Self-Explanation: How to Study by Explaining Each Step

Self-explanation means stopping as you study to explain to yourself why each step or statement is true and how it connects to what came before. A 2018 meta-analysis found a moderate overall benefit, and a 2017 meta-analysis in math found the evidence for classroom use and long-term retention more limited. It interrupts the feeling of nodding along to a worked solution that looks reasonable line by line.

1. What self-explanation is

The 2018 meta-analysis by Kiran Bisra and colleagues defines self-explanation as a process by which learners generate inferences about causal connections or conceptual relationships. A 2017 meta-analysis by Bethany Rittle-Johnson, Abbey Loehr, and Kelley Durkin describes it more simply as generating explanations for yourself in an attempt to make sense of new information.

The key word is "yourself." You are not explaining to a teacher or classmate, and you are not just restating the text. You are answering questions like "why does this step follow?" or "how does this fit with the rule we learned last week?"

2. What the big meta-analysis found

A meta-analysis combines results from many studies. Bisra and colleagues searched the research databases and found 69 effect sizes from 64 research reports in which participants received self-explanation prompts while studying or solving problems.

The overall weighted mean effect size was g = 0.55, a moderate and meaningful benefit by common standards in education research. The researchers coded 20 moderator variables, including the type of learning task (solving problems, studying worked problems, studying text), subject area, level of education, and how long the treatment lasted. They concluded that self-explanation prompts are a potentially powerful intervention across a range of instructional conditions.

3. What the math-focused meta-analysis adds

The 2017 meta-analysis looked specifically at prompting self-explanation in mathematics. Its findings were more cautious:

• Prompted self-explanation led to a small to moderate improvement in procedural knowledge, conceptual knowledge, and procedural transfer when tested immediately after the intervention.
• Evidence that it reliably helps within a regular classroom, or that the benefit lasts over a delay, was much more limited.
• The effect on immediate outcomes was stronger when learners got help producing high-quality explanations.

4. Where the broader evidence stands

The 2013 review of ten learning techniques by John Dunlosky and colleagues rated self-explanation as having moderate utility. The reason it did not rate higher was not that it fails, but that it had not yet been adequately evaluated in educational contexts. That matches the math meta-analysis: good results in research settings, with less known about long-term classroom use.

So the fair summary is this: self-explanation is a well-supported technique with a moderate effect, and it is worth using, but that review placed it a step below the techniques it rated most highly.

5. How to self-explain well

The 2017 math meta-analysis proposes four recommendations for mathematics educators. Each translates into something you can do on your own:

1. Learn what a good explanation looks like. It recommends scaffolding high-quality explanations. On your own, compare your explanation with the textbook's reasoning, or use a fill-in structure such as "This step works because ___."
2. Keep prompts short and focused. It says explanation prompts should not take attention away from other important content. One focused question per step beats a long essay.
3. Explain correct information. Self-explain from a correct worked example or a reliable text, so you are building understanding of the right method.
4. Explain why common mistakes are wrong. It also recommends prompting learners to explain why common misconceptions are incorrect. For example: "Why can I not cancel the x terms in (x + 3)/x?"

6. Useful prompts to write in the margin

• Why does this step follow from the one before it?
• What rule or definition is being used here?
• What would go wrong if this step were skipped?
• How is this example similar to, or different from, the last one?
• Why is the tempting wrong answer wrong?

7. How this differs from asking deep questions

You may have seen advice to ask yourself deep "why" questions about course material. That is a close cousin. The difference is focus: self-explanation works step by step through the material in front of you, such as each line of a worked solution, while broader questioning can range over a whole topic. They work well together.

8. Key takeaways

• Self-explanation means generating your own explanations of why each step or idea makes sense while you study.
• A 2018 meta-analysis of 69 effect sizes found an overall moderate benefit (g = 0.55) across many tasks and subjects.
• A 2017 math meta-analysis found small to moderate immediate gains, with limited evidence so far for classroom use and long-term retention.
• A 2013 review rated self-explanation as moderate utility mainly because it had not been tested enough in real classrooms.
• Use short, focused prompts on correct examples, and also explain why common mistakes are wrong.