This preview shows pages 1–3. Sign up to view the full content.
1
Homework # 1
ECO 7427, Spring 2011
ANSWER KEY
Prof. Sarah Hamersma
1.
The key to answering this question is to note that (1) the word “significant” should be
used with “statistically” when that is the proper meaning and (2) one should not talk about
the direction o
f an “effect” if the standard errors make the direction of the effect uncertain.
An ideal answer recognizes the relationship between the three variables: the change in
Medicaid coverage plus the change in private coverage must
add up to the change in “any
coverage
”
, and so the set of estimates taken together provide insight about crowdout.
My answer:
The coefficients on the interaction between the targetgroup dummy and the
postexpansion dummy presented in the second row provide our regression
estimates of the effects of the parentaleligibility expansions.
On average, the
expansions increased maternal Medicaid coverage by a statistically significant
2.7 percentage points, which is a large increase from the prior level of
participation.
Given this increase, we expect to find that either total coverage
(measured by “any”) increased as well, or it was crowded out by a decrease in
private coverage.
We are unable to precisely measure the effect on private
coverage; while the coefficient is .013, we cannot be confident that the effect is
negative since the coefficient has a tvalue of only about 1.
Given this
uncertainty in measuring crowd out, there is also uncertainty about the effects on
“any coverage.”
The coefficient on “any coverage” is positive, as expected based
on the other point estimates, but we cannot be confident that the true effect is
positive due to its large standard error.
Thus the key result that we can state with
confidence is that the expansions did appear to increase Medicaid coverage, but
we do not know the extent to which crowdout may have dampened these gains
overall.
2.
[ directions omitted]
Using STATA, produce a table of descriptive statistics (means,
standard deviations, mins and maxes) for each of the two data sets.
Briefly indicate your
reasoning for including these particular variables in your dataset.
. clear
. use CPS1993
. sum
Variable 
Obs
Mean
Std. Dev.
Min
Max
+
OCCURNUM 
14609
1.088028
1.428057
0
14
H_IDNUM 
14609
4.46e+11
3.30e+11
1.04e+07
1.00e+12
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document 2
PHF_SEQ 
14609
1.229721
.4906753
1
7
A_HGA 
14609
39.37799
2.74167
31
46
A_RACE 
14609
1.316038
.6872655
1
5
+
FOWNU18 
14609
.6000411
.9943722
0
9
A_FTPT 
14609
.1119858
.3534364
0
2
A_CIVLF 
14609
.7110001
.4533134
0
1
A_AGE 
14609
37.02587
13.79849
16
65
HG_ST60 
14609
51.67657
26.43724
11
95
+
A_SEX 
14609
2
0
2
2
A_MARITL 
14609
5.952769
1.159967
4
7
FKIND 
14609
3
0
3
3
year 
14609
1993
0
1993
1993
. use CPS2000
. sum
Variable 
Obs
Mean
Std. Dev.
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 06/07/2011 for the course ECO 7427 taught by Professor Hamersma during the Spring '08 term at University of Florida.
 Spring '08
 HAMERSMA

Click to edit the document details