Académique Documents
Professionnel Documents
Culture Documents
N
RACE. * Low Birth
Weight
189
Missing
Perce
nt
100.0
%
Total
Perce
nt
N
0
Perce
nt
.0%
189
100.0
%
othe
E.
rs
blac
Count
BWT<=2
500g
Total
115
48
163
Expected Count
112.1
50.9
163.0
% within RACE.
70.6%
29.4%
100.0
%
15
11
26
Expected Count
17.9
8.1
26.0
% within RACE.
57.7%
42.3%
100.0
%
130
59
189
Expected Count
130.0
59.0
189.0
% within RACE.
68.8%
31.2%
100.0
%
Count
Total
Count
Risk difference : we can see that black women have low birth weight baby than other
races.
Risk difference
Relative Risk (RR) : we will see RR of the black women to have low birth weight
baby compared with the other races.
Risk Estimate
95% Confidence
Interval
Value
Lower
Upper
.569
.244
1.329
.818
.580
1.153
1.437
.865
2.387
N of Valid Cases
189
From the table above, we can see that the RR of black women to have low birth
weight baby compared to other races women is 1.437. It means that the risk of
having low birth weight baby will
be 1.437 times higher for black
women.
Chi-Square Tests
Value
Pearson ChiSquare
1.727
Asymp.
Sig. (2sided)
df
1
.189
Continuity
Correction
1.180
.277
Likelihood Ratio
1.653
.199
Exact Sig.
(2-sided)
.254
1.718
.190
Exact Sig.
(1-sided)
.139
Association
N of Valid Cases
189
a. 0 cells (.0%) have expected count less than 5. The minimum expected count
is 8.12.
b. Computed only for a 2x2 table
increase the risk of having low birth weight baby.
Then we look at the CI is 0.865 to 2.387, so there is 1 (one) between them. Then we
can say that there is no significant difference for the risk of having low birth weight
baby from black race
women and other races.
After we did Chi-Square Tests, we can see from the table above in continuity
correction row, that the significance is 0.277 (more than 0.05). It means that there
is no significant difference for the risk of having low birth weight baby between the
black women and other races.
We didnt use the Fitcher`s test because there is no cell that has expected count
less than 5.
4. Now, a non-inference task: Determine the age cutoff for age of mothers which
increases risk of low birth weight. Use a method of your choice, and explain how
your method works.
The new cut off based on the histogram that increased low birth weight is at 20 years old.
Research Question : Is there any association between the history of hypertension
of the mother with the occurance of low birth weight baby?
N
History of
Hypertension * Low
Birth Weight
189
Missing
Perce
nt
100.0
%
Total
Perce
nt
N
0
.0%
N
189
Perce
nt
100.0
%
No
Count
BWT<=25
00g
Total
125
52
177
121.7
55.3
177.0
70.6%
29.4%
100.0
%
12
8.3
3.7
12.0
41.7%
58.3%
100.0
%
130
59
189
130.0
59.0
189.0
68.8%
31.2%
100.0
%
Hypertension
Expected Count
% within History of
Hypertension
Yes
Count
Expected Count
% within History of
Hypertension
Total
Count
Expected Count
% within History of
Hypertension
Chi-Square Tests
Value
Pearson ChiSquare
4.388
Asymp.
Sig. (2sided)
df
1
.036
Continuity
Correction
3.143
.076
Likelihood Ratio
4.022
.045
Exact Sig.
(2-sided)
Exact Sig.
(1-sided)
.052
Linear-by-Linear
Association
4.365
N of Valid Cases
189
.037
a. 1 cells (25.0%) have expected count less than 5. The minimum expected
count is 3.75.
b. Computed only for a 2x2 table
.042
Risk Estimate
95% Confidence
Interval
Value
Lower
Upper
3.365
1.021
11.088
1.695
.862
3.333
.504
.296
.856
N of Valid Cases
189
The Relative Risk (RR) of mother without history of hypertension to have low birth
weight baby compared to mother with history of hypertension is 0.504. Or in the other
word we can say that the
Relative Risk of mother with history of hypertension to have low birth weight baby
Fisher`s exact test was choosen because there is one expected value that is 3.75
(less than 5). The Chi-Squared test cannot be performed if 20% of the expected
values is less than 5.
We can see from fishers exact value is 0.052 then it is >0.05, then we conclude that
we accept H0.
H0 = There is no association between the history of hypertension of the mother with
the occurance of low birth weight baby.