A step carries a question about meaning
When solving an equation, the next transformation matters, but so does understanding why it preserves equality. In a word problem, important work happens before calculation: what do the quantities represent, and what relationship does the task describe? These decisions can be central to learning even when the final result looks simple.
This does not mean demanding an essay for every operation. A brief explanation at a useful point can begin a conversation. The purpose is to make the reason behind an important step understandable, while allowing more than one valid solution approach.
Worked examples need deliberate design
Atkinson and colleagues review worked-example research, discussing organised steps, integrated example components and connections between examples and practice. Useful examples let learners follow a method, especially early in skill development; merely displaying a complete solution does not guarantee learning. Original review.
A mathematics self-explanation meta-analysis found small-to-moderate benefits for immediate learning outcomes. Evidence for delayed retention and classroom use was more limited, and explanation quality and scaffolding matter. This does not establish that every request to explain an answer will help. Original research abstract.
From following a solution to choosing a step
Instead of asking learners to repeat every line of a worked solution, we can focus on a particular decision. Why was this operation used? How does one line relate to the previous line? How can the learner tell that the method suits the question? These are possible prompts selected for a need, rather than a compulsory checklist.
After the example, the learner needs an opportunity to try independently. A changed wording or arrangement of information can make choosing an idea part of the task. Transfer to independent practice should be investigated; it cannot be inferred merely from viewing a clear explanation.
Understand an error before judging it
An incorrect line may arise from incomplete understanding, a copied symbol, or a representation the system cannot read. Confusing a learner’s mistake with a system’s reading error could produce confident feedback about something the learner never wrote.
A valid solution may also differ from the reference example. Responsible step assessment therefore needs criteria that allow diverse correct paths, alongside clear handling of unreadable or incomplete attempts. These boundaries define responsible assessment in our approach.
The step in Warda’s approach
Warda’s approach connects curriculum-context explanations with practice that leaves room for a learner’s attempt. A suitable hint does not explain everything at once: it helps the learner return to a particular point in the solution. Its role is interpreted in relation to the activity’s purpose, whether learning practice or assessment.
Our mathematics vision distinguishes reading the learner’s work, verifying a transformation and interpreting its educational meaning. These are separate responsibilities. Conversation quality cannot replace correct content or evidence of understanding; separating them makes feedback more specific and examinable.
What matters in the experience?
Warda’s vision lets the learner return to a particular step, understand its purpose and try the idea without copying a solution. Explanation timing and assistance scope matter: support for a first encounter may interrupt someone practising a familiar idea.
Taking steps seriously clarifies design responsibilities: an explanation with a purpose, a judgement with criteria and room for the learner to work. Feedback is useful when connected to the written attempt and its evidence limits, followed by an understandable route to another attempt.
Sources and context
- Atkinson et al. (2000), Learning from Examples: Instructional Principles from the Worked Examples Research
A review of worked-example research and design, not evidence that an automated grader works.
- Rittle-Johnson, Loehr & Durkin (2017), Promoting self-explanation to improve mathematics learning
Based on the original meta-analysis abstract; delayed retention and classroom evidence are limited.
Sources describe research, specifications or documented product behaviour, as identified above. They did not evaluate Warda or establish its effectiveness.