Correlations

What is a correlation and how is it shown graphically?

• a statistic that identifies the strength and direction of a linear relationship between 2 co-variables

• Plotted on a scattergram

How does a correlation study differ from an experiment?

• An experiment involves manipulating the IV

• A correlation study involves co-variables instead if IV & DV

What would a positive, negative, and zero correlation look like on a scattergram

State and explain the main weakness of correlational research

• Can’t establish cause and effect because it doesn’t manipulate an IV

  • Variable A may cause B to change or vice versa

  • Variable C may cause both A and B to change

Explain why correlational studies are often used rather than experiments for ethical and practical reasons

• Ethical - can do a correlation when it’s not acceptable to manipulate variables since it wouldn’t protect participants from harm

• Practical - can do a correlation when it’s impossible to manipulate the IV

Explain how a zero correlation could involve a set of results which display a clear pattern on a scattergram

• Could show an inverted U shape

• E.g. performance and arousal

Explain how despite having the same correlation coefficient in 2 studies, it may not be possible to draw the same conclusions about the association between the covariables

• different patterns can lead to the same correlation coefficient

• But there patterns would lead to different conclusions