Model Answers to 08 ANOVA in Six Sigma Statistics using Minitab 17, Green Belt Edition.

Exercise 8.5.4 ANOVA GLM

A multi-media service provider wants to analyse some of their sales data. They have data on property size across 3 regions in a city. They want to know if sales are affected by  property size and city region. Are there differences within the property size bands and sales? Also are there differences within the city regions and sales?

Analyse the data in File 08 ANOVA.xlsx worksheet Ex 8.5.4 and answer the questions shown above and additionally those shown below. The data was collected randomly and is recorded in time order.

In this procedure are Fixed Factor s are Prop_size and Region. The response is Sales.

  1. Form the Main Effects Plot and Interactions plot. Can you guess which factors and interactions are significant?
  2. Which Factors and Interactions are significant when you use the ANOVA GLM ?
  3. How much of the variation in Sales does the model account for?
  4. Which Property Size and Region will generate the highest Sales?
  5. DO the VIFs indicate any problems with the model?
  6. Do the residuals indicate any issues with your model ?

Set-Up 1

Analysis 1

1.  Click Stat<<ANOVA<< Main Effects Plot

2. Complete the menu as shown below and then click OK.

As Prop_Size  has the greatest y-axis range we can say that as a Factor it is having more effect on Sales than Region. But Prop_Size has 5 levels and Region only has 3 so on a Factor level basis Region may be having more effect on Sales.

At this time we cannot say which of the factors are statistically significant.

There is a single interaction plot as there are only two Factors. It appears that the trend is for Region B to have the highest sales. However, this is reversed for Band4 and this could be a significant interaction.

Set-Up 2

1.  Click Stat<<ANOVA<< Interactions Plot.

2. Complete the menu as shown below and then click OK.

Analysis 2

Next Page