MCC-UE 1040 Chapter Notes - Chapter 7-8: Cancer Cluster, Infant Mortality, Maternal Death
Document Summary
Way to learn how certain people can be about what they think they know. Can establish the probability that a study finding is the result of chance is the p value. Even when a low p value indicates that a result is statistically significant, there is still a chance that the result is not valid, even if all sources of bias are ruled out. Law of small probabilities: even the most improbable things are bound to happen occasionally. Few people w/fatal illness may recover miraculously causing others to believe they had found a cure when they had not. If cluster is large, unlikely to be a random variation. Studies w/large #s of subjects are more likely to be valid than small studies. Power is probability of finding an effect if there is, in fact, an effect. Important in evaluating screening tests, used as a secondary prevention approach to detect diseases so that they can be treated at an early stage.