The correct statistical measures to reflect differences





--------------------------------------------------------------------------------

Scenario A:

treatment num percent
drug 75 54%
placebo 300 12%

as S.D. is not known for both groups, C.I. cannot be applied. what else
statistics can be used to reflect the difference?


Scenario B:

Site IDs accident number
1 1
2 0
3 2
4 3
5 0
.....

1995 0
1996 2
1997 1
1998 0
1999 3
2000 0


The accident event is rare but we want to know whether it occurs more than
purely by chance. What statistical measure can we use for reference?


.