Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Assessments And Data Analysis topic
No spam. Unsubscribe anytime.
District presents I-Ready predictions and finds weak correlation between high-school grades and SAT scores
Summary
Central Falls staff shared I-Ready district-wide predicted proficiency under target and stretch-growth scenarios and said final averages/GPA explain a small portion of SAT variation, limiting their use as strong proxies for state high-school assessment predictions.
Central Falls staff presented analysis on March 25 showing projected student proficiency under I-Ready target and stretch-growth scenarios for grades tested and reported findings about the relationship between high-school GPA/final averages and SAT scores.
District staff said the I-Ready prediction model, based on data from several Rhode Island districts, projects that if students meet target growth, district-wide predicted reading proficiency would be about 8% and math about 10%; under a stretch-growth scenario the projections rise to roughly 13% in both subjects. Staff cautioned that the predictions include caveats about prediction accuracy and are intended for planning, not as guarantees.
For high school students—where I-Ready predictions are not available—the district staff presented correlation analyses showing that GPA and class final averages are positively correlated with SAT results but that the explanatory power is low. Staff said only about 15% of variation in SAT scores could be explained by final averages in LA and math classes in their data, and that a one-point increase in final average corresponded on average to about a 1.5-point increase in SAT score—insufficiently strong to serve as a precise proxy.
Staff used R and Stata statistical tools for the analysis and said they examined breakdowns by class type (honors, multilingual learner sections, regular classes) and found some variation by course grouping. Trustees asked clarifying questions about prediction accuracy and methodology, and staff said the models include prediction-accuracy factors and that the district will continue refining predictive analytics to guide instruction.
No policy change was proposed; the presentation was informational and intended to inform planning, intervention and communications with educators and families.
