. The mean test score for all children was 13.76 (SD 4.35). There was a shift for the worse compared with the mean test scores from 1981 (4-year-olds: 11.19, SD 3.93; 5-year-olds: 13.95, SD 4.02; 6-year-olds: 15.81, SD 3.97). To determine reliability, the internal consistency (Cronbachs alpha=0.77) and split-half reliability were computed (r(tt)=0.86). The criterion-related validity with the convergent criterion "number recall" amounted to r=0.52 (p<0.0001) for all. CONCLUSION: It seems that young children nowadays have more difficulties successfully solving the items in the Mottier test than did children about 25 years ago.
BACKGROUND: The Mottier test to examine phonological processing is a widespread, frequently used instrument in clinical diagnostics. However, data on its validity and reliability are scarce. The reference scores for non-schoolchildren in 1981 (collected with small samples) have become outdated. METHOD: Three hundred and eight children (154 boys, 154 girls) from the German federal states of Lower Saxony and North Rhine-Westphalia were examined with the Mottier test (4-year-olds: n=86; 5-year-olds: n=126; 6-year-olds: n=96). RESULTS: The average item difficulty P for all children was 45.5 (range 5.8-90.6). On average, the items were easier to perform with increasing age (4-year-olds: P=37.3; 5-year-olds: P=46.5; 6-year-olds: P=52.7). The average item discrimination indices r(it) were 0.49 for 6-year-olds and 0.91 for 4-year-olds. The mean test score for all children was 13.76 (SD 4.35). There was a shift for the worse compared with the mean test scores from 1981 (4-year-olds: 11.19, SD 3.93; 5-year-olds: 13.95, SD 4.02; 6-year-olds: 15.81, SD 3.97). To determine reliability, the internal consistency (Cronbachs alpha=0.77) and split-half reliability were computed (r(tt)=0.86). The criterion-related validity with the convergent criterion "number recall" amounted to r=0.52 (p<0.0001) for all. CONCLUSION: It seems that young children nowadays have more difficulties successfully solving the items in the Mottier test than did children about 25 years ago.
Authors: Umberta Bortolini; Barbara Arfé; Cristina M Caselli; Luisa Degasperi; Patricia Deevy; Laurence B Leonard Journal: Int J Lang Commun Disord Date: 2006 Nov-Dec Impact factor: 3.020