Sociology 201 Third Quiztable of contents




Person
Height
Prejudice
Immunization
Sex
Dan
6 feet
Lo
Yes
M
Sam
5
Lo
No
M
Bill
6
Lo
Yes
M
Joe
5
Hi
No
M
Jim
6
Med
Yes
M
Ann
6
Hi
No
F
Kay
5
Hi
Yes
F
Sue
5
Hi
No
F
Kay
6
Hi
Yes
F

Use the above data for all problems on this exam which require data. Be sure to show all computations on these pages.  If you are asked to compute a statistic, compute it even though it may not meet the assumptions of the statistic and even though it may be wasteful to use that statistic on the data.  You may assume this is a random sample from some meaningful population.  Be sure you answer all 16 problems.



                                                                ________________________
                                                                        name

1. Why or why not use eta to answer this question: "What is the relationship between height and prejudice in this sample?"
a. Yes you can; this is the ideal statistic.
b. No, you cannot because prejudice is not interval.
c. Yes it is OK except it is wasteful to treat height as nominal.
d. Yes it is OK except it is wasteful to treat prejudice as nominal.
e. No you cannot because you need 5 or more cases in each nominal category.

2. Why or why not use the Kruskal-Wallis H to answer this question: "Can you generalize the association between height and prejudice to the whole population?"
a. Yes you can; this is the ideal statistic.
b. Yes it is OK except it is wasteful to treat prejudice as nominal.
c. No you cannot because you do not have 5 or more cases in each nominal category.
d. No you cannot because the Kruskal-Wallis H  does not measure relationship.
e. Yes it is OK except it is wasteful to treat height as nominal.
 


3. Why or why not use the Test of Significance for Gamma to answer this question: "Is the relationship between height and prejudice one you would expect to find in the whole population?"
a. Yes it is OK except it is wasteful to treat height as ordinal.
b. No you cannot because you need an N of 10 or more.
c. Yes it is OK except it is wasteful to treat prejudice as interval.
d. No you cannot because the Test of Significance for Gamma does not measure relationship.
e. No you cannot because prejudice is not ordinal.

4.  Why or why not use the Partial Gamma to answer this question: "What is the association between prejudice and height with immunization held constant?"
a. Yes it is OK except it is wasteful to treat height as ordinal.
b. No you cannot because immunization is not ordinal.
c. No you cannot because height is not ordinal.
d. Yes you can; this is the ideal statistic.
e. No you cannot because Partial Gamma does not measure relationship.

5. Why or why not use Theta to answer this question: "What is the association between prejudice and immunization?"
a. Yes it is OK except it is wasteful to treat prejudice as nominal.
b. No you cannot because you need 5 or more cases in each nominal category.
c. No you cannot because immunization is not ordinal.
d. No you cannot because Theta does not measure relationship.
e. Yes you can; this is the ideal statistic.

How would you interpret a Theta of .92 for the variables of height and sex?
a. Moderately small association between sex and height.
b. In 92% of the comparisons made, persons of different heights show systematic differences in their sex.
c. 92% more agreement than disagreement in the rank order of height and sex.
d. In 92% of the comparison made, persons of different sexes showed systematic differences in their height.
e. 8% or the variance in height cannot be explained by sex.

7. How would you interpret an eta of +.20 for the variables of prejudice and height? (you need to compute eta2  in order to know all three interpretations for eta)
a. 20% of the variance in height can be explained by prejudice.
b. Small negative relationship between prejudice and height.
c. 96% of the variance in height cannot be explained by prejudice.
d. 4% of the variance in prejudice can be explained by height.
e. 20% more agreement than disagreement in the rank order of height and prejudice.



8. How would you interpret a Z of 1.90 when your hypothesis says: "There is a positive association between prejudice and height in the whole population."
a. p = .01
b. The chances of being wrong if you generalize from the sample to the whole population are less than 5 in 100.
c. p = .05
d. You cannot generalize from the sample to the whole population because the chances of being wrong are too great.
e. p is less than 1 in 100.

9. How would you interpret an H of 2.12 for the variables of prejudice and sex?
a. p is greater than 5 in 100.
b. p = .05
c. The probability of being wrong if you generalize from the sample to the whole population is less than .05.
d. p = 1 in 100.
e. p is less than .01

10. How would you interpret a Partial Gamma of .55 for prejudice and height with sex held constant?
a. 55% more disagreement than agreement in the rank order of prejudice and height with sex held constant.
b. 55% of the variance in height can be explained by prejudice.
c. The chance of being wrong if you generalize form the sample to the whole population are 55 in 100.
d. Small negative association between prejudice and height with sex held constant.
e. 55% more agreement than disagreement in the rank order of height and prejudice with sex held constant.


11. Arrange your data in the proper form and compute Partial Gamma for the relationship between height and prejudice with immunization held constant.















12. Substitute the proper figures into the formula for the Kruskal-Wallis H, using the data in the table given below.  Do it even though you may think the data do not meet the assumptions for H.  Do not compute H; just put all the figures in the formula.



hi
med
lo
male
1
1
3
female
4
0
0

13. Compute Theta, using the data in the table in problem 12.












14.  Rearrange your data into the proper form for computing eta for the relationship between prejudice and height.  You may put it in either a grouped or ungrouped form.  Do not compute anything; just arrange the data.










15.  If your dad retired at age 60 when his generation's average age of retirement was 65 with a standard deviation of 3.4, when can you expect to retire if your generation's average age of retirement is 62 with a standard deviation of 2.1?










16.  Rearrange the tables below to control for alienation.

                                HI INCOME



hi alienation
med alienation
lo alienation
male
5
12
23
female
4
16
12

                                MEDIUM INCOME



hi alienation
med alienation
lo alienation
male
15
8
3
female
14
9
2

                                LOW INCOME



hi alienation
med alienation
lo alienation
male
25
15
1
female
14
12
5

answers