Scribd
Upload
123 views

Running Head: COMPUTER ASSIGNMENT 3 1

1. The document reports the results of four statistical analyses related to transportation costs, valve openings, and family size. 2. The first analysis uses ANOVA to find significant differences in transportation costs between four cities. Los Angeles costs differ from New York and Chicago. 3. The second uses two-way ANOVA to determine the impact of machine and shift factors on valve openings, finding only shift has an effect. 4. The third applies chi-square test of independence to reveal a dependency between family size and social class.

Uploaded by

AssignmentLab.com
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
123 views

Running Head: COMPUTER ASSIGNMENT 3 1

1. The document reports the results of four statistical analyses related to transportation costs, valve openings, and family size. 2. The first analysis uses ANOVA to find significant differences in transportation costs between four cities. Los Angeles costs differ from New York and Chicago. 3. The second uses two-way ANOVA to determine the impact of machine and shift factors on valve openings, finding only shift has an effect. 4. The third applies chi-square test of independence to reveal a dependency between family size and social class.

Uploaded by

AssignmentLab.com
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Running head: COMPUTER ASSIGNMENT 3

Computer Assignment 3 An Assignment Submitted by Name of Student Name of Establishment Class XXXX, Section XXXX, Fall 2012

COMPUTER ASSIGNMENT 3 Computer Assignment 3 #1 1. Research question: Due to the fact that family transportation costs are normally higher than anticipated by most people, it is of interest whether geographical location (e.g., place of residence) produces a significant impact on such costs, and if it does, whether the four major locations (Atlanta, New York, Los Angeles, and Chicago) differ significantly from each other. 2. The null hypothesis is that average transportation costs in all four cities are not significantly different from each other. H0: = = = . Alternative hypothesis is that one or more locations produce a significant impact on the transportation costs. In other words, there exist two means that are different from each other. HA: . 3. One-way ANOVA delivered F-statistics of 11.03:
One-way ANOVA: atl, ny, la, chi
Source Factor Error Total DF 3 16 19 SS 456630 220770 677400 MS 152210 13798 F 11.03 P 0.000

4. The obtained p-value is equal to zero, which implies that the null hypothesis can be rejected on a 95% confidence interval. 5. The rejection of null hypothesis implies that we have different means of transportation costs in some of the four cities. In order to find out which cities are significantly different from each other, Tukey method which utilizes pairwise comparisons of means was employed. The test delivered the following results:
Individual 95% CIs For Mean Based on Pooled StDev

COMPUTER ASSIGNMENT 3
Level atl ny la chi N 5 5 5 5 Mean 591.0 350.0 776.0 563.0 StDev 78.5 155.1 134.6 82.7 --------+---------+---------+---------+(-----*----) (-----*----) (-----*----) (----*-----) --------+---------+---------+---------+400 600 800 1000

Grouping Information Using Tukey Method N 5 5 5 5 Mean 776.0 591.0 563.0 350.0 Grouping A A B B C

la atl chi ny

Means that do not share a letter are significantly different.

Thus, we conclude that Los Angeles average transportation costs are significantly different from those in Chicago and New York, and also New York is significantly different from other three cities. #2 1. Research question: It is of interest whether the difference in mean measurements of valve openings is impacted by shift or by machine or by both of them simultaneously. 2. In order to analyze the problem, there are three possible hypotheses that can be generated with appropriate alternative hypotheses.
: The population means of machine factor are equal. H0: = = = . : Valve opening means are different for at least two machines. : : The population means of the shift factor are equal. H0: = = . : Valve opening means are different for at least two shifts. : : There is no interaction between the two factors (all possible combinations of

two factors do not differ in means).

COMPUTER ASSIGNMENT 3
: At least one combination of the two factors differs in mean from other

possible combinations. 3. Two-way ANOVA reported the following F-statistics: 0.14; 7.47; 0.38 in the same order as above. 4. The following p-values were obtained for each of the F-statistics. A p-value of 0.9368 corresponds to F-stat of 0.14; a p-value of 0.0078 corresponds to F-stat of 7.47; a p-value of 0.8760 corresponds to F-stat of 0.38. This implies that the only
null hypothesis that is rejected is due to the fact that p-value is lower than

anticipated significance level of 0.05. 5. Therefore, it was discovered that the shift factor impacts mean measurements of valve openings while machine factor does not produce such impact. There is no significant impact of interaction between two factors on valve openings measurements. This implies that the model is additive. #4 1. Research question: It is of interest whether the number of children in a family is related to the familys social class. 2. H0: Number of children in the family and the familys social class are statistically independent. HA: Number of children in the family and the familys social class are statistically dependent. 3. Chi-square test of independence reported the following chi-square value: 34.963.
Chi-Square Test: lower, middle, upper
Expected counts are printed below observed counts Chi-Square contributions are printed below expected counts lower middle upper Total

COMPUTER ASSIGNMENT 3
1 7 7.56 0.041 9 17.06 3.808 34 46.06 3.159 47 26.32 16.245 97 18 14.33 0.939 38 32.36 0.982 97 87.38 1.060 31 49.93 7.177 184 6 9.11 1.063 23 20.58 0.285 58 55.56 0.107 30 31.75 0.096 117 31

70

189

108

Total

398

Chi-Sq = 34.963, DF = 6, P-Value = 0.000

4. The following p-value was reported: 0.000 which is lower than anticipated confidence level of 0.05. This implies that the null hypothesis can be rejected. 5. The rejection of null hypothesis implies that the number of children in the family and the familys social class are statistically dependent. However, the nature of this test does not allow making any further conclusions about the relation of the two variables (e.g., whether it is positive or negative).

COMPUTER ASSIGNMENT 3

You might also like