What correlations are and their purpose in psychology

Correlations are a statistical method used in psychological research to examine the relationship between two variables. These variables are known as co-variables, which means they vary together in some way. Researchers use correlations to identify patterns and associations without manipulating any factors, making it a useful tool for exploring potential links in behaviour, personality, or other psychological phenomena.

This approach is particularly valuable when it is unethical or impractical to conduct experiments, such as studying the relationship between childhood experiences and adult mental health. By measuring how changes in one co-variable relate to changes in another, correlations help researchers decide if further investigation is warranted. For example, if a link is found between hours of sleep and exam performance, it might prompt more detailed studies.

Key terms in correlations

• Co-variables - The two factors being measured to see if they change together, such as self-rated extroversion and scores on a personality test.
• Scatter plot - A graph used to visually represent the relationship between co-variables, with one on the x-axis and the other on the y-axis; data points show individual scores.

Types of correlations

Correlations can show different patterns depending on how the co-variables relate.

Positive correlation

A positive correlation occurs when both co-variables increase or decrease together. This means as the value of one rises, so does the other. For instance, more time spent revising might link to higher exam scores, indicating a positive relationship.

Negative correlation

A negative correlation happens when one co-variable increases while the other decreases. This shows an inverse relationship. As an example, increased stress levels might correspond to lower concentration abilities, where higher stress leads to reduced focus.

Zero correlation

If there is no consistent pattern between the co-variables, this indicates zero or no correlation. The values do not change together in any predictable way, such as no link between shoe size and intelligence.

To visualise these, scatter plots are often used. In a positive correlation, points trend upwards from left to right; in a negative one, they trend downwards; and with no correlation, points are scattered randomly without a clear line.

Correlation coefficients and their interpretation

A correlation coefficient is a numerical value that quantifies the strength and direction of the relationship between two co-variables. It provides an objective measure, making it easier to compare relationships across studies.

Range and meaning of correlation coefficients

• Coefficients range from -1 to +1.
• A value of +1 indicates a perfect positive correlation, where high values in one co-variable perfectly match high values in the other (e.g., height and weight often show this tendency).
• A value of -1 shows a perfect negative correlation, where high values in one correspond to low values in the other (e.g., driving speed and fuel efficiency).
• A value of 0 means no relationship exists between the co-variables.

In psychology, perfect coefficients like +1 or -1 are rare due to the complexity of human behaviour. Instead, values closer to +1 or -1 suggest stronger relationships, while those near 0 indicate weaker or no links. For example, a coefficient of +0.8 would show a strong positive correlation, whereas +0.2 suggests a weak one.

Example data and scatter plot interpretation

Consider a study where participants rate their extroversion on a scale of 1 (very introverted) to 10 (very extroverted), and then take a personality test scoring from 0 (introverted) to 100 (extroverted). The table below shows sample scores:

Plotting these on a scatter plot would show points trending upwards, indicating a positive correlation. The coefficient might be around +0.88, suggesting a strong link between self-perception and test results.

Issues with using correlations in research

While correlations are useful, they have limitations that researchers must consider to avoid drawing inaccurate conclusions.

Key limitations

• Lack of cause and effect - Correlations show associations but cannot prove that one co-variable causes changes in the other. For example, a link between ice cream sales and crime rates does not mean one causes the other; a third factor like hot weather might influence both.
• Confounding variables - An unknown third variable could be responsible for the observed relationship. This is called a confounding variable, which affects both co-variables. In studies of monozygotic (MZ) twins and IQ, a correlation of +0.72 might suggest genetics play a role, but shared environments (e.g., similar upbringing) could be the true cause instead.
• Directionality issues - It may be unclear which co-variable influences the other, or if the relationship is bidirectional.

These issues mean correlations should be interpreted cautiously, often as a starting point for more controlled research like experiments.

Advantages of using correlations in research

Despite their limitations, correlations offer several benefits that make them a popular method in psychological studies.

Key advantages

• Efficiency and speed - They allow quick analysis of relationships using existing data, such as school records of GCSE and A-level scores, without needing to collect new information.
• Ethical suitability - Correlations can explore sensitive topics where experiments would be unethical, like the link between trauma and mental health outcomes.
• Objective measurement - The correlation coefficient provides a clear, numerical summary of the relationship's strength and direction, making results easy to understand and compare.
• Foundation for further research - A strong correlation can justify more in-depth investigations, helping researchers prioritise areas worth exploring with other methods.

By highlighting potential patterns, correlations guide psychologists towards deeper insights while being straightforward to implement.