Attention to Excellence or Need? An Experimental Investigation of How Teachers Prioritize Students

The study involves a survey of teachers, administered online. Participants will be presented with different hypothetical classroom scenarios featuring groups of students with varying performance levels (excellent, very good, somewhat good). In each scenario, teachers will be asked to indicate which students they would prioritize for attention during different classroom tasks. The study aims to (1) establish a benchmark stock of teachers’ attention practices across different classroom compositions, and (2) identify the mechanisms that drive teachers’ attention patterns, such as the perceived needs of students, opportunities for learning reinforcement, or motivations related to classroom management. Teachers participate anonymously and are randomly assigned scenarios to ensure variation in classroom composition and task type.

2025-04-30

2025-09-01

Attention Prioritization: The type of students (excellent, very good, somewhat good) to whom teachers report paying more attention across different classroom compositions.
Task-Specific Attention: Variation in teacher attention by classroom task (e.g., discussion participation, homework feedback, exam preparation).
Mechanisms of Attention: Stated reasons for prioritizing certain students (e.g., perceived need, potential for growth, maintaining classroom order).

Teachers are recruited to participate anonymously in an online randomized experiment. Each teacher is randomly assigned to a series of hypothetical classroom scenarios, varying the distribution of student performance (e.g., mostly excellent students, a mix, mostly somewhat good students). In each scenario, teachers are asked to select which students they would prioritize for attention across several classroom tasks. Scenario assignment and task types are randomized to prevent order effects and isolate causal impacts of classroom composition and task type on attention patterns. The design captures both direct choices (which students are prioritized) and stated rationales (why they are prioritized).

Randomization is conducted automatically through the survey platform (Qualtrics) using its built-in randomization tools. Each participating teacher is randomly assigned to a sequence of hypothetical classroom scenarios that vary the classroom task. The composition of student performance levels is fixed within participant. Answer option order is randomized to minimize ordering effects.

The unit of randomization is the individual teacher.

No

1000 teachers

1000 teachers

The design covers 7 experimental arms, corresponding to 7 different classroom compositions. Each teacher is randomly assigned to one classroom composition and completes all four classroom tasks. With a target sample of approximately 1,000 teachers, we expect about 140–150 teachers per classroom composition arm.

Seven Classroom Composition Arms:

1. All Excellent Students — The classroom consists entirely of excellent students.
2. All Very Good Students — The classroom consists entirely of very good students.
3. All Somewhat Good Students — The classroom consists entirely of somewhat good students.
4. Majority Excellent, Minority Very Good — Most students are excellent, a smaller share are very good.
5. Majority Very Good, Minority Somewhat Good — Most students are very good, a smaller share are somewhat good.
6. Even Mix Across All Types — The classroom is evenly balanced among excellent, very good, and somewhat good students.
7. Majority Somewhat Good, Minority Very Good — Most students are somewhat good, a smaller share are very good.