MGMT2263 Tutorial Sheet #6

1) A real estate agent wanted to see how much a home selling price (in thousands of dollars) depends on house size (in thousands of square feet), number of bedrooms and whether it has an attached garage or not (attached = 1). A sample of 8 homes produced the following results:

Home size	Bedroom	Garage	Price
1.36	2	1	2.49
1.6	3	1	3.02
1.88	3	0	3.65
1.38	2	0	2.34
1.02	2	1	2.16
1.33	2	0	2.94
2.04	4	1	3.75
1.56	3	1	2.84

Using the criteria of adjusted r², ANOVA p-value, t-test p-values (compared to 5%) and presence of multicollinearity, which one of the following models is the best? Fill in the following table and state why you chose the model you did based on these four criteria. (see key for solution)

Variables	Adj. R^2	ANOVA p-value	t-test: are all variables significant? (yes or no)	Any multicollinearity? (yes or no)
All variables
Home size, garage
Home size

2) A deli with 3 checkout counters took 100 sales from each counter and classified them as shown in the following crosstab:

	Counter A	Counter B	Counter C	Total
Under $30	48	35	27	110
$30 to under $40	34	38	35	107
$40 and over	18	27	38	83
Total	100	100	100	300

Based on organizing the data this way, do the sales depend on the counter? Test at 5%. To what degree does the amount of the sale depend on the counter? (chi-square = 13.623; conclude the sales depend on the counter; 15.07%)