Seven-lever course-design framework for shaping student AI use

Design · All designs
Evidence · no claims cited

Description

The framework defines a lever as "a recurring feature of the course environment that students experience, that can be configured differently by someone within the institution, and that may influence what students do with or without direct supervision." The seven levers are Structure (what pays), Material (what they work on and when), Practice (how the room runs), Response (what gets said back), Norms (what is normal), Frame (what they understand about why), and Relationship (what is safe). Every classroom has a setting on all seven, whether chosen or defaulted, and none of the levers is specific to AI.

Design Implications

Context

Requirements

• Each lever must be a recurring, experienceable feature of the course environment configurable by someone within the institution

Constraints

• The framework is offered as a heuristic taxonomy, not a validated or necessarily exhaustive model, and does not establish causal effects for particular configurations
• Class size, contact hours, and tracking are conditions of the cultural surround, not levers

Target Learners

• secondary and university mathematics students

Learning Goals

• protecting independent thinking when AI use is unobserved
• deepening rather than shortcutting reasoning

Claims

Related Designs

Examples

-

Key Sources

• Halani, A. (2026). Designing for Reach: Seven Levers and the Student Alone with AI. Tang Institute at Phillips Academy. https://www.tanginstitute.andover.edu. doi:10.35542/osf.io/t4cd2_v1

Page metadata

Field	Value
type	design
id	seven-levers-course-design-framework
title	Seven-lever course-design framework for shaping student AI use
description	The framework defines a lever as "a recurring feature of the course environment that students experience, that can be configured differently by someone within the institution, and that may influence what students do w...
status	review
generated	by process:settle-candidates on 2026-10-12
trust	unverified — no one has confirmed this page's content
sources	1 — halani-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.