0
$\begingroup$

Edit: Please Find the Table in Image

I am conducting research on 2 groups and lets call them A and B and these 2 groups are 2 Variables of my dataset. These are categorical data.. Below is sample of my data.

enter image description here

Now I want to know if group A is better than group B..

Using ChiSquare test of Independence I can know if there is any significant difference between these 2 groups. But how can I find which group is better than the other? Which Test I need to perform?

Improve this question
$\endgroup$
7
  • $\begingroup$ What do these counts represent? $\endgroup$
    – Demetri Pananos
    Commented Apr 14, 2020 at 3:18
  • $\begingroup$ It depends what you mean by better. And given that it would be important to know what the data means. Do you want a higher average? Do you want one to be more consistent than the other? $\endgroup$
    – ajax2112
    Commented Apr 14, 2020 at 3:30
  • $\begingroup$ @DemetriPananos Hi, I have edited my post with picture of table.. $\endgroup$
    – chaitanya
    Commented Apr 14, 2020 at 4:39
  • 1
    $\begingroup$ Group A has the higher observed survival rate. Your significant $\chi^2$ test says that the observed difference in survival rates groups A and B is unlikely if the survival rates are the same. In that case, the observed difference is exceptionally unlikely if B in fact has the higher survival rate. With the usual caveats about p-values, you would doubt that B has even as high of a survival rate, let alone an even higher survival rate. $\endgroup$
    – Dave
    Commented Apr 14, 2020 at 4:48
  • 1
    $\begingroup$ @Dave I am proposing the same methodolgy as well.. But I am wondering If I can use 2 sample proportion z test with H0: proportion of A = proportion of B and HA: proportion of A > proportion of B $\endgroup$
    – chaitanya
    Commented Apr 14, 2020 at 5:24

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.