Consider a prediction study if participation supports it

The advice. Consider completing a prediction study, if there are enough participants to support one.

Why. Raised in response to the observation that other systems make predictions of student performance on the summative test. A prediction study would replace the concordance table’s “could score” with a quantified relationship, which would answer the committee’s question about what could means and would let OEAA state the imperfection the table currently only gestures at.

Caveat carried with it. OEAA’s stated preference — and the committee’s — is for the present-tense framing: what a student would score today, rather than at the end of the year. A prediction study should not be allowed to convert the interim into a forecasting instrument at the expense of its instructional use.

Constraint. Conditional on participation volume. With roughly 14,000 students across 21 schools and flat year-over-year participation, sample adequacy is not obvious.

Status. Open.