Correlational research

The correlational method

A correlation measures the relationship between two co-variables, both measured as they naturally occur. Nothing is manipulated, so there is no independent or dependent variable.

Example: hours of screen time and hours of sleep. Each participant gives a score on both co-variables, and each pair becomes one point on a scatter diagram.

Types of correlation

• Positive: as one co-variable increases, the other increases (points slope up to the right).
• Negative: as one increases, the other decreases (points slope down).
• No correlation: no pattern.

The correlation coefficient runs from -1 (perfect negative) through 0 (none) to +1 (perfect positive). The closer to 1 or -1, the stronger the relationship.

Issues: cause and effect

Correlation does not show causation. Even a strong correlation cannot tell us:

• the direction of cause (does screen time reduce sleep, or does poor sleep lead to more screen time?), or
• whether a third variable (such as stress or age) causes both.

Only an experiment, where a variable is manipulated and others controlled, can show cause and effect.

Strengths and weaknesses

• Strengths: can study variables that cannot be manipulated ethically (such as brain damage or hormones); shows the strength of relationships; can suggest hypotheses for experiments.
• Weaknesses: no cause and effect; other variables are not controlled; a relationship that is not straight-line (curvilinear) may be missed.

Worked example

A study finds a correlation of -0.62 between age and self-rated aggressive tendencies. Interpret it.

1. The sign is negative: as age increases, self-rated aggression decreases.
2. 0.62 is a moderate to strong relationship.
3. It does not show that getting older causes lower aggression; other variables may be involved.

Answer: A moderate to strong negative correlation, with no conclusion about cause.

Check yourself

1. In correlational research the two variables measured are called
   • independent and dependent variables
   • extraneous variables
   • co-variables
   • behavioural categories
   Show answer

   co-variables. Neither is manipulated, so they are co-variables.

2. As time on social media increases, hours of sleep decrease. This is a
   • negative correlation
   • positive correlation
   • zero correlation
   • causal relationship
   Show answer

   negative correlation. One goes up as the other goes down.

3. Which coefficient shows the strongest relationship?
   • +0.45
   • +0.70
   • -0.88
   • -0.10
   Show answer

   -0.88. Strength depends on distance from 0, ignoring the sign.

4. Why can a correlation not show cause and effect?
   • Correlations always use small samples
   • Neither variable is manipulated, so direction and third variables are unknown
   • Correlations only use qualitative data
   • Scatter diagrams are inaccurate
   Show answer

   Neither variable is manipulated, so direction and third variables are unknown. Without manipulation and control, cause cannot be established.

5. On a scatter diagram, each point represents
   • the mean of a group
   • a whole condition
   • a tally of behaviours
   • one participant's scores on the two co-variables
   Show answer

   one participant's scores on the two co-variables. Each point is one pair of scores.

6. Researchers find a positive correlation between testosterone and aggression. Explain why they cannot conclude that testosterone causes aggression.
   Show answer

   It is only a correlation, so the direction is unknown (aggression could raise testosterone) and a third variable such as stress could affect both, so cause and effect cannot be shown.. Direction of cause or third variable.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification, Issue 5 (February 2021)
• Pearson IAL Psychology component guide: WPS02 (Issue 2, November 2024)

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