This preview shows pages 1–3. Sign up to view the full content.
Sociology 63993—Exam 1 Answer Key —Page 1
Sociology 63993
Exam 1 Answer Key
February 17, 2012
I.
TrueFalse.
(20 points) Indicate whether the following statements are true or false. If false, briefly explain why.
1.
Cohen and Cohen’s dummy variable adjustment method is useful when variables like gender or age have missing
values.
False. The method should not be used when values exist but are not known (and
values for gender and age surely exist). The method can be useful when values don’t
exist, e.g. father’s education is missing because there is no father in the family.
2.
R
2
is biased downwards.
False. It is biased upwards. Sampling error will always cause R
2
to be greater than
zero, i.e. even if no variable has an effect R
2
will be positive in a sample. When there
are no effects, across multiple samples you will see estimated coefficients sometimes
positive, sometimes negative, but either way you are going to get a nonzero positive
R
2
. Further, when there are many Xs for a given sample size, there is more opportunity
for R
2
to increase by chance. Adjusted R
2
corrects for this bias.
3.
The more “tolerant” a variable is (i.e. the less highly correlated it is with the other IVs), the smaller its unique
contribution to R
2
will be.
False. The more tolerant a variable is, the more unique (and higher) its contribution to
R
2
will be. You can see this via such formulas as
22
2
2
2
2
* (1
)
*
k
kk
k
YH
YG
k
X G
k
k
sr
R
R
b
R
b
Tol
′′
=
−=
−
=
4.
When you have more than one independent variable, random measurement error can cause coefficients to be biased
either upward or downward.
True. In bivariate regression, the bias will be downward, but once you have more than
one independent variable the bias can go in either direction.
5.
A DurbinWatson statistic of 4 or greater indicates that the case is an extreme outlier.
False. The DurbinWatson statistic checks for serial correlation.
II.
Short answer.
Discuss all three of the following problems. (15 points each, 45 points total.)
In each case, the
researcher has used Stata to test for a possible problem, concluded that there is a problem, and then adopted a strategy to address
that problem. Explain (a) what problem the researcher was testing for, and why she concluded that there was a problem, (b) the
rationale behind the solution she chose, i.e. how does it try to address the problem, and (c) one alternative solution she could
have tried, and why. (NOTE: a few sentences on each point will probably suffice – you don’t have to repeat everything that was
in the lecture notes.)
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentSociology 63993—Exam 1 Answer Key —Page 2
II1.
. use "http://www.nd.edu/~rwilliam/stats3/statafiles/rwm11.dta", clear
(German Health Care Panel Data, Riphahn Wambach Million (2003), Greene (2007))
. reg newhsat female age handdum educ married working if year==1984
Source 
SS
df
MS
Number of obs =
3874
+
F(
6,
3867) =
141.19
Model 
4483.9589
6
747.326483
Prob > F
=
0.0000
Residual 
20468.7796
3867
5.29319359
Rsquared
=
0.1797
+
Adj Rsquared =
0.1784
Total 
24952.7385
3873
6.44274168
Root MSE
=
2.3007

newhsat 
Coef.
Std. Err.
t
P>t
[95% Conf. Interval]
+
female 
.186618
.0870046
2.14
0.032
.3571973
.0160387
age 
.0388289
.0035637
10.90
0.000
.0458159
.031842
handdum 
2.341489
.1236374
18.94
0.000
2.58389
2.099088
educ 
.1089876
.0173546
6.28
0.000
.0749626
.1430125
married 
.2048268
.0918166
2.23
0.026
.0248133
.3848402
working 
.2955985
.0912629
3.24
0.001
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '11
 RichardWilliams
 Sociology

Click to edit the document details