Sentence-Level Writing Skills in Children With and Without Developmental Language Disorders

Abstract

In this issue of Topics in Language Disorders, guest issue editor Dr. Anthony Koutsoftas invited groups of authors to present research related to text generation, primarily at the sentence level, by preschoolers and school-age children, including children with language impairments. Five articles are included, representing systematic reviews of the literature, examination of syntactic and morphological complexity and accuracy, and assessment and intervention for sentence writing. In the first article, Williams and Larkin conducted a systematic review of 39 studies in which transcription (i.e., writing mechanics such as spelling) and/or translation (i.e., converting ideas into comprehensible language) processes were evaluated in children between 4 and 17 years of age. They found that children with developmental language disorders (DLD) exhibited delays in both translation and transcription processes relative to their same-age peers with typical language development. These delays were observed across different measures including writing quality, writing productivity, and spelling. The authors discuss the limitations they saw in the literature they reviewed and offer suggestions for researchers for further work. In the second article, Brimo et al. analyzed narrative writing samples produced by a group of elementary students with DLD and another group with typical language development who were on average nearly a year younger. They specifically examined morphological (e.g., past tense -ed and be verbs) and syntactic (e.g., word omission and word order) errors in both simple and complex sentences produced by the students. They found that children with DLD produced a higher percentage of regular and irregular past tense errors than children with typical language development but only in simple sentences. Children with DLD produced significantly more syntactic errors than children without language delay in simple and complex sentences. However, children with DLD did not produce significantly more errors in simple sentences than in complex sentences. They discuss implications for assessment by speech–language pathologists. Next, Ritchey et al. performed a systematic review of 16 sentence writing intervention studies. These studies included oral language instruction, written language instruction, or combined oral and written language instruction. Participants in the reviewed studies ranged from typically developing to students at risk for or identified with learning disabilities or attention-deficit/hyperactivity disorder to students with significant developmental disabilities such as autism. The core instructional procedures that appeared to yield positive outcomes in sentence writing included explicit instruction, self-regulation procedures, opportunities to link oral and written language, and sentence generation practice. The authors noted several significant limitations in the extant research, including the limited number of studies in this area. Then, in the study reported by Hall-Mills and Wood, the syntactic complexity of informational texts produced by fifth-grade students with and without language impairment, with and without native English-speaking proficiency, was compared. They focused on the frequency of utterances containing complex syntax and four specific clause types (conjoined, subordinate, relative, and full complement). Children with typical language development, regardless of English proficiency, wrote more words, utterances, and different word roots than their peers with language impairment. When text length in the writing samples was controlled, a significant difference between groups was found for use of relative clauses but not for other clause types. Specifically, English-proficient students with language impairment produced a greater proportion of utterances with relative clauses, which is not necessarily associated with more sophisticated sentences in informational texts. In addition, there were moderate positive correlations between variables of complex syntax and a measure of text narrativity, which was low given the informative genre of the writing samples. In the last article, Marble-Flint and Koutsoftas conducted a feasibility study of a virtual sentence writing probe task with 15 intermediate-grade children, some of whom were considered struggling writers. The probe task included visual (pictures), written (descriptive sentences for pictures plus key words, representing nouns, verbs, adverbs, and conjunctions, to be used in responses), and verbal (oral rehearsal) scaffolds. Scores examined included total number of words, a sentence accuracy score, and a word accuracy score, which were compared across three administration time points. Generally, there were no significant differences across time points on the measures for the sample of children. Moreover, scores on the sentence writing probes were correlated with standardized measures of oral language. The study supports the proof of concept that virtual assessment procedures can be used to assess sentence-level writing in late elementary-age students, an important finding, given the expansion of online education options since the onset of the COVID-19 pandemic. Overall, the articles in this issue of Topics in Language Disorders indicate children who struggle with writing, especially those of whom are diagnosed with DLD, exhibit morphological and syntactic errors in their written sentences (even when such errors are resolved in their spoken language), use more relative clauses when this may not be the best mechanism for conveying information through writing, and have more difficulties with transcription and translation processes involved in text production (and, correspondingly, produce less text than their typical peers). Fortunately, research reviewed in this issue also indicates that sentence-level writing instruction is generally effective for students with disabilities and assessments to examine the sentence level of written language production can be employed even in virtual settings. —Gary A. Troia, PhD, CCC-SLP —Sarah E. Wallace, PhD, CCC-SLP Co-Editors