Correlations strengths and weaknesses

Strengths

-correlations can show a relationship between two variables which can lead to hypothesises and be used to conduct further research on the cause

-reliability can be tested through replicating research

-correlation allows researchers to research naturally occurring variables that may be unethical or impractical to test experimentally eg. It would be unethical to conduct an experiment on whether smoking causes lung cancer

Weaknesses

we cannot establish cause and effect unlike an experiment (where IV is deliberately changed in order to observe the effect on DV) the variables in correlations can only be measured. Only experiments can imply causation due to strict control over the IV and DV. Therefore people often misinterpret correlations and assume that causing effect has been found, which is not possible. For example as sales of ice cream increase the number of crimes recorded increase but you cannot conclude the increase sales of ice cream caused an increase in crimes.

-There may be intervening variables that can explain why the co-variables are linked other variables involved that could provide another explanation of why co-variables being studied are linked. (Extraneous variables but we call the intervening in terms of correlation) eg. Our sales of ice cream increase the number of crimes recorded increase the other variable linking these maybe more people outside so it needs to more trouble due to the hot weather.

-Variables would have to be operationalised (how you define and measure a specific variable) which could make the results less applicable to real life. E.g. studying the relationship between a pet and happiness you would have to explain how you would scientifically measure happiness which may not be realistic or applicable to real life.

-Correlation is show linear relationships but do not reflect curvilinear ones. Eg, the effect of stress on exam performance (might go up then down)

e

e

d

d