Now run the survdiff function for any differences in education:
> survdiff(SurvObj ~ Xeducation, data = ChurnStudy)Call:survdiff(formula = SurvObj ~ Xeducation, data = ChurnStudy) N Observed Expected (O-E)^2/E (O-E)^2/VXeducation=Bachelor's Degree 1186 592 594.8 0.01291 0.07392Xeducation=Doctorate Degree 88 45 45.4 0.00301 0.00382Xeducation=Master's Degree 214 114 110.9 0.08896 0.12417 Chisq= 0.1 on 2 degrees of freedom, p= 0.94
Now look at the p value. The p value for the chi-square test is 0.94. That means that we cannot conclude that a significant difference exists for the education survival curves.
This also suggests having a closer look at the following plot. We can see ...