In Search of an Effective Method of Measuring Aboriginal Children's Speech and Language Development

• Main Author: Anderson, Ann
• Other Authors: Childs, Ruth, Applied Psychology and Human Development
• Format: Thesis
• Language: unknown
• Published: 2015
• Subjects: assessment; children; First Nations; 0460; Canada; Peltier
• Online Access: http://hdl.handle.net/1807/69216

The present study examines the effectiveness of the Fluharty Preschool Speech and Language Screening Test - Second Edition (FPSLST-2) as a speech and language measure for First Nations children. In the literature (Gould, 2008b; Klenowski, 2009; Pearce Williams, 2013; Peltier, 2009) and in practice, questions have been raised about whether any standardized speech language assessment is effective in measuring skills of First Nations children. The effectiveness of speech and language standardized tests was investigated by comparing test performance of two groups within the Rainy River District in Ontario, Canada: First Nations (FN) and Non-First Nations (NFN) children in the Rainy River District School Board (RRDSB) over a three year period (2009-2011). The study's target population included 429 Senior Kindergarten (SK) children comprised of 314 NFN children and 115 FN children from 11 elementary schools. There were overall differences in scoring patterns, with NFN children performing significantly better than FN children. A Mantel-Haenszel Differential Item Functioning (DIF) analysis provided a detailed picture of how individual items functioned psychometrically in the two groups. This analysis found six phonology items showing small and moderate amounts of uniform and non-uniform DIF and one language item showing a moderate level of non-uniform DIF. Review of these items by the Native Language and Curriculum Coordinator of the RRDSB suggested explanations for the difference in performance. In particular, items with the sounds /fl,v/ on the Articulation subtest exhibited a moderate level of uniform DIF in favour of the NFN group because the sounds f, l, r, and v do not exist in the Ojibwe language spoken in the Rainy River District. In addition, because the Ojibwe language is comprised of 80% verbs, it is reasonable to expect DIF for items describing verbs, which was noted; however, the DIF detected was non-uniform. Suggestions are provided for how the assessment's content as well as the administration and scoring might be adapted to better evaluate the speech and language development of FN children in the Rainy River District. Ph.D.