That's true. One option might be to do it with 20 different
Posted on: October 6, 2017 at 13:37:19 CT
JeffB
MU
Posts:
72409
Member For:
21.47 yrs
Level:
User
M.O.B. Votes:
0
sets of people per city, holding the variables as close to even as is reasonably possible, and doing so in 20 cities.
Then tally up the survivors and see if there is a statistically significant difference.