No abstract

The purpose of this study was to explore the production of infinitival complements by children with specific language impairment (SLI) as compared with mean length of utterance (MLU)-matched children in an effort to clarify inconsistencies in the literature. Spontaneous language samples were analysed for infinitival complements (reduced infinitives and true infinitives). Participants included children with SLI (n = 19; 5;2-7;10) and children with typical language (n = 19; MLU; 3;0-5;9). There was no group difference in the number of infinitival complements and the number of different complement-taking verbs. However, the SLI group produced more true infinitives than the MLU group. The SLI group was less accurate than the MLU group on inclusion of obligatory infinitival to, with 80.21% accuracy (SD = 29.42) and 99.81% accuracy (SD = 0.85), respectively. As a group, children with SLI did not have problems with the clausal structure of infinitives. However, they had difficulty with the specific grammatical requirement of infinitival clauses, that is, the inclusion of the infinitival marker.

School speech-language pathologists (SLPs) use norm-referenced measures for many purposes, including eligibility decisions for the disability category of speech or language impairment. The Individuals With Disabilities Education Improvement Act of 2004 (IDEA) defines conditions that must be met when conducting full and individual evaluations (see Appendix A). As such, SLPs must assure that chosen measures are "technically sound" and "selected and administered so as not to be discriminatory on a racial or cultural basis" (Sec. 614). Because the federal definition of speech or language impairment is broad, individual state departments of education publish guidelines or requirements for eligibility decisions. Typically, these guidelines give prominence to a child's performance on norm-referenced measures in determining eligibility as a child with speech or language impairment. For example, Missouri and Tennessee specify a standard score cutoff of 1.5 standard deviations below a measure's norming sample mean (see Appendix B) to meet eligibility requirements. The study reported herein contributes to the discussion on identification of and eligibility determination for language impairment within the IDEA category of speech or language impairment. We address the challenges that ensue from the single population-based norms provided in the majority of, if not all, commercially published norm-referenced measures. We argue that norms based on a single aggregated norming sample can lead to discrimination on a cultural basis because family socioeconomic status (SES), often indexed by maternal education (Bradley, Corwyn, McAdoo, & Coll, 2001; Hoff, 2006), is a critical component of culture (see Note 1), particularly as it relates to school achievement (Waldfogel, 2012). As a result, we argue further that a well-supported evidence-based practice is consideration of family SES in the interpretation of a child's 679886C DQXXX10.