Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Student Assessment topic
No spam. Unsubscribe anytime.
Board hears progress report on reducing zero scores for extended constructed responses
Summary
District instructional leaders reported a goal to reduce zero scores on STAR extended constructed responses by 10% by June 2026, presented 2025 baselines by grade and language, noted grade‑level successes and challenges, and described interventions including targeted instruction, STAR‑aligned boot camps, and a new AI writing‑feedback tool.
Instructional leaders gave the Castleberry ISD board an update on Priority 4 of the district improvement plan: reducing the frequency of zero scores on extended constructed responses (ECRs) across RLA‑tested grades.
Presenters explained that the district uses the most recent benchmark (Benchmark 3) as an apples‑to‑apples comparison with the 2025 STAR release, and that the district set a goal of a 10 percent reduction in zeros by June 2026. They reviewed baseline 2025 STAR ECR data: for example, the presenters reported that in one grade level 53 percent of English testers earned a zero and 36 percent of Spanish testers earned a zero on the ECR baseline, then showed benchmark‑3 results indicating improvements in certain grades (fourth and fifth grade and seventh grade showed notable reductions). Presenters noted that a lower percentage indicates better performance on these metrics.
Interventions described included a multi‑week STAR‑aligned boot camp for tested levels, expanded practice across assessed writing modalities, reinforcement of evidence‑based reading comprehension strategies, and adoption of a new AI writing‑feedback tool (referred to as "Magic School") to serve as a second scorer and provide targeted feedback. Presenters cautioned that ECR prompt rigor and language background can affect results and that tester counts can vary across years.
Board members asked clarifying questions about cohort comparisons and sample sizes; presenters said comparisons are by grade‑level accountability rather than a single tracked cohort and that tester counts may vary depending on local testing decisions. The board did not take action; staff will continue monitoring benchmark and STAR results toward the June 2026 goal.
