A correlation means an association or a link between two things. In psychology we say that a correlation is the measure of an extent to which two variables are related to each other. Therefore, we can say that if one variable increases and so does the other, then there is a positive correlation. If one variable increases and the other decreases, then there is a negative correlation. Have a look at the examples below:
- As an athlete spends more hours in the gym, their ability to lift heavier weights increases – positive correlation as hours in gym and weight lifted both increase
- As an athlete spends more time on the race track, their 100m speed time decreases – negative correlation as hours on race track increase but speed time decreases
- As someone’s calorie intake increases, their weight increases – a positive correlation as both variables are increasing.
When there is no correlation between two variables then this is known as a zero correlation. For example, there is no link between a person’s intelligence and the amount of biscuits that they eat!
Correlations can be used to make predictions about things, such as when the temperature increases, sales of bikinis will also increase.
A correlation is usually expressed visually in the form of a scatter gram, as shown in the diagram below:
As you can see, the scatter grams which show a correlation of some sort have the points in some form of line whereas the one where there is no correlation has points which are completely random.
- Researchers are able to explore variables which are naturally occurring and therefore do not need to be subject to laboratory experiments
- It is easy for researchers to clearly see that there is an association between two variables.
Weaknesses of correlations:
- Although they can show a relationship, correlations do not show cause and effect. For example, although they can show that people who have schizophrenia have higher levels of the neurotransmitter dopamine, we cannot be sure that this is what has caused their condition
- Correlations are quite limited as although they can show that two variables are linked, they cannot explain why this is the case.
