📌 What the policy did

A short-lived rule changed how teachers were evaluated by altering which student test scores counted. Students who failed an initial end-of-year test were retested a few weeks later, and only the higher of the two scores was used to calculate the teacher's evaluation score. That rule created a clear discontinuity in teachers' incentives when deciding how to allocate effort across assigned students.

📊 How the change reveals teacher behavior (quasi-experimental design)

This discontinuity functions as a quasi-experiment: otherwise similar students were treated differently solely because the retest rule made marginal effort on some students more valuable for a teacher's evaluation score. Administrative test timing and score-use rules isolate the effect of evaluation incentives on teacher effort and subsequent student outcomes.

🔑 Key findings

• Retested students scored 0.03σ higher than non-retested students one year later — a measurable, lasting increase in achievement.
• The gain persisted long after the retest rule (and the associated incentive discontinuity) had ended.
• The pattern implies teachers reallocated effort toward students affected by the rule, despite arguably equal returns to effort across students.

💡 Why it matters

The evidence shows that even short-lived, targeted changes to evaluation rules can shift teacher effort and produce durable student gains. This highlights the power of evaluation design to shape educator incentives and long-term student outcomes, with implications for how performance policies are structured and evaluated.