Re:Cellphone-Cancer Link Found in Government Study
by psychometry on Friday May 27, 2016 @03:43PM
(#11788408)
Biostats grad student here.
"But the probability of this study having failed to control for some other variable is higher than..."
It was a randomized block experiment. You don't need to assume that there were no unmeasured confounders in order to make causal inferences. That assumption is only needed in cross-sectional and cohort studies.
"But the probability of this study having failed to control for some other variable is higher than..."
It was a randomized block experiment. You don't need to assume that there were no unmeasured confounders in order to make causal inferences. That assumption is only needed in cross-sectional and cohort studies.