Psychometric properties of the English language version of the C-BiLLT evaluated in typically developing Canadian children.

Abstract

Purpose: This study aimed to 1) investigate the convergent and discriminant validity, internal consistency, and test-retest reliability of the Canadian English version of the Computer-Based instrument for Low motor Language Testing (C-BiLLT-CAN), and 2) explore feasibility of the C-BiLLT assessment for children with cerebral palsy (CP) and complex communication needs in the Canadian health care context.

Methods: Eighty typically developing children between 1.5 and 8.5 years of age completed the C-BiLLT-CAN, the Peabody Picture Vocabulary Test-IV (PPVT-4), the receptive language sub-test of the New Reynell Developmental Language Scales (NRDLS), and/or the Raven's 2. Correlations between raw scores were calculated for estimates of convergent and discriminant validity. Internal consistency was calculated for all items and separately for items pertaining to vocabulary and grammar. To calculate the standard error of measurement (SEM) and intraclass correlation coefficient (ICC), 33 participants were re-tested with the C-BiLLT within three weeks. Feasibility was explored with nine participants with CP.

Results: C-BiLLT-CAN's convergent validity was good to excellent (Spearman's rho > 0.78) and discriminant validity was higher than hypothesized (Spearman's rho > 0.8). Internal consistency (Cronbach's alpha = 0.96), test-retest reliability (ICC > 0.9), and measurement error (SEM < 5%) were excellent. The feasibility study could not be fully completed due to the COVID-19 pandemic. Preliminary data demonstrated some technical and practical barriers for using the C-BiLLT in children with CP in Canada.

Conclusion: The C-BiLLT-CAN showed good to excellent psychometric properties in a sample of typically developing children, indicating that it is an adequate test for measuring language comprehension in English-speaking Canadian children. Further research is needed to investigate the feasibility of the C-BiLLT-CAN in children with CP.