Resource Hub
Explore guides, research papers, policy briefs, and tools to aid decisions on standards and assessments.
Placing Multiple Early Literacy Screening Instruments on a Common Risk Metric: A Machine Learning Approach
Many states now require districts to administer an early literacy screener in the primary grades, selecting from a list of state-approved screeners. Given each screener’s scale, benchmarks, and definition of risk, states may question how to compare results across screeners. This article presents a machine learning approach to predicting student probability of scoring below proficiency on a grade 3 reading assessment. States may consider how such a predictive model can create a singular benchmark from multiple screening tools.
