De-emphasize Answers, Emphasize Reasoning¶

Strategy · All strategies
Evidence · 4 claims (4 for) · 10 studies (5 causal, 3 quant-synthesis, 1 associational, 1 theoretical), q2–q4 · 2 of 10 report an effect size

Description¶

This strategy shifts classroom attention from what the answer is to why it works and how it was reached. Teachers present worked examples — including deliberately flawed ones — and ask learners to find and explain the error, or restructure participation routines so that learners must articulate their thinking process before, or instead of, stating a final answer. The answer becomes evidence in an argument about reasoning, not the endpoint of the activity.

Design Implications¶

Requiring learners to explain and justify their steps triggers self-explanation, which improves conceptual understanding beyond what answer-checking alone produces Self-explanation improves conceptual understanding. [+S]. Studying erroneous examples — worked solutions containing a plausible mistake — builds conceptual knowledge and error-detection skill better than studying only correct solutions Erroneous examples build conceptual knowledge. [+M]. The strategy also reframes evaluation around process rather than outcome, which benefits novices who lack the knowledge to reason from outcome goals Process goals outperform outcome goals for novices. [+M].

Context¶

Requirements¶

• A classroom culture in which reasoning is publicly valued and errors are treated as informative, not shameful — this must be established before learners will expose partial thinking
• Teachers skilled at facilitating discussion of strategies, including responding to wrong reasoning without simply supplying the answer
• Tasks with reasoning worth examining: multi-step problems, contrasting solution methods, or cases where plausible errors occur
• Time: reasoning-focused discussion is slower than answer-checking; fewer problems must be covered
• Tasks with multiple defensible solution paths or interpretable errors, not single-step recall items
• A classroom culture where errors are treated as information; teachers must respond to wrong answers by probing the reasoning rather than immediately correcting
• Assessment and grading practices that reward explanation and justification, or the strategy will be undermined by what students perceive as counting
• Teacher facility with facilitating discussion and with Coaching moves that press on thinking ("Why did you choose that operation?")

Constraints¶

• Difficult to implement in answer-focused settings where high-stakes assessment rewards speed and correctness; learners may perceive reasoning talk as off-task [-M]
• For learners with very low prior knowledge, open reasoning discussion can exceed working memory capacity and produce unproductive search; some structure (worked examples, Fading) is needed first [~M]
• If facilitation is weak, discussion can drift to a single confident voice or degenerate into unfocused talk without consolidation of the target concept [-M]
• Undermined when high-stakes assessments reward speed and correct answers only — students rationally optimize for what is graded [-M]
• Time-intensive: examining several reasoning paths takes longer than confirming one answer, creating pressure to revert to answer-checking under pacing constraints [-M]
• Less effective for learners with very low prior knowledge, who may lack the schema needed to evaluate reasoning at all and benefit more from direct modeling first [~M]
• Can confuse novices if multiple student strategies are compared without teacher synthesis; unstructured sharing of flawed reasoning can propagate misconceptions [~M]

Implementation Variability¶

• Error-finding tasks: present a worked solution with a planted error; learners locate, explain, and repair it
• Answer-last routines: learners explain their approach before revealing the answer (Think-Aloud, Articulation)
• Multiple methods: compare two or more valid solution paths and discuss why each works
• Think-pair-share: pairs negotiate reasoning before whole-class discussion, lowering the risk of public error
• Erroneous examples: present a fully worked but incorrect solution; students find and fix the error, then explain why it was wrong
• Answer-delayed discussion: collect answers to a conceptual question, have students defend or revise them in Class Discussion before the teacher confirms
• Reasoning-first norms: teacher opens with "Tell me the first thing you thought about," explicitly banning immediate answer-giving
• Two approaches compared: display two student solutions side by side and ask which is more efficient or generalizable — a Comparing Cases move that supports abstraction Multiple contrasting cases support abstraction. [+M]

Target Learners¶

• Novices and learners who struggle with mathematics or lack confidence in problem-solving; de-emphasizing the single correct answer reduces anxiety and legitimizes partial thinking [~M]
• Learners with some prior knowledge, who benefit most from explaining and comparing methods Self-explanation improves conceptual understanding. [+S]
• Less effective for complete beginners without worked examples to anchor the discussion [~M]
• Students who experience mathematics as answer-production and have developed unproductive beliefs about speed and correctness; the strategy broadens participation beyond fast answerers
• Novices with minimal prior knowledge need worked modeling before error analysis becomes productive [~M]

Target Learning Goals¶

• Conceptual understanding: why procedures work, not just how to execute them
• Error detection and self-monitoring: recognizing when a solution path has gone wrong
• Mathematical discourse and argumentation: justifying, critiquing, and comparing reasoning
• Metacognitive monitoring: evaluating one's own solution path for validity
• Mathematical argumentation and justification

Instructions¶

1. Select a problem with rich reasoning and plausible errors; solve it (or present a prepared solution) containing a deliberate mistake
2. Ask learners to work through the solution and identify where and why it fails, using Non-Examples to sharpen discrimination
3. Have learners Articulation their own reasoning aloud or in writing before stating an answer; the teacher probes with "why?" and "how do you know?" rather than confirming correctness
4. Facilitate Class Discussion comparing different solution strategies, treating each as a candidate argument to evaluate
5. Consolidate: name the key idea the reasoning revealed, and follow with Practice that requires the same reasoning on new problems

Related Strategies¶

• Productive Failure — lets learners generate and discuss flawed solutions before canonical instruction, using the same reasoning-first logic
• Think-Pair-Share — a participation structure that makes reasoning discussion low-risk
• Teach students to intentionally choose from alternative algebraic strategies when solving problems, after they have procedural fluency
• Emphasizing reasoning pairs naturally with error-based learning: diagnosing incorrect worked examples is its most structured enactment
• Works alongside formative assessment practices, since eliciting reasoning is formative evidence gathering
• Question students about why a correct answer was chosen, rather than why an error was made
• Error-injection exercises: diagnosing subtly wrong simulations as graded practice
• Teacher follows learners' reasoning path and poses questions that alert learners to their geometric met-befores
• Have learners develop their own problem-solving templates instead of replicating instructor templates

Related Elements¶

• Articulation — the mechanism by which reasoning becomes visible and discussable
• Think-Aloud — models expert reasoning so learners have a target for their own explanations
• Non-Examples — erroneous solutions function as non-examples that sharpen conceptual boundaries
• Class Discussion — the forum in which competing reasoning is compared and evaluated
• Practice — consolidation; reasoning talk without subsequent application rarely transfers

Examples¶

• Error-finding in algebra: present a worked equation-solving sequence with a sign error; learners locate the step, explain why it is wrong, and correct it — a form of erroneous-example learning shown to build conceptual knowledge Erroneous examples build conceptual knowledge. [+M]
• Number talks (K–6 mathematics): a short daily routine in which learners solve a computation mentally and the teacher records and compares multiple strategies, with the class discussing why each works; the answer is noted but never the focus
• Kapur's Productive Failure: students attempt novel problems in groups, generating reasoning that is often wrong, before receiving canonical instruction — outperforming direct instruction followed by practice on conceptual and transfer measures [+S]
• Illustrative Mathematics curriculum: tasks are structured around comparing student solution strategies and explaining reasoning, with answers treated as claims to justify
• Erroneous-example routine: a teacher projects a solved equation with a sign error and asks, "Find the step where this went wrong, and convince your neighbor." Students must locate and justify the error before any correct solution appears.
• "My favorite no": a widely used middle-school routine (popularized by Levi & Teaching Channel) in which the teacher collects anonymous wrong answers, displays one, and has the class identify what is right about the reasoning before fixing the error.
• Cognitively Guided Instruction — teachers elicit and build instruction around students' own solution strategies rather than demonstrating the canonical algorithm first.
• Case-Based Learning — in professional education, discussion of reasoning and diagnosis is the point; the "answer" is often deliberately withheld or contested.

Key Sources¶

• Chi, M. T. H., de Leeuw, N., Chiu, M.-H., & LaVancher, C. (1994). Eliciting self-explanations improves understanding. Cognitive Science, 18(3), 439–477. doi:10.1207/s15516709cog1803_3
• Kapur, M. (2014). Productive failure in learning math. Cognitive Science, 38(5), 1008-1022. doi:10.1111/cogs.12107
• Hiebert, J., & Wearne, D. (2003). Developing understanding through problem solving. In J. Kilpatrick, W. G. Martin, & D. Schifter (Eds.), A research companion to Principles and Standards for School Mathematics (pp. 227–236). NCTM.
• Isotani, S., Adams, D., Mayer, R. E., Durkin, K., Rittle-Johnson, B., & McLaren, B. M. (2011). Can erroneous examples help middle school students learn decimals? Towards Learning and Instruction in Web 3.0, Springer.
• Boaler, J., & Brodie, K. (2004). The importance, nature, and impact of teacher questions. Proceedings of the 26th PME-NA.
• Chi, M. T. H., Roy, M., & Hausmann, R. G. M. (2008). Learning from observing and self-explaining alone: Are both as effective as learning from tutoring? Applied Cognitive Psychology, 22(6), 795–822.
• Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. doi:10.3102/003465430298487
• Durkin, K., & Rittle-Johnson, B. (2012). The effectiveness of using incorrect examples to support learning about decimal magnitude. Learning and Instruction, 22(3), 206-214. doi:10.1016/j.learninstruc.2011.11.001
• Boaler, J., & Brodie, K. (2004). The importance, nature, and impact of teacher questions. Proceedings of the 26th Annual Meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education.
• Kapur, M. (2014). Productive failure in learning math. Cognitive Science, 38(5), 1008–1022. doi:10.1111/cogs.12107