Assessing Potential Bias in Population Level Research with Aboriginal Children: Literature Review, Psychometric, and Administration Report
2016 Directions Evidence and Policy Research Group, LLP
The Early Development Instrument (EDI) is a tool for assessing and monitoring the state of early childhood development within communities. To be informative and useful to communities, any population monitoring tool must provide valid and reliable data that is free of bias. Aboriginal children tend to receive lower EDI scores than their non-Aboriginal peers (Janus, 2002; Lapointe et al., 2007: McTurk et al., 2008). A number of researchers and stakeholders have suggested that this result reflects a bias against Aboriginal children within the EDI, and argue that EDI scores underestimate the true abilities of Aboriginal children (Walton, 2007). In response to concerns about the validity and potential bias of the EDI for use with Aboriginal children, the First Nations Education Steering Committee commissioned an independent assessment of the EDI. This report details the findings of a review of the literature, psychometrics, and administrative procedures regarding the EDI with respect to Aboriginal children, and of an experimental simulation of the EDI assessment.