Human nutrition produces some of the most frustrating written answers in science. Learners know the content, describe it accurately, and still lose most of the available marks. The gap is not knowledge; it is that nutrition questions are almost always asking for a mechanism or a comparison while learners answer with a definition. Across a Year 9 class in Wales working towards progression step 4, four recurring errors account for the bulk of the lost marks, and each one is fixed by modelling rather than by more content.
Error one: describing when the question says explain
Asked why a diet lacking protein causes poor growth, learners write that protein is needed for growth. That is the question restated, and it earns nothing. The command word explain requires a chain: protein provides amino acids, amino acids build new tissue, without them cells cannot produce new structural proteins, so tissue growth slows.
Model this as a physical chain. Four cards, four links, laid on the desk in order, then written down. Do it three times with different examples and require learners to count their own links before submitting. A three-link answer to a three-mark question is a reliable habit and it removes this error faster than any amount of feedback on individual scripts.
Error two: naming a nutrient instead of naming a function
Learners have nutrient lists memorised and reach for them under pressure. Asked about the consequence of a deficiency, they produce the name of the nutrient and the name of the deficiency disease, then stop. The examiner wanted the physiological role that was lost.
The fix is a two-column drill that never becomes a matching exercise. Left column: the nutrient. Right column: the sentence beginning "without it, the body cannot...". Learners fill the right column from memory and are not allowed to write a disease name in it. Macronutrient Metabolism: Carbohydrates, Lipids & Proteins is a strong source for building that drill, because the material is organised around what each macronutrient does rather than around lists to be recited.
Error three: treating a data question as a knowledge question
Give a table of intakes against recommended values and learners write everything they know about balanced diets without quoting a single figure. This is the most costly error of the four because data questions carry a lot of marks and are the most predictable part of any paper.
- Rule one: quote two numbers. Not one. A comparison needs both sides.
- Rule two: use the units. A number without a unit is treated as an unsupported claim.
- Rule three: state the direction and the size. Higher is a weak answer; nearly double is a strong one.
- Rule four: only then bring in the biology. The explanation comes after the data, never instead of it.
Practise this on real cases where the data genuinely matters, since invented tables produce lazy reading. Food Intolerances & Human Nutrition Bundle is useful here because intolerance cases come with figures that have to be interpreted rather than recalled, and the differentiated levels mean weaker readers still meet a real table rather than a simplified one.
Error four: writing about health rather than about biology
Nutrition is the topic where learners bring the most outside information, and much of it is advice rather than science. Answers drift into what people should eat, which is a moral claim, when the question asked what happens in the body, which is a mechanism.
Name the drift explicitly and give it a label the class uses. Call it the magazine answer. Once learners can identify it in someone else's script they stop producing it in their own. Then reteach the mechanism underneath the claim, because usually the drift is covering a genuine gap. Conditions where the biology and the popular commentary diverge sharply are the best teaching ground for this, and Diabetes & Obesity Teaching Resources | High School Biology | Human Physiology & Health Science handles that territory with the physiology kept front and centre, which is exactly what the written answers need.
The modelling routine that removes all four
One question a week, fifteen minutes, in this order. You write an answer live on the board and get it wrong on purpose in one of the four ways. Learners identify which error it is using the class labels. You rewrite it correctly, thinking aloud, while they annotate the difference. Then they answer a parallel question and swap for a peer check against the same four labels.
Fifteen minutes a week for a term is roughly six hours, and it does more for results on this topic than six hours of additional content ever would. The signal that it has landed is when a learner reads a question, underlines the command word before writing anything, and says out loud that this one needs numbers.


Comments
No comments yet — be the first to share your thoughts!
Leave a comment
Comments are reviewed before being published.
Thanks for your comment!
Your comment is being reviewed and will appear here shortly.