computer-packaged programs such as SAS. In performing this test, we refer
the value of thezscore to percentiles of the standard normal distribution; for
example, we compare the absolute value ofzto 1.96 for a two-sided test at the
5% level.
Example 11.14 Refer to the data for low-birth-weight babies in Example
11.11 (Table 11.14). With all four covariates, we have the results shown in
Table 11.17. Only the mother’s weightðp¼ 0 : 0942 Þand uterine irritability
ðp¼ 0 : 0745 Þare marginally significant. In fact, these two variables are highly
correlated: that is, if one is deleted from the model, the other would become
more significant.
The overall tests and the tests for single variables are implemented simulta-
neously sing the same computer program, and here is another example.
Example 11.15 Refer to the data for vaginal carcinoma in Example 11.10
(Table 11.13). An application of a conditional logistic regression analysis yields
the following results:
- Likelihood test for the global hypothesis:
wLR^2 ¼ 9 :624 with 2 df;p¼ 0 : 0081- Wald’s test for the global hypothesis:
wW^2 ¼ 6 :336 with 2 df;p¼ 0 : 0027- Score test for the global hypothesis:
w^2 S¼ 11 :860 with 2 df;p¼ 0 : 0421For individual covariates, we have the results shown in Table 11.18.
In addition to a priori interest in the e¤ects of individual covariates, given a
continuous variable of interest, one can fit a polynomial model and use this
TABLE 11.17
Variable Coe‰cient
Standard
Error zStatistic pValueMother’s weight 0.0191 0.0114 1.673 0.0942
Smoking 0.0885 0.8618 0.103 0.9182
Hypertension 0.6325 1.1979 0.528 0.5975
Uterine irritability 2.1376 1.1985 1.784 0.0745
422 ANALYSIS OF SURVIVAL DATA