Psychometric Characteristics of the Visual Communication and Sign Language Checklist

Abstract Since its publication in 2013, the Visual Communication and Sign Language (VCSL) Checklist has been widely utilized to assess the development of early American Sign Language skills of deaf children from birth to age 5. However, little research has been published using the results of VCSL assessments. Notably, no psychometric analyses have been conducted to verify the validity of the VCSL in a population whose characteristics are different from those of the small sample of native signing children from whom the published norms were created. The current paper, using data from the online version of the VCSL (VCSL:O), addresses this shortcoming. Ratings of the 114 VCSL items from 562 evaluations were analyzed using a partial-credit Rasch model. Results indicate that the underlying skill across the age range comprises an adequate single dimension. Within the items’ age groupings, however, the dimensionality is not so clear. Item ordering, as well as item fit, is explored in detail. In addition, the paper reports the benefits of using the resulting Rasch scale scores, which, unlike the published scoring strategy that focuses on basal and ceiling performance, makes use of the ratings of partial credit, or emerging, skills. Strategies for revising the VCSL are recommended.