Assessing Pearson assumptions
For a Pearson correlation to be appropriate, the data must follow two assumptions: linearity and normally distributed.
A company is interested in running a Pearson correlation on the amount of time spent on a website and the amount of money spent purchasing items on the website, first assessing the correlation ignoring groups. Assess whether a Pearson correlation is appropriate given the linear and normal distribution assumptions that must be met.
The ggplot2 package and SiteSales dataset have been loaded for you.
This exercise is part of the course
A/B Testing in R
Exercise instructions
- Create a scatter plot with
AmountSpenton the x-axis andTimeSearchingon the y-axis to assess the linearity of the relationship. - Use
shapiro.test()to determine whetherAmountSpentis normally distributed. - Use
shapiro.test()to determine whetherTimeSearchingis normally distributed.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Assess the assumption of linearity
ggplot(___) +
___
# Assess the normality of the `AmountSpent` variable
___
# Assess the normality of the `TimeSearching` variable
___