ANOVA and ANCOVA tests using R Programming
Problem Description:
30 stores sales data has been recorded. The first 10, second10, third 10 stores are undergoing low, medium, high promotion respectively.
We would like to find out if Stores are increased their sales based on the client ratings, promotion and coupons using the ANOVA and ANCOVA test
Data Set
ANOVA - Analysis of Variance
One way ANOVA
It is used to compare the two samples.
Dependent Variable- Continuous
Independent Variable- Categorical
One dependent and one independent Variable
Sale is a Dependent variable. 30 stores have taken in the example.
Promotion is an independent variable that has three categories(High, Medium, Low).
We are undergoing different levels of promotion. Do sales affected by these promotions?
H0(Null Hypothesis)- sales are not affected by the promotion.
Ha(Alternative Hypothesis)- Sales are affected by the promotion.
Conclusion:
Decision according to the p-value:
Calculated P value(1.47e-06) is less than 0.05.
We reject the null hypothesis. Data fails to provide enough evidence to support the null hypothesis.
Yes, Sales are affected by the different levels of promotions.
Two way ANOVA
It is used to compare two or more sample means.
One Dependent Variable- Continuous.
Two Independent Variables- Categorical.
If the customer purchases more than 5000 rupees, Coupon 1 will be provided or Coupon 2 will be issued if more than 10000 rupees.
We need to find out the changes in the sales depends upon the promotion and coupon.
The decision according to P-value
The calculated P-value is less than 0.05.
We reject the null hypothesis. Data fails to provide enough evidence to support the null hypothesis.
Yes, Sales depends on different levels of promotions and coupon.
ANCOVA- Analysis Of Co Variance
It is used to compare the covariance of the sample mean.
One Dependent Variable: Continuous
Three Independent Variable: Continuous and Categorical
The decision according to P-value
Promotion and coupon are less than 0.05 whereas client ratings are greater than 0.05.
In conclusion, Client rating does not affect sales whereas promotion and coupons affect sales.
Happy learning…