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.
Este exercício faz parte do curso
A/B Testing in R
Instruções do exercício
- Create a scatter plot with
AmountSpent
on the x-axis andTimeSearching
on the y-axis to assess the linearity of the relationship. - Use
shapiro.test()
to determine whetherAmountSpent
is normally distributed. - Use
shapiro.test()
to determine whetherTimeSearching
is normally distributed.
Exercício interativo prático
Experimente este exercício completando este código de exemplo.
# Assess the assumption of linearity
ggplot(___) +
___
# Assess the normality of the `AmountSpent` variable
___
# Assess the normality of the `TimeSearching` variable
___