StATS: Protocol changes.
Dear Professor Mean, After I collected my data, I noticed a problem that I had not anticipated. I want to make some protocol changes and analyze my data differently. Can I do this?
Changes in a research protocol are always troublesome, but don't overreact. There's a cute quote by Stephen Senn in Statistical Issues in Drug Development (1997).
Medical statistician: one who will not accept that Columbus discovered America ... because he said he was looking for India in the trial plan.
So you don't want to be too obsessive about protocol deviations. On the other hand, look at all the controversy in the Florida vote recount for the year 2000 presidential election. One of the biggest issues was the lack of a single standard for vote counting that everyone agreed on prior to the election.
Protocol changes are not just limited to data analysis choices. Other changes might involve
Note that you if you allow for a protocol change in the protocol itself, that is not a problem. For example, you can say in the protocol, we will recruit patients for six months or until we recruit 200 patients, whichever comes first.
For any protocol deviation, you need to do is to assure yourself (and anyone who might be reading your research) that what you propose is a sensible deviation. There are several questions you can ask.
There are no right or wrong answers here. Protocol changes will hurt the credibility of your research, but you can mitigate the damages if you can show that your changes are rational.
This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Early stopping.