Although there is extensive documentation of the morphological limitations of children with specific language impairment (SLI), few studies have reported on complex syntax acquisition in children with SLI. This case study examined the development of complex syntax in a child with SLI between 3 and 7 years. Twelve conversational samples were analysed to describe emergence of complex syntax types, proportional use of complex syntax, and complex syntax production errors. Earliest emerging complex syntax types were catenatives and let's clauses, not always considered true complex syntax, as well as simple infinitives. These three types accounted for 88% of complex syntax tokens through age 4;8 (MLU = 3.12). A diverse range of complex syntax types was produced consistently at age 5;9 (MLU = 4.27), including wh clausal complements, relative clauses, and full propositional clauses. Production errors on complex syntax included omissions of infinitival to, omissions of wh pronouns in wh clausal complements, omissions of obligatory relative markers, and an omission of a complementizer. Production errors continued to be prevalent even in the last language sample at 7;10 (MLU = 5.46). Documentation of complex syntax development in children with SLI is critical to the clarification of linguistic deficiencies that characterize specific language impairment.

Background: There is a tradition of social validation studies in the adult clinical treatment literature that has not yet been extended to narrative evaluation. This social validation process begins with lay listener holistic judgements of narrative quality for non-brain-injured narrators, and compares these judgements to clinician-researchers' assessments of the same narratives. The resulting narratives rated for quality reflect the premorbid range of skills in clinical populations, and overall narrative performance. Narratives whose quality has been assessed holistically by laypersons can be further examined for the qualitative and quantitative dimensions that characterise them. Aims: This study explores the social validity of clinical narrative assessment. Specifically, it examines the degree to which clinicians' holistic and analytic assessments of narrative quality parallel holistic narrative quality judgements of a lay audience, for a set of narratives produced by non-brain-injured narrators. It also examines the characteristics of narratives along a continuum of quality. Methods & Procedures: A total of 14 African American and 15 Caucasian lay individuals holistically rated the quality of personal narratives produced by ethnicity-matched groups of 37 African-American and 34 Caucasian non-brain-injured narrators, respectively. Five clinical raters holistically rated all narratives. Additionally, lay raters were asked for their intuitions about what contributes to narrative quality. They also ranked six researcherspecified dimensions for their importance in determining narrative quality. High-rated and low-rated narratives were compared on dimensions of length, content, and narrative structure.Outcomes & Results: Agreement between lay and clinical rater groups was found for a subset of high-rated and low-rated stories, which constituted one-quarter to one-third of the total narratives. A similar level of agreement was found for male and female lay raters. Ratergroup disagreements at the extremes of narrative quality were rare. Certain of the lay raters' characterisations of good and poor narratives corresponded to narrative dimensions discussed in the literature and defined by the researchers. High-rated stories were longer, more conventionally structured, and different in content from low-quality stories. Conclusions: For personal narratives at the extremes of perceived quality (high and low), evidence for social validity in narrative assessment is manifested both at the holistic assessment level, and at the level of the dimensions thought to contribute to narrative quality. The results contribute to our understanding of the normal range of narrative performance.