A validity study of the early language inventory: investigating ecologically appropriate communication assessment for aboriginal children in western Sydney

The overall aim of this thesis was to undertake research into the validity of the Early Language Inventory (ERLI), with parents of Aboriginal children (aged approximately 12-36 months) in western Sydney. ERLI was originally developed with Aboriginal parents in northern Australia as an authorised adaptation of the MacArthur-Bates Communicative Development Inventories (CDI), but it is known that parents in western Sydney prefer this tool to the mainstream Australian CDI (the OZI) as it features culturally familiar items and is done through yarning. The thesis had three goals: (1) describe how Aboriginal children in western Sydney score on ERLI; (2) measure the ecological validity of ERLI through comparison of children's ERLI scores with their language use at home via sampling of parent-child interaction in free play in the child’s home environment; (3) assess the relationship between children's ERLI scores and Hearing and Talking Scale (HATS) scores, to explore the extent to which both tools identify the same or a similar set of children as those who are typically developing or those who may be at risk of language delay and how this relates to parent concern. The results described typical language development on ERLI for Aboriginal children in western Sydney (N=81). Further assessment of ERLI is needed, however, to develop norms in which a 10th percentile cutoff can be confidently identified so that children with strong spoken language are not identified as at risk of language delay, or vice versa. Ecological validity was established in the home-language samples in that children's ERLI scores were significantly related to several linguistic measures of their spontaneous language use at home. Items on ERLI (vocabulary and gestures in local Aboriginal English varieties) were identified throughout the home language sample audiovisual recordings. The relationship between ERLI and HATS was assessed, and despite a small number of children who were identified as at-risk on HATS, there was substantial agreement between the tools for the 2- and 3-year-old children. The findings of this thesis are relevant to frameworks and policies which advocate for fairer and quality outcomes in Aboriginal and Torres Strait Islander issues, including access to fair and quality assessments for children.