analysis and interpretation of correlations

Question 1

What is a correlation in psychology?

Answer 1

A correlation is a statistical technique used to assess the relationship between two co-variables. It tells us whether an increase or decrease in one variable is associated with an increase or decrease in another.

Question 2

What are the two main types of correlation?

Answer 2

• Positive correlation: As one variable increases, the other also increases.
• Negative correlation: As one variable increases, the other decreases.
(No correlation means no consistent relationship.)

Question 3

How are correlations visually represented?

Answer 3

Using a scattergram. Each dot represents one participant’s data on two variables. The pattern of the dots indicates the type and strength of the correlation

Question 4

What is a correlation coefficient?

Answer 4

A numerical value between -1 and +1 indicating the strength and direction of a correlation.
• +1 = perfect positive
• 0 = no correlation
• -1 = perfect negative

Question 5

How do you interpret a correlation coefficient?

Answer 5

• The closer to ±1, the stronger the correlation.
• The closer to 0, the weaker the correlation.
E.g., r = +0.85 is a strong positive correlation; r = -0.30 is a weak negative correlation.

Question 6

Can correlations establish causality?

Answer 6

No. Correlation does not imply causation — it only indicates a relationship, not a cause-effect link.
Examiners often penalise students for confusing correlation with causality. ‘

Question 7

What are possible reasons for a correlation besides causality?

Answer 7

• Third variable problem (extraneous variable)
• Coincidence
• Bidirectional relationships (A affects B and B affects A)

Question 8

What are strengths of using correlations?

Answer 8

• Useful for initial investigation before conducting experiments.
• Quick and economical — especially using existing data.
• Can highlight potential relationships worth exploring further.

Question 9

What are limitations of correlations?

Answer 9

• Cannot show cause and effect.
• May be affected by extraneous variables.
• Misinterpretation by assuming causality

Question 10

What statistical test is used to assess correlations?

Answer 10

• Spearman’s rho (for ordinal data or non-parametric).
• Pearson’s r (for interval/ratio data with parametric assumptions met).

Question 11

What type of data is needed for a correlation?

Answer 11

Two sets of quantitative data (ordinal or interval/ratio) that are measured, not manipulated.

Question 12

What’s the difference between a correlation and an experiment?

Answer 12

• Correlation: Measures relationship between variables — no manipulation.
• Experiment: Manipulates an independent variable to observe effect on a dependent variable — can infer causality.

Question 13

What should you look for when interpreting a scattergram in an exam?

Answer 13

• Overall trend (positive/negative/no correlation)
• Outliers
• Tightness of data points (strong/weak relationship)
• Use correct terminology when describing (e.g., “moderate positive correlation”)

Question 14

In a correlation, what does it mean if the points on a scattergram form a loose oval shape?

Answer 14

This suggests a weak correlation — the variables are somewhat related, but not strongly.

Question 15

How should you answer a 4- or 6-mark question on interpreting a correlation?

Answer 15

• Clearly state type of correlation (positive/negative/none)
• Quote the coefficient or refer to the scattergram
• Interpret what this means for the relationship
• Avoid claiming causality
• Link interpretation to the research aim or context