June 6, 2012

Disentangling the impact of speech and language impairment from speaking a language other than English

While in Brisbane I have been able to discuss and progress the analysis of data regarding children in the Longitudinal Study of Australian Children (LSAC) who speak languages other than English. My colleagues, Dr Sue Walker (Queensland University of Technology), Chrystal Whiteford (QUT), Linda Harrison (CSU) and I have been disentangling the impact of speaking a language other than English from having a speech and language delay in any/all languages.

Here is a summary of our findings to date:
Within Australia educational provisions are made to support children who speak languages other than English (LOTE) as their primary language. For children who speak a LOTE and have speech and language difficulties in each language, additional support services are required. This study aimed to determine the impact of primarily speaking a language other than English at home when 4- to 5-years-old on teacher-rated literacy, numeracy, and social-emotional outcomes at 8- to 9-years-old for children identified with and without overall speech and language concern. Participants were from the Kindergarten cohort of The Longitudinal Study of Australian Children (LSAC) (n = 4983). A matched-case control design (controlled for sex, socio-economic position (SEP), and age) enabled comparison between four groups of children: a) English + typical speech and language (n=109), b) LOTE + typical speech and language (n=109), c) English + speech and language concern (n=109), d) LOTE + speech and language concern (n=109). Next, multiple regression analyses were undertaken for the entire LSAC sample (n=4983) constructed in a hierarchical framework, with variables entered in four steps: child factors (including LOTE), family factors, competencies when 4- to 5-years-old, then competencies when 6- to 7-years-old. Three teacher-rated outcome measures at age 8- to 9-years were considered. The results indicated at 4- to 5-years there were significant differences between groups a) and b) on the Peabody Picture Vocabulary Test (PPVT), a direct assessment measure, and the parent-rated Strengths and Difficulties Questionnaire (SDQ). Further, there were significant differences between groups c) and d) on the PPVT. Children speaking LOTE performed more poorly than English speaking children on both these measures. There were no significant differences between groups on the Who am I? (WAI). However, by 8-to 9-years speaking a LOTE did not predict school-age academic and social-emotional difficulties. Significant predictors for Language and Literacy included SEP, parent and teacher reported use of speech pathology services, expressive language concerns, PPVT at 4- to 5-years and children’s competencies at 6- to 7-years. Significant predictors for Mathematical Thinking included child sex, SEP, LOTE, parent and teacher reported use of speech therapy, 4/5 PPVT and children’s competencies at 6- to 7-years. For the SDQ significant predictors included child sex, SEP, receptive language concerns and children’s approaches to learning at 6-7 years. These results indicate that presence of speech and language concern at 4- to 5-years was a greater predictor of school-age academic and social-emotional difficulties than speaking a LOTE.

Sue Walker, Chrystal Whiteford, Linda Harrison & Sharynne