How to Write a Lab Report Discussion Section

To write a lab report discussion, explain what your results mean, compare them with theory or published values, and analyse the errors that explain any difference. It is where you show understanding, and it often carries the most marks in the report.

This guide gives a paragraph-by-paragraph structure, shows how to compare results and analyse error, and works through an example from a pendulum experiment. It ends with common mistakes, a checklist and how STEM Donkey can help.

What the Discussion Section Is For

The results section reports what you found. The discussion explains why it came out that way and what it means.

Some courses combine results and discussion, or discussion and conclusion. Follow your lab handbook, but keep the jobs distinct within the text.

A Paragraph-by-Paragraph Structure

1. Main finding. Restate the key result with its uncertainty and say whether it supports the hypothesis or aim.
2. Comparison. Compare it with the theoretical, accepted or published value, using percentage difference and uncertainty.
3. Patterns and trends. Explain the shape of graphs, gradients, intercepts and any anomalous points.
4. Error analysis. Identify specific random and systematic errors, and state which way each would shift the result.
5. Limitations. Note assumptions in the method or theory that limit the conclusions.
6. Improvements. Suggest specific changes linked to the errors you identified.

Short reports may combine some of these into a single paragraph. Longer ones may give several paragraphs to error analysis alone.

Comparing Results with Theory

Comparison should be numerical. Two quantities help.

• Percentage difference: |measured − accepted| / accepted × 100. It shows how far off the result is.
• Agreement within uncertainty: check whether the accepted value lies within your stated uncertainty range. If it does not, there is probably a systematic error, or the uncertainty was underestimated.

A result can be precise but inaccurate (small uncertainty, far from the accepted value) or accurate but imprecise (close to the value, large uncertainty). Say which describes your result, because they point to different problems.

For the pendulum example below, the percentage difference is |9.62 − 9.81| / 9.81 × 100 ≈ 1.9 percent, while the relative uncertainty is 0.08 / 9.62 × 100 ≈ 0.8 percent. The gap is more than twice the uncertainty, which signals a systematic effect worth investigating.

Comparing with Published Studies

In upper-level reports, compare your findings with published work, not only with a textbook value. Say whether your result agrees, and if it does not, whether differences in method, materials or conditions could explain the gap.

Cite each study you compare with. One or two well-chosen comparisons, discussed properly, are worth more than a long list of sources mentioned in passing.

Random and Systematic Errors

Do not list "human error" as an explanation. Name the specific action, such as starting the stopwatch late, and state its likely size and direction.

A good test: if a source of error would push your result in the direction opposite to the difference you observed, it cannot explain that difference on its own.

A Worked Example: Measuring G with a Pendulum

Notice what the paragraph does. It gives the percentage difference, checks agreement against the uncertainty, names specific errors, states their direction and links each improvement to an error.

Discussing Graphs, Trends and Anomalies

• Explain what the gradient and intercept represent physically, with units.
• Compare the gradient with its theoretical value, using the uncertainty from the line fit.
• Explain a non-zero intercept where theory predicts zero; it often reveals a systematic error.
• Discuss anomalous points honestly. Say whether you excluded any and give the rule you used.
• Comment on R² only as a measure of fit, not as proof that the model is correct.

Limitations and Improvements

Improvements earn marks when they are specific and tied to an identified error. Generic suggestions earn little.

Also note limitations in the theory itself, such as the small-angle approximation, an ideal gas assumption or a neglected air resistance term. Explain how each assumption could affect your result and whether its effect is likely to be large or small compared with your measurement uncertainty.

Common Discussion Mistakes

• Repeating the results without interpreting them.
• Claiming results "prove" a theory; experiments support or fail to support a hypothesis.
• Blaming vague "human error" or "faulty equipment" with no detail.
• Listing errors whose direction does not match the observed difference.
• Introducing new data that was not in the results section.
• Ignoring uncertainty when judging agreement with theory.

Writing a discussion is steady work: one result, one comparison, one error at a time. Taken in order, even a difficult set of results becomes manageable.

Discussion Checklist

1. The main result is restated with its uncertainty and units.
2. It is compared numerically with the expected value.
3. Agreement within uncertainty is checked.
4. Graph features and anomalies are explained.
5. Specific errors are named, with their direction and likely size.
6. Improvements are specific and linked to those errors.
7. Claims are supported by the data in the results section.

How STEM Donkey Helps with Lab Report Discussions

Send your lab brief, your raw data and results, and any marking criteria. A writer matched to your subject prepares a custom discussion, or a full lab report, that interprets your findings, compares them with theory, analyses errors quantitatively and suggests real improvements.

Your measurements are used exactly as recorded; data is never invented or adjusted. Free revisions within the original scope are included for 14 days, and you see the price before you pay.