Identifying risk for language impairment in children from linguistically diverse low-income schools

Abstract

Purpose: To improve screening procedures for children in a linguistically diverse context, we combined tasks known to reveal grammatical deficits in children with language impairment (LI) with training to facilitate performance on a verb elicitation task.

Method: Sixty-four first grade children participated. The objective grammatical measures included elicitation of 12 past tense regular verbs preceded by a teaching phase (teach-test), the sentence recall (SR) subtest of the Clinical evaluation of language fundamentals (CELF-4), and a tally of all conjugated verbs from a narrative retell task. Given the widespread reliance on teacher observation for the referral of children suspected of having LI, we compared our results to the spoken language portion of the CELF-4 teacher observational rating scale (ORS).

Result: Using teacher observation as a reference for comparison, the past tense elicitation task and the SR task yielded strong discriminating power, but the verb tally was relatively weak. However, combining the three tasks yielded the highest levels of sensitivity (75%) and specificity (92%) than any single measure on its own.

Conclusion: This study contributes to alternative assessment practices by highlighting the potential utility of adding a teaching component prior to administering informal grammatical probes.

Keywords:

• language impairment
• past tense
• screening
• linguistic diversity

Acknowledgements

The American Speech-Language-Hearing Foundation’s New Century Scholars Award provided support for this study. Support for development of the video stimuli came from the National Institutes of Health/National Institute on Deafness and other Communication Disorders (Grant 5RO3DC 07018-02) awarded to the first author. Gratitude is extended to Deanna Kokotos, Sapna Narang, Amanda Pellegrino and Marta Gielarowiec who assisted with data collection and preparation of spreadsheets, and to Lauren Kiraly and Amanda Sherwood for proofreading and editing. I also acknowledge the assistance of Ralph Larkin, statistical consultant. Finally, we are most indebted to the children and families who participated, and to the principals and teachers for their cooperation.

Supplementary material

Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/17549507.2017.1406987.