Emphasis on Reasoning over Answers¶

Strategy · All strategies
Evidence · 5 claims (5 for) · 11 studies (4 causal, 4 quant-synthesis, 1 review, 1 associational, 1 theoretical), q2–q4 · 3 of 11 report an effect size

Description¶

This strategy reorients instruction and classroom discourse away from evaluating final answers and toward examining the reasoning that produced them. Teachers elicit initial ideas before answers are settled, put worked examples with errors on the board for students to diagnose, and ask learners to describe their first steps rather than their conclusions. The result is a more discursive classroom in which being wrong is treated as informative rather than as failure, which builds confidence and supports conceptual learning Self-explanation improves conceptual understanding. [+S].

Design Implications¶

Valuing reasoning over answers converts errors from endpoints into learning opportunities: analyzing why an answer is wrong produces stronger conceptual knowledge than simply being told the right one Erroneous examples build conceptual knowledge. [+M]. It also aligns feedback with what actually drives improvement — feedback at the process level outperforms feedback at the answer level Feedback is most effective at task and process levels. [+S].

Context¶

Requirements¶

• Tasks and discussion norms that make reasoning visible and discussable, not just answers
• Assessment and grading schemes that reward process quality, not only correctness
• Instructor skill in probing ("What made you try that?") rather than evaluating ("Right/wrong")
• Psychological safety: learners must believe that sharing incomplete thinking carries no penalty

Constraints¶

• Learners with strong answer-focused habits may initially experience the shift as ambiguous or unfair, reducing engagement until norms are established [-W]
• If reasoning talk is unstructured, low-quality or circular explanations can be reinforced rather than corrected [-M]
• Requires more instructional time per problem than answer-checking; coverage pressure can erode the practice [-W]
• For routine procedural skills with a single correct method, answer accuracy remains a legitimate target and reasoning emphasis adds little [~M]

Implementation Variability¶

• Error analysis: present a worked example with a wrong answer and have students find and explain the error Erroneous examples build conceptual knowledge. [+M]
• Answer-deferral discourse: ask for initial ideas and first steps before anyone states an answer ("I don't want the answer yet — tell me the first thing you thought about")
• Struggle-first task design: let learners attempt problems before instruction, then compare approaches [~M] — productive failure can outperform direct instruction first for deep conceptual goals, though it can depress performance on routine procedures
• Process-focused grading: rubrics that score strategy selection, justification, and revision separately from the final answer

Target Learners¶

• Learners who are answer-focused or fear being incorrect; normalizing reasoning reduces avoidance and risk-taking costs [~W]
• Novices benefit especially from process-level feedback and explanation prompts Feedback is most effective at task and process levels. [+S]; process goals also outperform outcome goals for novices Process goals outperform outcome goals for novices. [+M]
• Learners with high confidence in a wrong idea benefit from error diagnosis, since high-confidence errors are especially memorable when corrected High-confidence errors improve retention. [+M]

Target Learning Goals¶

• Conceptual understanding: why procedures work, not just how Self-explanation improves conceptual understanding. [+S]
• Metacognition and self-reflection: monitoring one's own strategy choices
• Mathematical discourse and argumentation practices
• Growth-oriented beliefs about ability and error

Instructions¶

1. Select or design a task with multiple solution paths or a plausible error, so reasoning is genuinely discussable.
2. Elicit initial thinking before answers: have learners describe their first step or idea, in pairs or aloud, before any answer is confirmed.
3. Present an erroneous worked example and have students locate and explain the error Erroneous examples build conceptual knowledge. [+M].
4. Prompt self-explanation: ask learners to justify each step in their own words Self-explanation improves conceptual understanding. [+S].
5. Give feedback on the process — strategy choice, justification quality, where reasoning went off track — before or instead of verdicts on the answer Feedback is most effective at task and process levels. [+S].
6. Align assessment: use rubrics and formative assessment that score reasoning explicitly Assessment for Learning.

Related Strategies¶

• Error Analysis with Erroneous Examples — a specific enactment: diagnosing wrong answers builds conceptual knowledge more than studying correct ones alone
• Productive Failure — structured struggle before instruction, leveraging the same principle that reasoning attempts have value independent of answers
• Self-Explanation Prompts — the individual, written counterpart to whole-class reasoning discourse
• Question students about why a correct answer was chosen, rather than why an error was made
• Design learning-oriented LLM prompts around guided questions, step-by-step support, reflection, and a thinking partner rather than direct answers

Examples¶

• A teacher puts a worked algebra problem with a sign error on the board and asks, "Find where the reasoning went wrong" — students must articulate the error, not just supply the right answer.
• A teacher responds to a raised hand with, "I don't want to hear the answer just yet. I want to hear about the first thing you thought about," making the classroom norm explicit.
• Youcubed (Stanford, Jo Boaler) — published number-sense tasks and teacher resources built around valuing reasoning, multiple representations, and normalizing mistakes in mathematics classrooms.
• Kapur's Productive Failure studies — Singapore math classrooms where students first attempt unsolved problems in groups, then compare their (often wrong) reasoning to canonical solutions.

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
• 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
• Kapur, M. (2014). Productive failure in learning math. Cognitive Science, 38(5), 1008-1022. doi:10.1111/cogs.12107
• Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. doi:10.3102/003465430298487
• 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.