Psychology: Correlations

Psychology: Correlations

1. What

   • A statistical technique that involves measuring two variables whilst looking for a relationship between the two. It is not to say that one thing causes the other, just that one thing varies in accordance with another
   • There are no independent and dependant variables, but co-variables
   • Allows us to statistically analyse naturally occurring events so it can be used when it is impossible and unethical to do experimental research
   • Refers to a measure of how strongly the variables are related to each other so it is really a method of analysing data already collected

2. Types of correlation

   • Positive: relationship between variables where as one increases, the other increases
   • Negative: relationship where as one variable increases the other decreases
   • No correlation: the variables are said to be uncorrelated

3. How is it presented and measured?

   • They are displayed on scatter graphs where each variable is plotted against each other in dots to represent each pair of scores.
   • When drawing a scattergraph the axis should be fully operationalised and labelled and indicate what co-variable they represent and how it was measured
   • Association does not mean a cause and effect relationship
   • 4 things for drawing a graph: title including both variables, label both axis, plot data correctly

4. Evaluation of conducting a correlation

   • Strenghts
     • We can establish relationships between variables without having to manipulate behaviour so correlations can be used in situations where experiments can not
     • Correlations are often used to suggest ideas for future experimental research
   • Weaknesses
     • A cause and effect cannot be established, they can only establish a relationship between two variables so further research would be needed
     • Other unknown variables may have been used and caused the link between the co-variables being measured so correlations may lack validity

5. Correlational hypothesis

   • Must include both variables that the correlation is testing and be fully operationalised
   • If it is one tailed it must say whether it is positive or negative
   • If it is two tailed it must say that there is a significant relationship/correlation but no direction stated
   • If it is a null hypothesis it must say that there is no significant relationship and any relationship is due to chance
   • Must never say the words difference, effect or link

6. Reliability and validity

   • Reliability
     • All variables must be measured in a way that produces quantitative data
     • probably be something to comment on in the way that there should be a way to retest to check reliability; i.e it can be replicated
     • But if a self report had been used it may not have been answered in a consistent way by all
   • Internal validity
     • There may be risk of social desirability bias or demand characteristics which may have implications on the result
     • If a rating scale has been used to collect thought or opinion that may be obscure in meaning then people may not have understood, if the scale was not labelled or if leading questions are used
   • External validity
     • Consider of the behaviour has been measured in an artificial or natural environment
     • participants may have been asked to take part in a task that was not natural