Individual Problem Set Fall SPSS

profilenjosh
individual_problem_set_fall_2015.docx

Follow all instructions carefully in presenting your answers. Be sure to show all your working. (Handwritten responses are fine.) You will not need SPSS for questions 1-3. For question 4, please download the housing dataset (from Latte), then import it into SPSS for analysis.

1) Jet Blue Airlines examined the bags of 80 passengers and found that 20% of the bags were overweight.

a) Based on this sample, what is the 95% confidence interval for the proportion of bags that are overweight? [6 points]

b) What is the minimum sample size the airline would need to estimate with 95% confidence to obtain a margin of error of +/- 3% for this estimate of the percentage of overweight bags? [6 points]

2) A factory recently took a sample to assess the quality of its candy output, looking at three different types of candy, and how many of each type of candy were damaged during the manufacturing process:

Candy

# damaged

Total # candies counted

Apple hard candy

15

50

Chocolate chew

18

50

Nut cluster

30

100

The factory management would like to determine whether the proportion of candy that is damaged is different for these three types of candy.

a) Construct a contingency table for these data. [2 points]

b) Is the proportion of candy that is damaged different for these three types of candy? (Calculate the appropriate statistic, give the p-value, and state your conclusion.) [6 points]

3) A manufacturer of headphones is interested in the sales of a particular headphone model in its stores in 8 airports. Some of these stores are located on the West and some on the East coast of the U.S. Also, the manufacturer recently conducted an advertising campaign. The sales before and after the advertising campaign, which it ran in February using billboards in the airports, are shown below (i.e., data for sales in those stores in January and data for sales in the same stores for March.)

(Some descriptive statistics have also been provided in the table. You will need to decide which ones you need for your calculations in answering the questions below.)

Store

Location

Sales in Jan

Sales in March

Change in sales

1

East coast

195

230

35

2

East coast

220

240

20

3

East coast

220

250

30

4

East coast

245

265

20

5

West coast

130

157

27

6

West coast

130

140

10

7

West coast

80

99

19

8

West coast

185

207

22

Summary statistics

All stores

Mean

175.63

198.50

22.88

SD

56.72

59.65

7.68

East coast

Mean

220.00

246.25

26.25

SD

20.41

14.93

7.50

West coast

Mean

131.25

150.75

19.50

SD

42.89

44.71

7.14

To get full points when answering each part below be sure to: calculate an appropriate statistic, state the result of the test, and state your conclusion.

a) Looking at all the stores, is there a difference in sales between January and March? [6 points]

b) Did the campaign have a different effect on sales for stores on the East coast versus on the West coast? [6 points]

c) Was there a difference in sales in January for stores on the East coast versus on the West coast? [6 points]

4) Below are data for 40 houses located in one of two neighborhoods (A or B).

(This data is also provided in an Excel spreadsheet on the website for the class. Open the data in SPSS and conduct the analyses required to answer the questions. Be sure to paste output (i.e., tables) from SPSS into your answers where that is requested or else you will lose points.)

Neighborhood

Appraised Land Value

Appraised Value of Improvements

Sale Price

Has a yard? (yes/no)

A

56658

53806

255000

no

A

93200

11121

422000

no

A

76125

78172

290000

no

A

28996

5864

305900

no

A

30000

64831

118500

yes

A

30000

50765

93900

yes

A

46651

8573

191500

yes

A

45990

91402

184000

yes

A

42394

98181

168000

yes

A

47751

3351

169000

yes

A

63596

2182

208500

yes

A

51428

72451

264000

yes

A

54360

61934

237000

yes

A

65376

34458

286500

yes

A

42400

15046

202500

yes

A

40800

92606

168000

yes

A

12170

22786

375000

yes

A

24637

90598

169900

yes

A

30600

80858

135000

yes

A

44730

99047

176000

yes

B

38979

25946

140000

no

B

14861

59258

74900

no

B

14976

48957

57300

no

B

15244

55169

87500

no

B

18260

59267

82000

no

B

16680

55525

78000

no

B

53421

19792

175000

no

B

31417

99413

185000

no

B

32311

75343

123000

no

B

26817

78726

108000

no

B

24564

66533

108000

no

B

24564

71149

112900

no

B

27640

85347

106000

no

B

29656

78968

147500

no

B

13440

41177

61000

yes

B

45765

81227

320000

yes

B

16680

72867

99500

yes

B

17020

61935

93000

yes

B

25751

82259

110000

yes

B

25751

64568

100500

yes

a) Give appropriate summary statistics (one measure of central tendency and one measure of

variation) for each of the 3 variables Appraised Land Value, Appraised Value of Improvements, and Sale Price, calculated separately for neighborhoods A and B. Important: PROVIDE ONLY ONE (APPROPRIATE) CENTRAL TENDENCY MEASURE AND ONE (APPROPRIATE) MEASURE OF VARIATION FOR EACH VARIABLE FOR EACH NEIGHBORHOOD. [6

points]

b) Based on this data sample, do neighborhoods A and B differ in the number of houses with and without yards? In your answer be sure to calculate an appropriate statistic, state the result of the test, and state your conclusion. (Paste the output from SPSS for the statistical test that you do in your answer, as well as stating your conclusion and writing out the appropriate statistic that supports your conclusion.) [6 points]

c) Based on this data sample, do houses in neighborhoods A and B have different sale prices? (In your answer be sure to calculate an appropriate statistic, state the result of the test and state your conclusion.) (Paste the output from SPSS for the statistical test that you do in your answer, as well as stating your conclusion and writing out the appropriate statistic that supports your conclusion.) [6 points]

d) Provide a correlation matrix for Appraised Land Value, Appraised Value of Improvements and Sale Price for neighborhood B only (you will need to split the data to do this – in SPSS under the Data menu use the "split file" command, split by neighborhood, and select "organize output by groups"). In words, explain the meaning of the correlation between Sale price and Appraised Land Value and the meaning of the correlation between Appraised Land Value and Appraised Value of Improvements. [6 points]

Note: make sure you deselect "split file" after doing this question part, so that you analyzing all the cases for the next two parts.

e) Imagine you are interested in the relationship between house Sale price and Appraised Land Value while controlling for any effects of Appraised Value of Improvements. Conduct a linear regression that allows you to test this relationship (using data for all the houses, i.e., from both neighborhoods). State your conclusion about the relationship, and provide the statistics that support your conclusion. (Paste your SPSS output for this regression into your answer.) [6 points]

f) Imagine you are interested in the relationship between house Sale price and Neighborhood, while controlling for any effects of Appraised Land Value and Appraised Value of Improvements on Sale price. Conduct a linear regression that allows you to test this relationship. State your conclusion about the relationship, and provide the statistics that support your conclusion. (Paste your SPSS output for this regression into your answer.) [6 points]