Eligibility criteria for language impairment: is the low end of normal always appropriate?

Abstract

Purpose: The assumption that children with language impairment will receive low scores on standardized tests, and therefore that low scores will accurately identify these children, is examined through a review of data in the manuals of tests that are intended for use in identifying such children.

Method: Data from 43 commercially available tests of child language were compiled to identify whether evidence exists to support their use in identifying language impairment in children.

Results: A review of data concerning the performance of children with impaired language failed to support the assumption that these children will routinely score at the low end of a test's normative distribution. A majority of tests reported that such children scored above 1.5 SD below the mean, and scores were within 1 SD of the mean for more than a quarter (27%) of the tests. The primary evidence needed to support the purpose of identification, test sensitivity and specificity, was available for 9 of the 43 tests, and acceptable accuracy (80% or better) was reported for 5 of these tests.

Implications: Specific data supporting the application of "low score" criteria for the identification of language impairment is not supported by the majority of current commercially available tests. However, alternate sources of data (sensitivity and specificity rates) that support accurate identification are available for a subset of the available tests.