Why self-assessment belongs in open resources
Many users of open educational resources study without a teacher. They have no one to mark their work or tell them whether they have understood. Self-assessment tools fill part of that gap, and they also help students inside classes by giving fast, private feedback. Done well, they encourage learners to notice what they know, what they do not, and what to do next.
There is a limit worth stating early. People are often poor judges of their own understanding, especially when material feels familiar. Self-assessment supports learning but does not replace external feedback, and its effect varies between learners and subjects. The designs below aim to make honest self-checking easier, without promising particular results.
Quizzes that help learning
A quiz inside a resource has two possible jobs: showing the learner where they stand, and helping them remember. Research on learning suggests that trying to recall information, rather than simply re-reading it, tends to help memory, an approach often called retrieval practice, although how much it helps depends on the task and learner. A well-designed self-check quiz makes use of this.
Design suggestions:
- Keep quizzes short and frequent. A few questions at the end of each unit works better than a long test at the end of a course.
- Match questions to objectives. If you wrote learning objectives for the unit, each question should connect to one, and the learner should be able to see which.
- Vary question types. Multiple choice is easy to mark but can be answered by guessing or recognizing an answer without truly knowing it. Short-answer and “explain in a sentence” questions require recall, and can be self-marked against a model answer.
- Write good wrong answers. In multiple choice, wrong options should reflect common misunderstandings, so a chosen error tells the learner something.
- Mix old and new. Including a few questions from earlier units supports revisiting material and fits with the idea of interleaved practice, in which topics are mixed and not studied in blocks. Questions that require some effort, sometimes called desirable difficulties, can aid long-term learning, but should not be so hard that learners give up.
- Allow retries. Letting learners try again, perhaps with different questions, supports improvement.
Feedback is where most quizzes succeed or fail. Do not stop at “correct” or “incorrect”. Explain why the right answer is right and, for common mistakes, why the other options are not. Point to the place in the resource where the idea is explained. Ask a colleague to check that answer keys are accurate, since an error in a key misleads self-studying learners who have no way to catch it.
For flashcard-style recall practice, advice on flashcard design applies: one idea per card, clear wording, and an answer you can check. Learners may also benefit from revisiting items over increasing intervals, which ties to the forgetting curve and the principle of spaced repetition.
Rubrics learners can use themselves
A rubric lists criteria and describes what different levels of performance look like. It is usually a marking tool, but it also works as a self-assessment tool when written in language students can apply to their own work.
To write a student-friendly rubric:
- Choose a small number of criteria, perhaps three to five, that matter for the task.
- Describe each level in concrete terms. “Uses at least two pieces of evidence and explains how each supports the point” is more usable than “good use of evidence”.
- Avoid vague words such as “excellent” or “poor” without describing what they mean.
- Present the rubric before the task, not afterward, so students know the target.
- Include a space for students to note evidence from their own work for each criterion, and what they plan to improve.
Provide an example of work at different levels where you can, with a note that these are illustrative. Remember that a rubric describes qualities, so it should not make creative or open-ended work formulaic.
Students often rate themselves generously at first. Comparing their rating with a peer’s or teacher’s view, where available, can help them calibrate. Where no one else is available, encourage them to find specific evidence for every rating rather than relying on a feeling.
Journals and reflective prompts
A learning journal is a regular, short, written reflection. For self-assessment, useful prompts include:
- What can I now do that I could not do before this unit?
- Which questions did I get wrong, and why?
- What is still unclear? What will I do about it?
- How confident do I feel, and what is that based on?
- How did I study, and what would I change next time?
Such prompts encourage metacognition skills: thinking about one’s own thinking and learning. A simple confidence rating before checking an answer, followed by a comparison with the result, can show learners where their confidence and their performance differ. Frame this as information, not judgment.
Make journals optional or flexible in length, and let learners decide what to share if a teacher reads them. Some prompts can invite personal disclosure, so avoid asking for sensitive details. Provide the journal template in editable and printable formats so it works on or off a device.
Tracking progress over time
Self-assessment is more useful when it accumulates. A checklist of objectives that learners can tick and date, or a simple table of quiz scores by unit, lets them see a pattern. This links to approaches such as progress tracking in study habits more generally. In courses organized around competency-based education, where learners move on when they can show mastery, such checklists are especially natural. Keep tracking light and private, and avoid presenting scores as a judgment of ability.
Offer a template learners can copy, in an editable format, and explain that all of this can be done on paper.
Online and offline considerations
Automated quizzes in a browser are convenient, but not every learner can use them. Provide a printable version with an answer key, placed separately so learners are not tempted to peek. If you build quizzes as self-contained pages, they can run offline, though results cannot then be sent anywhere. Say plainly what is and is not stored. Learners should know that their quiz responses stay on their own device unless told otherwise, and any tool that collects data should say what it collects and why.
For online assessment more generally, use well-established design rather than complex features, and check that tools work with keyboards and screen readers.
Common weaknesses to avoid
- Questions that test trivia, not understanding.
- Answer keys with errors or no explanation.
- Rubrics full of vague words.
- Reflection prompts so generic that learners write the same thing every time.
- Claims that using the tool will raise grades.
- Testing so often that it discourages learners.
A short checklist
Before publishing, check that each unit has a few questions linked to its objectives, feedback that explains, a rubric or checklist where a task is open-ended, a reflection prompt or two, and a way to record progress. Try the materials yourself as a learner would, and ask someone else to try them, noting where instructions confuse. Authentic assessment, which asks learners to apply knowledge in realistic tasks, can supplement these tools, particularly for skills that a quiz cannot capture.