Suppose You Are Investigating the Effect of the Neighborhood, Improvement

1 Exam 7 - A

Suppose you are investigating the effect of the neighborhood, improvement

value of a property and the land value of a property on the sales price of

the property. A random sample of 42 recently sold properties was collected

and the following variables were measured.

DY = 1 if > $48,500 , 0 if not (the sales price category)

(NEIGBRHD) = One of three possible neighborhoods C, D, or E

The results are below:

CONTINGENCY TABLE OF NEIGHBORHOOD BY PRICE CATEGORY

Frequency| less than| >

| $48,500 | $48,500| Total

------+------+------+

C | 4 | 6 | 10

------+------+------+

Neighborhood D | 16 | 3 | 19

------+------+------+

E | 4 | 9 | 13

------+------+------+

Total 24 18 42

Chi-Square value=10.577 with 2 d.f. PVALUE= 0.005

7. The test statistic value for determining if the neighborhood is related

to the sales price category would be

A. 4.706485

B. 0.05

C. 10.577

D. 2

E. 10.787

8. The contingency table (Dy being the columns and neighborhood being the

rows) would be analyzed using a(n)

A. chi-square test of variance

B. analysis of covariance

C. Bartlett's test

D. chi-square test of independence

E. analysis of variance

9. At alpha = 0.05 what is the rejection region value for the contingency

table test of neighborhood versus sales price category?

A. Reject the null hypothesis if the test statistic > 3.84146.

B. Reject the null hypothesis if the test statistic > 7.37776 or if the

test statistic < -7.37776.

C. Reject the null hypothesis if the test statistic < 5.99147.

D. Reject the null hypothesis if the test statistic > 5.99147.

E. Reject the null hypothesis if the test statistic > 5.99147 or if the

test statistic < -5.99147.

10. Which of the following is an assumption of the contingency table test?

A. The neighborhood dummy variables have the same variance for any

combination of sales price category.

B. The errors in neighborhoods are related to the previous errors in

neighborhoods.

C. The probability that a property falls into one of the six combinations

of neighborhood and sales price category does not change from property to

property.

D. The neighborhood dummy variables are normally distributed for any

combination of sales price category.

E. The expected number of properties in each of the six combinations of

neighborhood and sales price category should be smaller than 0.05.

Answers to Exam 7 -A

7. c

8. d

9. D

10. c

Exam 7 - B

Suppose you are investigating allegations of sex discrimination in the

hiring practices of a particular firm. An equal-rights group claims

that females are less likely to be hired than males with the same

background, experience, and other qualifications. A random sample of

13 males and a random sample of 15 females former applicants was

collected. The following variables were measured.

Y = 1 if hired, 0 if not (HIRED STATUS)

X3 = 0 if female, 1 if male

The results of the different analysis are

------Contingency Table ------

Hired Not Hired

Male 6 7 Chi-square

Sex test statistic

Female 3 12 value = 2.1841

1. Considering only Y and X3, what procedure would answer the

question: "Is there a relationship between the gender and whether they are hired or not?"

(A) multinomial test

(B) t-test of a slope

(D) chi-square test of mystification

(E) F-test of model usefulness

2. Consider only Y and X3. If there is independence,

what is the expected number of females who would not be hired?

(A) 4.82

(B) 12.00

(D) 5.00

(E) 10.18

3. Consider only Y and X3. Is there independence between the two?

(A) Fail to reject H0 since 2.1841 < 3.84146

(B) Fail to reject H0 since 2.1841 < 9.48773

(D) Fail to Reject H0 since 2.0573 < 3.84146

(E) Fail to reject H0 since 2.0573 < 9.48773

4. What procedure in NCSS would have calculate the test statistic value

for a contingency table?

(A) the one way anova procedure.

(B) cross-tabulation

(D) descriptive statistics.

(E) scatterplots.

Answers to Exam 7 - B

1. c

2. e

3. A

4. b

Exam 7 - C

A firm wants to determine what variables affect the number of

hours of labor needed to ship its products. Weekly data for the

last 20 weeks are available with the following variables

measured:

Y (LaborCat) = Labor category: 1 if >100 hours and 0 if not,

X3 (TruckCat) = 0 if percent of shipments by truck <= 50 and

1 if 50%

TRUCKCAT

| 0 1 Chi-square

------test statistic

LABORCAT 0| 3 9 value = 0.3571

1| 3 5

1. Given only the Labor category and the Truck Category, what

contingency table procedure could be used to analyze the data?

A. chi-square test of independence

B. simple linear regression

C. chi-square test goodness of fit

D. two independent sample t-test

E. chi-square test of sadism

2. If labor had been divided into four categories instead of

two and you wanted to test the relationship between Y and X3,

what would be the degrees of freedom of the test?

A. 1, 18

B. 1

C. 16

D. 3, 16

E. 3

3. Which of the following assumptions for the chi-square test

appears to be violated for the contingency table?

A. trials are identical

B. probability remains the same from trial to trial

C. all expected counts > 5

D. number of hours can fall into one of two categories

E. none appear to be violated

4. After finishing this exam, what will be the moral that you

will take with you?

A. The teacher gets a kick out of watching me suffer during

these exams.

B. With enough effort and examination any business problem can

be simplified beyond all understanding.

C. The next time I see a statistic, I will shoot it.

D. I'm free! I'm free! I'm free! I'm free! I'm free! I'm free!

I'm free! I'm free!

E. With effort, many business decision can be set up as a

statistical problem allowing a measure of the chance of a wrong

decision. (Hint, this is the correct answer)

1 a

2 e

3 c

4 e