Correlation coefficients

• What are correlation coefficients?

  Correlation indicates the strength of a linear relationship between 2 sets of data.
• What are the 2 correlation coefficients?

  • PMCC
  • SRCC
• How are the correlation coefficients calculated?


  Since data is for 2 variables, we use menu, 6, 2

  • For PMCC use raw data
  • For SRCC use ranked data
• Scatter diagrams 


  When using bivariate data we are best suited to using a scatter diagram because:

  • we can see the shape of the distribution
  • we check for outliers
  • we estimate the correlation value
• Correlation vs causation


  2 variables have a casual relationship if a change in one variable causes a change in the other. Just because there is a correlation does not mean that a casual relationship exists.
• Exam technique for correlation coefficients


  • PMCC requires bivariate normal
  • Value for PMCC and SRCC is unaffected by linear scaling
• Why might SRCC be preferred to PMCC


  • if our data is qualitative
  • if relationship is not linear
  • if population distributions are not normal
  • if data is already ranked