Writing a lab report that scores in the discussion section

Lab report marking schemes are unusually transparent. Method and results carry a fixed and fairly mechanical set of marks that most students collect. The discussion carries a larger share, is harder to write, and is routinely the section produced at midnight in a hurry. That is where the grade is decided.

Results and discussion are different sections

Results state what happened. Discussion says what it means. Mixing the two is the most common structural fault in student lab reports, and it costs marks in both sections because the results become cluttered and the discussion becomes repetitive.

What a discussion has to contain

• Whether the results support the hypothesis, stated directly
• How the findings compare with published values or expected theory
• Explanations for any deviation, specific rather than generic
• An honest treatment of uncertainty and its dominant source
• What you would change if you repeated the experiment, and why

Anomalies are opportunities

A result that came out wrong is not a problem for your mark. Failing to engage with it is. Explaining why a value deviated, whether it is within uncertainty, and what mechanism could account for it demonstrates exactly the understanding the discussion section is testing. Quietly omitting it demonstrates the opposite.

Human error is not an error analysis. Name the actual source, and say whether it would bias the result high or low.

Error analysis, done properly

Uncertainty should be quantified where possible and attributed where not. Distinguish systematic from random error, identify which dominates, and say what it would take to reduce it. Reporting a value to six significant figures when the instrument resolves three is its own small error, and markers notice.