Forge
markdowne8ad0934
1# Evidence: Rosenshine (2012) — Principles of Instruction
2
3## Provenance
4
5- source_refs:
6 - primary: Rosenshine, B. (2012). Principles of Instruction: Research-Based Strategies That All Teachers Should Know. *American Educator*, 36(1), 12–19, 39–45.
7 - open_pdf: https://www.aft.org/sites/default/files/Rosenshine.pdf
8 - eric: EJ971753
9- created_at: 2026-06-06
10- updated_at: 2026-06-06
11- extraction_note: **read_depth: full** — open AFT PDF (2026-06-06).
12
13## Metadata
14
15- card_id: kb-pedagogy-rosenshine-2012-evidence-v1
16- world: pedagogy.general
17- layer: knowledge
18- tags: [paper-evidence, rosenshine, principles-of-instruction, explicit-instruction]
19- status: active
20
21## Epistemic linkage
22
23- epistemic_basis: synthesis_from_primary (review integrating cognitive science + master-teacher studies)
24- evidence_type: peer-reviewed practitioner journal + research synthesis
25- confidence: **high** for principle list as teaching heuristics; **medium** for effect sizes (not meta-analysis itself)
26- read_depth: **full**
27- transfer_boundary: principles → gate/spacing/scaffolding `allow-with-check`; **not** license for drill-only without held-out probe
28- falsification_trigger: large RCT program contradicts specific principle operationalization
29
30---
31
32## Fundamentals
33
34Rosenshine distilled **10 principles** from: (a) cognitive science on working memory; (b) studies of master teachers (high success rate); (c) cognitive supports (models, scaffolds).
35
36| # | Principle (condensed) | Operational signal |
37|---|----------------------|-------------------|
38| 1 | Begin with review | spaced retrieval before new material |
39| 2 | New material in small steps | one construct per instructional moment |
40| 3 | Ask questions | check understanding, not monologue |
41| 4 | Provide models | worked example before problem |
42| 5 | Guide student practice | success rate high during acquisition |
43| 6 | Check responses | detect misconceptions early |
44| 7 | Obtain high success rate | ~80% correct during guided practice (cited range in article) |
45| 8 | Provide scaffolds | temporary supports, faded |
46| 9 | Independent practice | after sufficient guided success |
47| 10 | Weekly/monthly review | cumulative spacing |
48
49## Core unit
50
51- context: designing subject instruction or agent gate curriculum
52- signal: lesson too large; no review; no model before problem
53- action: map step to Rosenshine #; add held-out probe (see `kb-pedagogy-evidence-global-v1.md`)
54- outcome: higher success rate on **new** items, not just immediate recall of taught example
55- lesson: explicit instruction ≠ rote; includes questioning and fading
56
57## Operationalization
58
59- first_adoption_task: label each curriculum moment with Rosenshine principle #
60- validation_check: held-out probe exists for target construct
61- success_criterion: agent/journal shows spacing + independent probe pass
62- rollback_or_mitigation: if drill without transfer — add worked example + new context probe
63
64## Связанные файлы
65
66- Index: [`kb-pedagogy-scientific-evidence-v1.md`](kb-pedagogy-scientific-evidence-v1.md)
67- Synthesis: [`kb-pedagogy-evidence-global-v1.md`](kb-pedagogy-evidence-global-v1.md)
68
69*Версия: v1.0 · 2026-06-06*
70
71
View only · write via MCP/CIDE