Use Pearson and Spearman correlation
Choose and interpret the two correlation tests for relationships in ecological data.
A Level; Cambridge International Biology 9700, 2028–2030. Written responses are self-checked; this does not certify supervised fieldwork or practical competence.
Before you begin
- Classification compares shared characteristics; populations occupy environments and contain variation.
Work on paper. Try each question before revealing help, and explain why your method works.
Quick readiness check
Which test is designed for ranked data that are not normally distributed?
What you will learn
- Use Pearson correlation for suitable normally distributed continuous data
- Use Spearman rank correlation for suitable ordinal or non-normal data
Build the explanation
Pearson’s linear correlation tests association between two sets of continuous, normally distributed data when the scatter suggests a linear relationship; the syllabus specifies at least five paired observations. Spearman’s rank correlation tests association using ranks when data are ordinal or not normally distributed, with independent randomly selected pairs and a monotonic trend. The formulae are supplied.
Use evidence within scope
A coefficient ranges from −1 to +1 and describes strength and direction of association. Correlation does not prove that one environmental factor caused the other variable to change; other variables may contribute.
Pause and explain
Plant abundance and soil moisture are correlated. What additional evidence is needed before claiming moisture caused the abundance difference?
Worked example
A researcher has non-normal rankable data on shade and plant abundance with 12 paired sites. Which test is appropriate?
Show the worked solution
- The data can be ranked and are not normally distributed.
- Use Spearman’s rank correlation.
Answer Spearman’s rank correlation is appropriate for these rankable, non-normal data, subject to independent and randomly selected observations.
Try a topic test
Choose Easy, Normal or Difficult. Save your answers and open worked review after finishing.
Test this topic →From guided practice to a new situation
Write your answer and reasoning first. Use a hint only if you are stuck. The model solution is for self-checking; your written work is not automatically marked.
Paired continuous data are approximately normal and show a linear trend. Which correlation test is suitable?
Give me a hint
Identify the required evidence and sequence the reasoning.
Compare my reasoning
- Check distribution and relationship shape.
- Pearson’s linear correlation fits these conditions.
Pearson’s linear correlation is suitable, assuming the observations and sample design meet the required conditions.
Look for these in your work
- Relevant evidence selected
- Reasoning linked to conclusion
A scatter plot curves upward but has no straight-line pattern. Why might Pearson be unsuitable?
Give me a hint
Use the supplied example or data to justify a conclusion.
Compare my reasoning
- Pearson tests linear correlation.
- The relationship is not linear.
Pearson’s linear correlation may not describe a curved relationship well; inspect the pattern and choose a method consistent with the data and question.
Look for these in your work
- Concept applied
- Limit or assumption stated
Plant abundance and soil moisture are correlated. What additional evidence is needed before claiming moisture caused the abundance difference?
Give me a hint
Apply the concept in a new context and qualify its limits.
Compare my reasoning
- Correlation alone does not establish cause.
- Consider other biotic and abiotic factors or a controlled test.
A correlation does not establish causation. Other factors could influence both variables; controlled evidence or a stronger causal design is needed.
Look for these in your work
- New context handled
- Conclusion appropriately bounded
Common mistakes
- Using Pearson for any two numerical columns without checking assumptions
- Claiming a correlation proves causation
What can you explain now?
Which test is designed for ranked data that are not normally distributed?
Compare with the explanation
Spearman’s rank correlation
Pearson’s test is for suitable continuous, normally distributed data with a linear relationship.
After trying it yourself, choose your next review. This is your self-assessment.
Your review choice appears on Today. Sign in to sync it across devices.
Make it stick
When should Pearson or Spearman correlation be considered, and what can neither prove?
If you needed the solution, close it and solve the problem again from a blank page. A correct answer is stronger when you can explain the reason for each step.