Rafe adds this, as well: The clinic approach is great, perhaps even phenomenal, if we can get the others to see that collaboration is key, and that they need to show up. If they are waiting until they run up against a wall they cannot climb, they have waited too long and it is too late. We have got to get to them early.

And the outcome from the biostatistics classes we teach ought to be a realization on the part of the investigator that they ought to consult a data professional as soon as the germ of a reasearch idea implants in the lining of their brains. Teaching investigators to point and click a two-way anova just furthers the illusion that statistics and data science are things one can do once or twice a year. We need to get the investigators to think (1) about distributions, not just means and (2) "oh goodness, this idea will deal with evidence and data and decisions! I'm just a phd physiologist; I'd better call a data scientist."

That's why we need more money.

Cindy: The clinics are designed to increase the quality and quantity of research in the Medical Center. It is a place the investigators can go for the help from biostatisticians no matter what their funding situation, although they are encouraged to have statisticians involved in their study as early as possible. On the other hand, the clinical researchers are encouraged to educate statisticians the scientific background before address their questions in the clinics with the possibility of inviting closly working statistician having lab-visit outside clinic. Simple analysis with careful thinking scientific background is better than advanced, well designed model with nonsensical scientific knowledge.
Topic revision: r1 - 16 Feb 2006, RafeDonahue

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback