All subjects · Biology · WBI16 Practical Skills in Biology II

Hypotheses, null hypotheses and significance

Edexcel IAL Biology · WBI16 Practical Skills in Biology II

A statistical test checks whether a difference or correlation is big enough that it is unlikely to be due to chance. It starts from a null hypothesis.

Practise this in the app: instant marking and a fix for every mistake. Free, no email needed to start.

Practise this lesson

Hypothesis and null hypothesis

A hypothesis is a testable prediction: 'Increasing light intensity increases the rate of photosynthesis.'
A null hypothesis (H₀) states there is no difference, no correlation or no association, so any pattern is due to chance.

Write it with the actual variables:

difference (t-test): 'There is no significant difference between the mean mass of leaves eaten of variety A and variety B' (the October 2023 mark scheme wording);
correlation (Spearman's rank): 'There is no significant correlation between light intensity and percentage cover of the plant';
frequencies (chi-squared): 'There is no significant difference between the observed and expected numbers of woodlice in damp and dry areas'.

Significance and the 5% level

Biologists usually use the p = 0.05 (5%) significance level: a result counts as significant if there is less than a 5% probability it happened by chance.

1. Calculate the test statistic (t, rₛ or χ²).

2. Find the critical value for p = 0.05 and the right degrees of freedom or n.

3. If the calculated value is greater than or equal to the critical value, reject the null hypothesis: the difference or correlation is significant.

4. If it is less than the critical value, accept (do not reject) the null hypothesis: any difference may be due to chance.

Writing the conclusion

A full conclusion has three parts:

compare: 'The calculated t of 4.87 is greater than the critical value of 2.23 at p = 0.05';
decide: 'so the null hypothesis is rejected';
interpret in context: 'there is a significant difference in mean zone of inhibition between antibiotics A and B'.

Significant does not mean 'proved' and does not show what causes the difference. A non-significant result does not prove there is no effect: the sample may have been too small.

Worked example

Write a null hypothesis for an investigation into whether the mean heart rate of Daphnia differs between water with and without caffeine.

Use 'no significant difference' and name both groups and the measured variable.

Answer: There is no significant difference between the mean heart rate of Daphnia in water with caffeine and in water without caffeine.

Key idea: H₀: no difference, correlation or association, named in context. At p = 0.05: calculated value ≥ critical value, reject H₀ (significant); less than critical value, accept H₀ (may be chance). Conclude in context.

Check yourself

Which is a correctly written null hypothesis?
- There is no significant difference between the mean number of colonies with and without the antibiotic
- The antibiotic will kill the bacteria
- The antibiotic has a big effect
- There is a difference between the two groups
Show answer
There is no significant difference between the mean number of colonies with and without the antibiotic. It states no difference and names the variable and groups.
Which null hypothesis suits a Spearman's rank test of soil moisture and number of mosses?
- Moss numbers increase with soil moisture
- There is no significant difference between the mean soil moisture and moss number
- There is no significant correlation between soil moisture and number of mosses
- Moisture causes moss growth
Show answer
There is no significant correlation between soil moisture and number of mosses. Correlation tests use 'no significant correlation'.
What does a significance level of p = 0.05 mean?
- A result is accepted as significant if there is less than a 5% probability it is due to chance
- The results are 5% accurate
- 5% of results are anomalies
- The experiment must be repeated 5 times
Show answer
A result is accepted as significant if there is less than a 5% probability it is due to chance. 5% is the conventional cut-off for chance.
The calculated value of a test statistic is greater than the critical value at p = 0.05. What should be concluded?
- Accept the null hypothesis
- The test was done wrongly
- The result is due to chance
- Reject the null hypothesis; the result is significant
Show answer
Reject the null hypothesis; the result is significant. Bigger than critical means unlikely to be chance.
A calculated t is 1.6 and the critical value is 2.31. What is the correct conclusion?
- Reject the null hypothesis
- Accept the null hypothesis; there is no significant difference, so the difference may be due to chance
- The hypothesis is proved
- There is a significant correlation
Show answer
Accept the null hypothesis; there is no significant difference, so the difference may be due to chance. 1.6 is less than 2.31.
Why should a significant result not be described as 'proving' the hypothesis?
- Statistics are never useful
- Only non-significant results can prove anything
- Biology does not use hypotheses
- There is still a small probability it is due to chance, and other factors may be involved
Show answer
There is still a small probability it is due to chance, and other factors may be involved. Significance is about probability, not proof.

Practise with instant feedback

Sources

Last updated 8 November 2026. Spotted a mistake? Tell us.