This preview shows pages 1–3. Sign up to view the full content.
Sociology 63993—Exam 2—Page 1
S o c i o l o g y 63993
E x a m 2 A n s w e r K e y
A p r i l 1 , 2 0 1 1
I. TrueFalse.
(20 points) Indicate whether the following statements are true or false. If false, briefly explain why.
1.
A researcher computes a variable X
4
= X
2
+ X
3
. She then estimates the following two models using OLS regression:
Y =
1
X
1
+
2
X
2
+
3
X
3
+
Y =
1
X
1
+
4
X
4
+
She can use an incremental F test to determine which of these two models is better.
True. Since X
4
= X
2
+ X
3
, the incremental F is a test of whether or not
2
=
3
.
2.
A researcher runs the following:
. webuse nhanes2f, clear
. gen femage = female * age
. reg health female age femage
Source 
SS
df
MS
Number of obs =
10335
+
F(
3, 10331) =
549.60
Model 
2069.28161
3
689.760537
Prob > F
=
0.0000
Residual 
12965.7398 10331
1.2550324
Rsquared
=
0.1376
+
Adj Rsquared =
0.1374
Total 
15035.0214 10334
1.4549082
Root MSE
=
1.1203

health 
Coef.
Std. Err.
t
P>t
[95% Conf. Interval]
+
female 
.2752255
.0648373
4.24
0.000
.4023191
.1481319
age 
.0280887
.0009315
30.15
0.000
.0299146
.0262627
femage 
.0043295
.0012822
3.38
0.001
.0018162
.0068428
_cons 
4.78594
.0469616
101.91
0.000
4.693886
4.877994

These results show that age has a negative effect on the health of males and a positive effect on the health of females.
False. The effect of age is less negative for females (.0280887 + .0043295 =
.0237592) but it is still negative.
3.
A researcher has included several extraneous variables in her model. The larger her sample, the more serious this problem
will be.
False. Adding extraneous variables increases standard errors. Larger sample sizes
decrease standard errors.
4.
A researcher regresses income on education. She does not include any dummy variables or interaction terms involving
gender. One implication of this model is that, if it is true, the mean income for men will be the same as the mean income for
women.
False. If men and women differ in their mean levels of education, they will also differ in
their mean incomes.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document Sociology 63993—Exam 2—Page 2
5.
A researcher is interested in the relationship between bmi (Body Mass Index) and health. She does the following:
. webuse nhanes2f, clear
. gen bmi = weight/ (height/100)^2
. gen bmi2 = bmi * bmi
. reg health bmi bmi2
Source 
SS
df
MS
Number of obs =
10335
+
F(
2, 10332) =
111.24
Model 
316.928298
2
158.464149
Prob > F
=
0.0000
Residual 
14718.0931 10332
1.4245154
Rsquared
=
0.0211
+
Adj Rsquared =
0.0209
Total 
15035.0214 10334
1.4549082
Root MSE
=
1.1935

health 
Coef.
Std. Err.
t
P>t
[95% Conf. Interval]
+
bmi 
.0072049
.0152456
0.47
0.637
.0226794
.0370892
bmi2 
.0007416
.0002646
2.80
0.005
.0012601
.000223
_cons 
3.731409
.2147848
17.37
0.000
3.31039
4.152429

Based on these results, she should conclude that bmi is not related to health.
False. The results indicate that there is a curvilinear relationship between bmi and
health. Increases in body mass index are good up to point, but after that further
increases are harmful. (In other words, it isn’t good to be obese.)
II.
Path Analysis/Model specification (25 pts).
A sociologist believes that the following model describes the relationship between X1, X2, X3, and X4. All her variables are in
standardized form. The estimated value of each path in her model is included in the diagram.
a.
(5 pts) Write out the structural equation for each endogenous variable, using both the names for the paths
(e.g. β
42
) and the estimated value of the path coefficient.
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 02/29/2012 for the course SOC 63993 taught by Professor Richardwilliams during the Spring '11 term at Notre Dame.
 Spring '11
 RichardWilliams
 Sociology

Click to edit the document details