FRAC Feedback Literacy Script (Filter–Reason–Act–Check)

Design · All designs
Evidence · 3 claims (3 for) · 1 study (1 causal), q3 · 0 of 1 report an effect size · 3 claims rest on one study

Description

FRAC is a four-step feedback literacy script that guides students' processing of external AI feedback: "Filter, identifying the type of problem addressed by the feedback; Reason, judging whether the feedback was valid and relevant to the current task", then Act (implementing adopted feedback as a specific revision) and Check (reviewing whether the revision achieved the intended result). Training lasted approximately 15 to 25 minutes before T1 with brief reviews before T1 and T2; students recorded decision sheets at T1 and T2 as process data, with return rates exceeding 90% per round.

Design Implications

Context

Requirements

• Students must process AI feedback systematically through the script and complete decision sheets recording how they handled each feedback item

Constraints

• In this study it was implemented with undergraduate English majors using Claude Sonnet 4 feedback during T1 and T2 only

Target Learners

• undergraduate English majors with basic academic writing skills

Learning Goals

• critical evaluation and effective uptake of AI-generated writing feedback
• deep revision of argumentative essays

Claims

• Frac Combined Highest Writing Quality Gains [+M]
• Interventions Raise Effective Adoption Rate [+M]
• Frac Shifts Revision Toward Deep Levels [+M]

Related Designs

Examples

-

Key Sources

• Dai (2026). Rethinking AI-assisted writing instruction: feedback literacy scripts, calibration training, and student writing development. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2026.1829268

Page metadata

Field	Value
type	design
id	frac-feedback-literacy-script
title	FRAC Feedback Literacy Script (Filter–Reason–Act–Check)
description	FRAC is a four-step feedback literacy script that guides students' processing of external AI feedback: "Filter, identifying the type of problem addressed by the feedback; Reason, judging whether the feedback was valid...
status	review
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content
sources	1 — dai-2026

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