Correlations

Psychologists are not alone in their use of correlations, in fact many disciplines will use the method. A correlation checks to see if two sets of numbers are related; in other words, are the two sets of numbers corresponding in some way.

In the case of psychology, the numbers being analysed relate to behaviours (or variables that could affect behaviour) but actually any two variables producing quantitative data could be checked to establish whether a correlations exists.

Each of the two sets of numbers represents a co-variable. Once data has been collected for each of the co-variables, it can be plotted in a scattergram and/ or statistically analysed to produce a correlation coefficient.

Scattergrams and coefficients indicate the strength of a relationship between two variables, which highlights the extent to which two variables correspond.

The relationship between two variables will always produce a coefficient of between 1 and -1.

Coefficients with a minus in front of them highlight a negative correlation which means that as one set of numbers is increasing the other set is decreasing or as one decreases the other increases, so the trend in the data from one variable opposes the other.

In contrast, coefficients which are positive indicate that both sets of data are showing the same trend, so as one set of data increases so does the other or as one set decreases the same trends is observed in the second set of data

Experiments Vs Correlations

The most fundamental difference between experiments and correlations is that experiments assess the effect of one variable, (I.V.) on another variable which is measured (D.V.).

This necessitates that data is discrete or separate and the effect of this on something else is being measured.

In contrast, correlations do not use discrete separate conditions, instead, they assess how much of a relationship exists between two co-occurring variables which are related.

For example, if a psychologist was interested in investigating stress and illness, they could generate stress scores and illness scores for 20 participants and assess how these two sets of numbers relate to each other, thereby adopting a correlational method. This could be turned into an experiment though if the researcher allocated 10 participants with low scores for stress (eg. 10/50 or less) and 10 participants with high stress scores (eg. 40/50 or more). There are now two conditions, one for low stress and one for high stress. If the researcher were to take illness scores for all 20 participants and compare the low stress against the high stress participants, this would be assessing the effect of stress on illness experimentally.

Strengths of Correlations

Correlations are very useful as a preliminary research technique, allowing researchers to identify a link that can be further investigated through more controlled research.

Can be used to research topics that are sensitive/ otherwise would be unethical, as no deliberate manipulation of variables is required.

Limitations of Correlations

Correlations only identify a link; they do not identify which variable causes which. There might be a third variable present which is influencing one of the co-variables, which is not considered.

Eg. stress might lead to smoking/ alcohol intake which leads to illness, so there is an indirect relationship between stress and illness.