The first analysis used age, sex, race, stage,
histology, initial treatment, and subsequent treatment as
independent variables. The number of months from diagnosis
to death was the dependent variable. Because of the ex-
clusion of cases with missing values for stage or histology,
the analysis was of 221 of the 266 cases in the sample. A
comparison of values on all variables revealed only slight
differences between the sample of 266 and the group of 221.
A detailed comparison of the values on all variables for this
site and the other two cancer sites is given as Appendix F.
The regression was significant with F (28,192) = 3.23,
p = .0001. The multiple correlation coefficient was .57 and
its square was .32.
The results of partial regression were used to evaluate
the predictive efficacy of each independent variable. Table
13 summarizes these results.
Table 13
Regression Analysis of the Main Effects of Independent
Variables for Lung Cancer
Variable df F p
Age 1 4.38 .0376
Sex 1 0.24 .6240
Race 1 0.57 .4522
Stage 7 2.99 .0054
Histology 4 1.39 .2395
Initial treatment 11 1.95 .0357
Subsequent treatment 3 4.87 .0029