Multiple-Choice vs Structured Questions: When to Use Each Under CBC

Every exam-setter eventually faces the same question for every learning outcome on their blueprint: multiple choice, or structured? The honest answer is that neither format is universally better — they're built to do different jobs, and CBC papers at Junior School level typically use both in the same paper for exactly that reason.

What Multiple Choice Does Well

MCQs are fast to mark — a real practical advantage when you're grading a full class, or several streams, under time pressure. They're also good for covering breadth: because each question takes a learner relatively little time to answer, you can test a wide spread of topics across a single paper rather than concentrating your marks on just a handful of areas.

Cognitively, MCQs work best at the Understand and early-Apply levels — recognising a correct fact, applying a single rule, classifying something correctly. This maps directly onto the Upper Primary band, where the standard format is a 4-option A-D question with exactly one correct answer, worth one mark each.

The catch is construction quality. An MCQ is only as good as its distractors — the three wrong options need to be plausible and in the same grammatical form as the correct answer, or the question stops testing knowledge and starts testing guessing.

What Structured Questions Do Well

Structured questions can do things an MCQ simply can't: they can test multi-step reasoning, require a learner to show their working, and reach higher up Bloom's ladder into Analyse and beyond. A structured question can ask a learner to explain why something happens, work through a calculation across several steps, or compare two situations and justify a conclusion — none of which fits into a single tick-one-box answer.

They also let you see the learner's process, not just their final answer, which matters most in subjects like Mathematics and the sciences where a wrong final answer might still reflect mostly-correct reasoning. That's exactly why structured marking schemes are built around per-sub-part marks with nested (a)/(b)/(c) parts — the format itself is designed to reward the steps along the way, not just the destination.

The tradeoff is time: structured questions take longer for learners to answer and longer for you to mark, so a paper built entirely of structured questions covers far less ground than one that mixes in MCQs.

How Real KNEC Papers Combine Both

This is exactly why Junior School papers commonly pair an MCQ section (often Paper 1) with a structured-question section (Paper 2) rather than choosing one format for the whole subject. The MCQ section gives broad coverage across many topics at the Understand/Apply level; the structured section goes deeper on a smaller number of outcomes that call for Analyse-level reasoning or multi-step work. Together, the two sections give a more complete picture of a learner's ability than either format alone.

Choosing Per Learning Outcome, Not Per Subject

The most useful way to decide is to look at each individual learning outcome on your blueprint and ask what it's actually asking learners to do:

• If the outcome is about recognising, identifying, or applying a single rule — an MCQ will test it efficiently and let you cover more outcomes in the same paper.
• If the outcome is about explaining a process, comparing two things, working through a calculation, or reasoning through a scenario — a structured question is the only format that can actually capture whether a learner can do that.

Trying to force a genuinely analytical outcome into an MCQ usually produces a weak question — either the four options give away too much, or the "correct" analysis gets flattened into something a learner can guess. Equally, writing a structured question for something as simple as a single fact wastes time that could have covered three or four other outcomes as an MCQ.

The Practical Takeaway

Don't start by deciding "this paper will be all MCQ" or "this section will be all structured." Start with your blueprint of learning outcomes, match each one to the cognitive level it calls for, and let that match decide the format. The result — a paper that blends both, the way real KNEC papers do — will give you a fairer and more complete read on what your learners can actually do.