Why not just start with problems?
A common way to learn a skill like algebra or programming is to read a short explanation and then try many problems. For someone who is new to the topic, that can be a poor use of effort. With no clear method in mind, a beginner often tries random approaches, gets stuck, and spends mental energy on searching rather than learning. A worked example turns this around: you first study a problem with its complete solution, step by step, and only then try similar problems yourself.
The advantage was identified within cognitive load theory, developed by John Sweller and colleagues, and it is usually called the worked example effect. Studies, mostly in areas such as mathematics, physics and programming, have often found that novices learn more efficiently from studying examples than from solving the same number of problems unaided. The mechanism fits what is known about working memory: solving from scratch takes up limited capacity with trial and error, while studying a solution lets that capacity go to understanding the structure.
Not a trick for everyone, all the time
The benefit depends on the learner’s stage. As people gain skill, the examples that helped begin to get in the way. This is known as the expertise reversal effect: detail that is helpful to a beginner can be redundant for someone who already knows the steps, and may slow them down. So worked examples are best used early, with support removed as competence grows. That removal is known as fading.
The effect is also not guaranteed in every subject or context. It has been studied most in well-structured domains where problems have clear solution steps. In open-ended areas, such as essay writing or design, the idea is less cleanly tested, though models of good work still play a role.
What a good worked example looks like
A worked example is more than an answer. It should show the reasoning. Features that tend to help:
- Clear steps. Each step is shown and, where it is not obvious, briefly explained.
- A single idea. The example should illustrate one method, not several at once.
- Visible structure. Group related steps and label the parts, for instance “set up”, “calculate”, “check”.
- Close integration. Put explanations next to the step they refer to, rather than forcing the reader to flip between a diagram and a paragraph. This is a point from cognitive load research about split attention.
- Variety over time. After a few examples of the same type, include examples with different surface features so learners see what stays the same.
A small example
Here is a short, easily checked example from algebra. Problem: solve 3x + 5 = 20.
- Subtract 5 from both sides, keeping the equation balanced: 3x = 15.
- Divide both sides by 3: x = 5.
- Check by substituting: 3 × 5 + 5 = 15 + 5 = 20, which matches.
The aim is not just the answer, but the reasoning pattern: undo operations in reverse order and do the same thing to both sides. Notice also the final check, a habit worth modeling. The article on solving linear equations step by step goes through this kind of procedure in more depth, and the first lessons of algebra basics cover what the letters mean.
Fading support
Fading means that, across a series of problems, you gradually take over more of the work. One common sequence, sometimes called backward fading, removes steps from the end:
- Full example. Study a completed solution.
- Last step missing. A second problem is shown with all but the final step done; you finish it.
- Last two steps missing. The next problem leaves more to you.
- Full problem. Eventually you solve it from the start, without help.
Using the equation above, step 2 might show “4x + 3 = 19, so 4x = 16” and ask you to finish. In step 3 you might get only the first line, and then you would solve “2x + 7 = 15” alone. Check: subtracting 7 gives 2x = 8, so x = 4, and 2 × 4 + 7 = 15.
Making it active: self-explanation
Reading an example passively can create the same illusion of understanding as rereading. A useful habit is self-explanation: after each step, ask yourself why it is valid and what it achieves. Cover the next step and try to predict it. Compare two examples and ask what is the same. This links to the broader advice on metacognition skills, which stresses checking understanding, not just following along.
It also pairs naturally with retrieval practice. After studying an example, close it and try to reproduce the solution from memory, then compare.
Where this appears outside math
- Programming. Reading a short, working piece of code, tracing what each line does, then modifying it or completing a half-finished version.
- Science. Following a worked calculation in physics or chemistry, then solving a similar one with only a first-step hint.
- Practical skills. Watching a demonstration, then doing it with a checklist, then without.
Cautions
- Do not stop at examples. Learning to solve problems requires doing them. If you only ever read solutions, you may find you cannot start unaided. The fading sequence exists to prevent that.
- Choose examples carefully. A poorly explained or incorrect worked example can teach errors. Check against a reliable source.
- Mix once you are ready. When you can handle each type separately, practice them in shuffled order, as described in interleaved practice.
What to take from this
For beginners, studying well-explained solutions before attempting problems is a well-supported way to learn procedures without being swamped. The skill is in the handover: step by step, the hints come off and the work becomes yours. Research results vary with subject and learner, so watch your own progress, and let it decide how quickly you move on.