CCEA Biology practical skills: AS 3 and A2 3 without guesswork
AS 3 and A2 3 are worth 20% of each qualification, and they are the units students revise last. They are also the most learnable, because the same handful of question shapes appear every year.
Variables: name it and control it
A control variable is only worth a mark when you say how it was kept constant. 'Temperature was controlled' scores nothing; 'the tubes were kept in a water bath at 30 °C' scores.
Identify the independent variable (the one you change), the dependent variable (the one you measure, with its unit), and at least two control variables with the method used to fix each.
Accuracy, precision and reliability are three different things
| Term | Meaning | How to improve it |
|---|---|---|
| Accuracy | How close a measurement is to the true value | Calibrate apparatus; use a more suitable instrument |
| Precision | How finely and consistently you can measure | Use apparatus with a smaller scale division |
| Reliability | Whether repeats give a similar result | Repeat and calculate a mean; discard anomalies with justification |
| Validity | Whether the experiment tests what it claims to | Control the other variables; include a suitable control |
Processing data
- Tables: quantity and unit in the column heading, never repeated beside each value.
- Graphs: independent variable on the x-axis, labelled axes with units, sensible scale, line of best fit where a trend is expected.
- Percentage change = (final − initial) ÷ initial × 100. State whether it is an increase or a decrease.
- Rate = change ÷ time, with the unit written out, for example cm³ min⁻¹.
- Round to the number of significant figures justified by your least precise measurement.
Choosing a statistical test at A2
Then interpret it properly. Compare the calculated value with the critical value at p = 0.05 and state whether the result is significant and what that means for the hypothesis. Standard deviation questions usually want you to say that overlapping error bars suggest the difference may not be significant.
- Looking for a difference between two means of continuous data: t-test.
- Looking for a relationship between two continuous variables: a correlation coefficient.
- Looking for an association between categories, or comparing observed with expected counts: chi-squared.
Evaluations that actually score
Generic evaluations — 'human error', 'do more repeats' — rarely earn credit. Name one specific limitation of the apparatus or method used, say what error it introduces, and give the change that removes it.
For example: reading the meniscus of the potometer by eye introduces a parallax error in the volume measurement; using a graduated capillary tube with a finer scale and reading at eye level would reduce it.
One precise, specific improvement beats four vague ones. Mark schemes credit the error you have identified, not the number of suggestions you make.