1.
Robert
Putnam’s research has stimulated
interest in the role played by voluntary associations
in American democracy. Putnam’s work seems to suggest that, when
people get involved
in groups and help make collective decisions
for the group, they develop participatory skills. These
participatory skills, in turn, cause
people to participate more in politics – voting at higher rates
than people who are not involved
in any groups.
A.
This explanation says that the
causal relationship between the independent variable and
the dependent variable is
mediated by an intervening
variable.
ANSWER: The independent variable is?
ANSWER:
The dependent variable is?
ANSWER: The
intervening variable
is?
B.
Based
on the explanation, write a hypothesis in which the intervening variable
is the dependent variable.
ANSWER:
C.
Based
on the explanation, write a hypothesis in which the intervening variable
is the independent variable.
ANSWER:
2.
Explain a
linear relationship without references to a graph.
ANSWER:
3.
Explain the
difference between reliability and validity of
measurements. Elaborate.
ANSWER:
4.
When is it appropriate
to use comparison of means instead
of cross tabulation?
ANSWER:
5.
Each
of the following conclusions is based on a relationship between X and Y
that could be spurious.
For each one:
(i)
Identify a
plausible control variable,
Z.
(ii) Briefly describe
how Z might be affecting the relationship between X and Y.
A.
The level
of ice cream sales (X) and crime rate (Y) are strongly related: As sales go
up, so does the crime rate. Conclusion: To reduce
the crime rate, ice cream should be prohibited.
ANSWER: I
ANSWER: II
B.
When one looks
at the relationship between
marital status (X) and party identification (Y), one finds: Married
people are more likely
to be Republican than people that are not married.
Conclusion: Getting
married causes people to become Republican.
ANSWER: I
ANSWER: II
C.
Individuals’
attendance at religious services (X) is related to the number of children they have (Y). Conclusion: Declining religious
attendance causes declining birthrates.
ANSWER: I
ANSWER: II
6.
Suppose you want to model a set of interaction relationships between
Catholicism, religious attendance,
and abortion beliefs. You think that the
positive effect of religious attendance on anti-abortion attitudes is significantly stronger
for Catholics than non-Catholics. To construct the
interaction model, you
will build on the base effects of the model
y = a + b1 (Catholic) + b2
(high attendance), where “Catholic” is a Catholic/non-Catholic dummy (Catholics
are coded 1, non-Catholics coded 0) and
“high attendance” is a high attendance /low attendance dummy (frequent attenders are coded 1, infrequent attenders are coded 0). Before you
specify the model, you will
need to compute an interaction variable.
Answer:
A. The interaction variable is computed by multiplying times .
B.
Which of the following groups of respondents will have a value of 0 on the interaction variable?
i)
Catholic
low-attenders
ii) non-Catholic low-attenders
iii)
Catholic
high-attenders
iv)
non-Catholic
high-attenders
C.
Which of the following groups of respondents will have a value of 1 on the interaction variable?
i)
Catholic
low-attenders
ii) non-Catholic low-attenders
iii)
Catholic
high-attenders
iv)
non-Catholic
high-attenders
D. Write out the
interaction model to be estimated.
Answer:
E.
Focus
on the coefficient that estimates the interaction effect. If your idea is
correct – that the positive
effect of religious attendance on anti-abortion attitudes is significantly stronger for Catholics than non-Catholics – then would you
expect the sign on the coefficient
to be:
i)
negative
ii) positive
iii)
close to
0
7.
For this
question, please refer to Attachment 1.
A.
Write out
the regression equation for
this model.
Answer:
B.
Which variable
has the strongest relationship with Frequency of contributing money to charity?
Answer:
C. How much variation
in the dependent variable is explained
by
the model?
Answer:
D.
If you
remove “Age of respondent” from
the model,
will R2 probably increase, decrease, or stay about the
same? Justify your answer.
Answer:
8.
For this question, please refer to
Attachment
2.
A.
Summarize in
a sentence the relationship shown in the tables.
Answer:
B.
Identify the control variable.
Answer:
C.
What is the
effect of the control variable? (e.g., change in
intensity, direction, little or no change).
Justify your answer.
Answer:
D. What
does 23.1% in the crosstabs table represent?
Answer:
9.
For this
question, a regression model was created
and run to test the effect of a respondent’s parents
and/or spouse’s level of educational attainment on the
respondent’s own level
of educational
attainment, measured in years
of school completed. The output
was as follows:
MODEL RESULTS: R2 (adj.) = .378 F= 109.28 p< .001
Table of Coefficients:
|
VARIABLE
|
B
|
Beta
|
t
|
p
|
|
(Constant)
|
5.574
|
|
|
|
|
Father’s
Highest Yr. School
Completed
|
.095
|
.136
|
2.995
|
.003
|
|
Mother’s Highest
Yr. School Completed
|
.104
|
.121
|
2.600
|
.010
|
|
Spouse’s Highest
Yr. School Completed
|
.445
|
.477
|
12.491
|
.000
|
Dependent: Respondent’s Highest Yr. School Completed
Answer the following
questions about this model:
A.
Bob’s
mom has 12 years of formal education,
but his dad found school difficult and dropped out after only 10 years.
His wife Margaret, a true braniac,
has 20 years of formal education.
Make your best prediction of the number of years
of formal
education Bob has based
on the information provided
above. How did you
arrive at the value of that prediction?
Answer:
B.
Of the independent variables listed, which one
has the strongest effect on an individual’s level
of educational attainment? Which one has the weakest effect? If you were asked based solely
on these variables whether
it was more important for someone’s schooling to
“be born into brainy” or “marry brainy,”
which would you say? What’s
the tricky causal issue with that question?
Briefly explain your answers.
Answer:
Attachment 1: The following regression
model uses data from the General Social Survey to estimate the frequency of contributing money
to charities based on socioeconomic status,
empathy, degree, and age.
Coefficientsa
|
Mode
|
l
|
Unstandardized Coefficients
|
Standardized Coefficients
|
t
|
Sig.
|
|
|
B
|
Std. Error
|
Beta
|
||||
|
1
|
(Constant)
|
.699
|
.758
|
.055
|
.922
|
.359
|
|
|
Socioeconomic Index of Respondent
|
.004
|
.011
|
.418
|
.677
|
|
|
|
Applicability of description as empathic
|
.257
|
.124
|
.200
|
2.072
|
.041
|
|
|
Highest Degree of Respondent
|
.513
|
.157
|
.438
|
3.273
|
.002
|
|
|
Age of Respondent
|
.005
|
.008
|
.058
|
.594
|
.554
|
a. Dependent Variable: Frequency of contributing money to charity
Model Summary
|
Model
|
R
|
R Square
|
Adjusted R Square
|
Std. Error of the Estimate
|
|
1
|
.514a
|
.264
|
.227
|
1.353
|
a. Predictors: (Constant), Age of Respondent, Applicability of description as
empathic, Socioeconomic Index of Respondent, Highest Degree of Respondent
Attachment 2: Variables Degree
(Recoded Degree) and Frequency
of giving (recoded
Frequency of contributing money
to charity).
Frequency of giving * Degree Cross tabulation
|
|
Degree
|
Total
|
||||||
|
High School or Less
|
Some College
|
At least college grad.
|
||||||
|
frequency
|
Not
|
at
|
all
|
Count
|
21
|
0
|
0
|
21
|
|
of giving
|
% within
Degree
|
35.0%
|
.0%
|
.0%
|
23.6%
|
|||
|
1-3+ past year
|
Count
|
26
|
4
|
10
|
40
|
|||
|
% within
Degree
|
43.3%
|
100.0%
|
40.0%
|
44.9%
|
||||
|
At least once a
|
Count
|
13
|
0
|
15
|
28
|
|||
|
month past ye
|
ar % within
Degree
|
21.7%
|
.0%
|
60.0%
|
31.5%
|
|||
|
Total
|
Count
|
60
|
4
|
25
|
89
|
|||
|
|
% within
Degree
|
100.0%
|
100.0%
|
100.0%
|
100.0%
|
|||
Chi-Square Tests
|
|
Value
|
df
|
Asymp. Sig. (2-sided)
|
|
Pearson Chi-Square
|
22.578a
|
4
|
.000
|
|
Likelihood Ratio
|
28.401
|
4
|
.000
|
|
Linear-by-Linear Association
|
16.910
|
1
|
.000
|
|
N of Valid
Cases
|
89
|
a. 3 cells (33.3%) have expected
count less than 5. The minimum expected
count is .94.
Symmetric Measures
|
|
Value
|
Asymp.
Std. Erro a
r
|
Approx. b
T
|
Approx. Sig.
|
|
|
Nominal by
|
Phi
|
.504
|
.102
|
4.809
|
.000
|
|
Nominal
|
Cramer's V
|
.356
|
.000
|
||
|
Ordinal by Ordinal
|
Gamma
|
.707
|
.000
|
||
|
N of Valid
Cases
|
|
89
|
|||
a. Not assuming the null hypothesis.
b. Using the asymptotic standard error assuming
the null hypothesis.
Symmetric Measures
|
Gender of Respon
|
dent
|
Value
|
Asymp.
Std. Erro a
r
|
Approx. b
T
|
Approx.
Sig.
|
|
|
MALE
|
Nominal by
|
Phi
|
.592
|
.148
|
3.574
|
.004
|
|
|
Nominal
|
Cramer's V
|
.418
|
.004
|
||
|
|
Ordinal by Ordinal
|
Gamma
|
.670
|
.000
|
||
|
|
N of Valid
Cases
|
|
44
|
|||
|
FEMALE
|
Nominal by
|
Phi
|
.487
|
.129
|
3.240
|
.031
|
|
|
Nominal
|
Cramer's V
|
.344
|
.031
|
||
|
|
Ordinal by Ordinal
|
Gamma
|
.785
|
.001
|
||
|
|
N of Valid
Cases
|
|
45
|
|||
a. Not assuming the null hypothesis.
b. Using the asymptotic standard error assuming the null hypothesis.
Chi-Square Tests
|
Gender of
Respondent
|
Value
|
df
|
Asymp. Sig. (2-sided)
|
|
|
MALE
|
Pearson Chi-Square
|
15.404a
|
4
|
.004
|
|
|
Likelihood Ratio
|
19.685
|
4
|
.001
|
|
|
Linear-by-Linear Association
|
9.123
|
1
|
.003
|
|
|
N of Valid
Cases
|
44
|
||
|
FEMALE
|
Pearson Chi-Square
|
10.672b
|
4
|
.031
|
|
|
Likelihood Ratio
|
11.986
|
4
|
.017
|
|
|
Linear-by-Linear Association
|
8.552
|
1
|
.003
|
|
|
N of Valid
Cases
|
45
|
||
a. 5 cells (55.6%) have expected count less than 5. The minimum
expected count is . 82.
b. 6 cells (66.7%) have expected count less than 5. The minimum
expected count is . 20.
frequency of giving * Degree * Gender of Respondent Crosstabulation
|
Gender of Respondent
|
Degree
|
Total
|
||||
|
High School or Less
|
Some College
|
At least college grad.
|
||||
|
MALE
|
frequency of giving
|
Not at all Count
% within Degree
|
12
46.2%
|
0
.0%
|
0
.0%
|
12
27.3%
|
|
1-3+ past year Count
% within Degree
|
8
30.8%
|
3
100.0%
|
7
46.7%
|
18
40.9%
|
||
|
At least once
a Count
month past year % within
Degree
|
6
23.1%
|
0
.0%
|
8
53.3%
|
14
31.8%
|
||
|
Total Count
% within Degree
|
26
100.0%
|
3
100.0%
|
15
100.0%
|
44
100.0%
|
||
|
FEMALE
|
frequency of giving
|
Not at all Count
% within Degree
|
9
26.5%
|
0
.0%
|
0
.0%
|
9
20.0%
|
|
1-3+ past year Count
% within Degree
|
18
52.9%
|
1
100.0%
|
3
30.0%
|
22
48.9%
|
||
|
At least once
a Count
month past year % within
Degree
|
7
20.6%
|
0
.0%
|
7
70.0%
|
14
31.1%
|
||
|
Total Count
% within Degree
|
34
100.0%
|
1
100.0%
|
10
100.0%
|
45
100.0%
|
||
No comments:
Post a Comment