ext_105594 ([identity profile] unclebooboo.livejournal.com) wrote in [personal profile] rmd 2008-05-24 04:02 pm (UTC)

Correlation is not causation.

Since this is an observational study, and particularly since there's no hypothesis that might explain how the use of talcum powder would cause ovariance cancer it's really not appropriate to infer from this study that using talcum power causes ovarian cancer. It's possible that use of talcum powder is correlated with some other variable that does actually predict ovarian cancer. It's also quite possible that a simple type I error has occured and that this finding is the result of chance. Or, it might be that there really is something in talcum powder that causes ovarian cancer. The approriate response is for physicians and scientists replicate the results and to come up with hypotheses that might explain this result, followed by more carefully controlled studies.

Unfortunately, this is a good example for pointing out two problems. First, the majority of the public doesn't really understand much about statistics and the design of ressearch studies like this one. Second, people who communicate the results of such studies often don't do a good job of communicating the results in ways that the public can understand.

Post a comment in response:

(will be screened)
(will be screened if not validated)
If you don't have an account you can create one now.
HTML doesn't work in the subject.
More info about formatting

If you are unable to use this captcha for any reason, please contact us by email at support@dreamwidth.org