Correlation

Biology: Cambridge International A Level, topic 18.1.4

Correlation and Causation

Correlation means that there is a relationship between two variables. Causation means that changing one causes the other to change.

Causation

Causation

  • Correlation does not imply causation.
  • If two variables are correlated that does not mean that changes in one directly cause changes in the other.
Example

Example

  • The number of ice cream sales might be positively correlated with temperature but this does not mean that ice cream sales cause higher temperatures.

Pearson's Linear Correlation

Two variables are correlated if you can draw a straight line that lies close to most of the points.

Variables and constants

Variables and constants

  • The variable x is the independent or explanatory variable.
  • The variable y is the dependent or response variable.
  • The slope of the regression line is given by the constant b.
  • The sign and size of this number tell us how y changes for every unit increase in x on average.
Pearson's correlation coefficient

Pearson's correlation coefficient

  • The Pearson's correlation coefficient, rr, provides a measure of strength and direction of the correlation between x and y.
  • The value of rr is always between –1 and +1.
Pearson's correlation coefficient

Pearson's correlation coefficient

  • Values of rr close to –1 or to +1 indicate a stronger linear relationship between x and y.
  • If r=0r = 0, there is likely no linear correlation.
  • If r=1r = 1, there is a perfect positive correlation so all points lie on a straight line.
  • If r=-1r = \text{-}1, there is a perfect negative correlation so all points lie on a straight line.
Pearson's correlation coefficient

Pearson's correlation coefficient

Formula

Formula

  • The formula for the correlation coefficient is:
  • r=1n−1∑(xi−x‾sx)(yi−y‾sy)r = \frac{\large 1}{\large n-1}\sum\left(\frac{\large x_i-\overline{x}}{\large s_x}\right)\left(\frac{\large y_i-\overline{y}}{\large s_y}\right)
  • Where x‾,y‾\overline{x},\overline{y} are the mean and sx,sys_x,s_y are the standard deviations of the explanatory and response variables respectively.
  • It is much more common to calculate rr using a calculator.

Test yourself on Correlation

  • Do these two variables have a correlation or a causal relationship?
    Time spent practising tennis & accuracy of tennis serve.

  • Do these two variables have a correlation or a causal relationship?
    Time spent practising tennis. Average daily temperatures.

  • Do these two variables have a correlation or a causal relationship?
    Lemonade sold & Average daily temperatures.

  • Which feature of a regression line tells us how the independent variable changes for every unit increase in the dependent variable on average?

  • Which variable is the response variable in the graph below?

Practise this topic for free