CORRELATIONS

• A mathematical technique in which a researcher investigates an ASSOCIATION between two variables (co-variables) e.g. the amount of caffeine drank and hours of sleep 
• Correlation vs experimental method


  Co-variables are not referred to as IV and DV because a correlation investigates the association between the variables rather trying to show cause and effect. There is no manipulation of the co variable by the researcher in correlation, like in experiments to study the effects of something. 
• How are correlations represented?

  Relationships between variables are plotted on a scattergram. One co-variable is represented on the x-axis and the other on the y-axis. Each point or dot on the graph is the x and y positions of each co-variable. 
• Correlations tell us about the direction and strength of an association between two or more co-variables.  

  The direction can be positive, negative or zero. 
• Positive correlation – as one variable increases so does the other e.g. no. of people in a room is positively correlated to the noise in a room 
• Negative correlation – as one variable increases the other decreases e.g. no. of people in a room is negatively correlated to the amount of space. 
• Zero correlation – no relationship between the co-variables 
• Correlation coefficients are calculated by doing a statistical test, so they are inferential statistics (not a descriptive statistic). This produces a correlation coefficient between –1 and +1, which can tell us about the strength relationship between two variables. 

  +0.4 --> weak positive 

  +0.8 --> strong positive

  +1 --> perfect 

  -0.2 --> weak negative

  -0.9 --> strong negative

  -1 --> perfect 
• ‘r’ stands for correlation coefficient 
• Strengths of correlations

  1. Provides a quantifiable measure of how two variables are related 
  2. May suggest ideas for future research if the variables are strongly correlated and demonstrate an interesting pattern, before a researcher commits to an experimental study. 
  3. Quick and economical to carry out – no need to control variables 
  4. Secondary data (e.g. gov statistics) can be used – less time consuming 
• Weaknesses of correlations

  1. Only tell us how variables are related and not why they are related (a link – not cause and effect) 
  2. The third variable problem - another untested variable is causing the relationship between the two co-variables 
  3. Correlations can then be misused or misinterpreted 
• Correlation hypothesis


  Hypothesis for correlational studies states the expected relationship between variables, but co variables in this case, which are operationalised. Like experiments, the hypothesis can be directional or non-directional. 
• Directional hypothesis example – use it when there is already pre-existing data indicating a particular outcome e.g. there is a positive correlation between the price of a chocolate bar and its tastiness rating 
• Non-directional hypothesis example – use when there is no previous research, or the research is conflicting/contradictory e.g. there is a correlation between the price of a chocolate bar and its tastiness rating