A lab report is not an essay about an experiment. It is a record of an experiment, written so that another scientist can see exactly what you did, what you found, and what you think it means. Markers judge it on different criteria from essays: reproducibility, precision, honest handling of uncertainty, and interpretation that goes beyond restating numbers. This guide works through each section in the order you should write it — which is not quite the order it appears on the page — with the mistakes that cost marks alongside the advice. If you want a worked model for your own experiment, our lab report help page explains how that support works.
Before you write: turn lab notes into a plan
Good reports are written from lab notes, not from memory. Before drafting, gather your notes, raw readings, calculations and any graphs you plotted during the practical, and write a one-sentence aim: what question was this experiment trying to answer? Then list, in order, what you actually did — including anything that went wrong, any reading that looked odd at the time, and any step you repeated. That list becomes the skeleton of your methods section, and the anomalies on it become material for your discussion. Students who skip this step usually discover, halfway through the discussion, that they cannot remember a concentration or a setting, and the report suffers for a gap that good notes would have prevented.
Title and abstract: short, specific, informative
The title should name the experiment and, where possible, the key variable or system studied — “The effect of temperature on the rate of amylase activity” rather than “Enzyme practical”. The abstract, where your department requires one, is a miniature of the whole report in around 150 to 250 words: aim, method in brief, headline result, and conclusion. Write it last, even though it appears first, because you cannot summarise a report you have not yet written. A common fault is an abstract that promises results the report never delivers, or a conclusion the data never supported; check each abstract sentence against the body before submitting.
Introduction: the question, the theory, the aim
The introduction does three jobs and no more. First, it states the scientific background a reader needs: the theory, the key concepts, and any previous findings relevant to this experiment. Second, it narrows to the specific question this practical addresses — a gap, a relationship to test, a value to determine. Third, it states the aim or hypothesis precisely, in terms that the results section can actually answer. What it must not do is become a general essay on the topic: pages of textbook background that never connect to the experiment waste words and patience. Cite the sources you actually used, in your department’s style, and keep the introduction proportionate — roughly a tenth to a fifth of the report for most undergraduate practicals. End the introduction with the aim stated plainly, so the reader knows exactly what the rest of the report is trying to show.
Methods: write so someone else could repeat it
The methods section is the test of reproducibility. Write in the past tense and in your own words, even when the procedure is standard — copying a lab manual word for word is plagiarism, however standard the steps. Include the materials and equipment with enough detail to matter: concentrations, volumes, instrument settings, sample sizes and number of replicates. Describe the procedure in the order performed, as a coherent account rather than a numbered recipe, unless your department prefers numbered steps. State how data were recorded and analysed: which quantities were calculated, which statistical tests were used, and how uncertainties were estimated. Leave results out entirely — no “the solution turned blue, showing the reaction worked” belongs here; that is a result, and it goes in the next section. A useful check: hand the methods to a classmate and ask whether they could repeat the experiment from it alone. If they cannot, something is missing.
Results: show the data, then say what stands out
Present results in text, tables and figures that complement each other rather than repeating the same numbers three times. Tables are for exact values a reader may need to check; figures — graphs, spectra, photographs — are for patterns, trends and comparisons. Every table and figure needs a number, a self-contained caption, labelled axes, stated units and, where relevant, error bars whose meaning is explained. In the text, guide the reader to the point of each display: “Absorbance rose linearly with concentration up to 0.8 mol dm⁻³, then levelled off (Figure 2)” — rather than listing every reading the figure already shows. Report statistics properly: the test used, the sample size, and the significance level, not a bare claim that a difference was “significant”. Include results that did not fit expectations, and do not round away precision your instruments never had: significant figures should match the measurement, not your calculator’s display.
Discussion: interpret, compare, own the limitations
The discussion is where most lab report marks are decided, and it is where most students underperform, because they restate the results instead of interpreting them. Work through four questions in order. First, what do the results mean for the aim or hypothesis — supported, contradicted, or partly supported, and how strongly? Second, how do they compare with theory and published values — and if they disagree, is the experiment, the model or a shared assumption the more likely source of the gap? Third, what are the real limitations, and what do they imply: does a particular uncertainty weaken the conclusion, or does the effect survive even allowing for it? Fourth, what would you change or test next — a larger sample, a tighter control, a different range — stated specifically rather than as “more research is needed”. Own anomalies rather than hiding them: an outlier noted in your lab book and explained is worth more than one quietly deleted. And never end with a claim the data cannot carry: a single undergraduate practical rarely overturns a theory, and markers trust a measured conclusion far more than a sweeping one.
Conclusion and references: close cleanly
The conclusion, where required separately, is short: restate what the experiment established, what remains uncertain, and any recommendation that follows. Introduce no new data here. The reference list then follows your department’s style exactly — ACS or RSC numbered styles in much of chemistry, Vancouver in biomedical work, Harvard author-date in some biology and environmental programmes. Check every in-text citation against the list and vice versa: an uncited source in the list, or a citation with no entry, is an easy mark lost. Cite lab manuals and manufacturer protocols where you relied on them, and cite data sources for any published values you compared your results against.
Common mark-losing mistakes to check before submitting
Run through this list on your finished draft. Missing units or wrong significant figures — check every number. Methods that cannot be repeated — check concentrations, settings and quantities. Figures with no interpretation — every figure should be discussed in the text. Uncertainty mentioned but never quantified — a sentence about “errors” with no numbers earns nothing; state the dominant uncertainty and its size. Discussion that repeats results — every discussion paragraph should contain interpretation, not just numbers. “Human error” as an explanation — name the actual source of error or do not claim one. Copied methods — rewrite procedures in your own words from your own notes. Fixing even three of these typically lifts a report by a visible margin, because markers see them so often that their absence is itself a signal of care.
Frequently asked questions
How long should each section be?
It depends on the experiment and your department’s template, but a useful rough guide for a standard undergraduate report is: introduction 10 to 20 per cent, methods 20 to 30 per cent, results 20 to 30 per cent, discussion 25 to 35 per cent. The discussion is usually the longest analytical section. Always follow your own brief where it gives section word counts or a template, since that overrides any general guide.
Should I write in the first person or the passive voice?
Follow your department’s preference, and be consistent. Many science departments still prefer the passive voice in methods (“The solution was heated to 60 °C”), while others accept “We heated the solution”. What matters most is clarity and consistency: do not switch between the two within a section, and never write instructions (“Heat the solution”) in a report of work already done.
What do I do with an anomalous result?
Report it, flag it, and discuss it. Check first whether it was a recording or calculation mistake; if it was, correct it and say so. If it is genuine, keep it in the data set unless your department’s stated outlier criteria justify exclusion, and explain in the discussion what might account for it. Deleting inconvenient data without comment is scientific misconduct in miniature, and markers treat it seriously.
How many references does a lab report need?
Enough to support the introduction’s background, the methods’ sources and the discussion’s comparisons — often a handful for a short practical, more for a project report. Quality matters more than quantity: a few well-chosen, genuinely consulted sources beat a long list padded with papers you never read. Every reference must be cited in the text and every citation must appear in the list.
Can I use the same figure in the results and the discussion?
There is no need to repeat a figure: the discussion refers back to the results figures by number (“as shown in Figure 2”). Reproducing the same figure twice wastes space and suggests the sections were written without a plan. If the discussion needs a new view of the data, such as a comparison with published values, that is a new figure with its own caption.
Write the report from your notes, let each section do only its own job, and give the discussion the time it deserves — that is most of what separates a competent lab report from a strong one. For subject-specific support, our lab report help page covers models across the laboratory sciences, and if your analysis involves larger data sets, our SPSS data analysis help page explains the statistical support available.