While the name “Brian Wansink” may not be immediately familiar, some of the research he’s known for probably is. Dr. Wansink’s research into how people make decisions about what and how much to eat gets a lot of play in the popular media. He’s one of those rock star scientists.
At the end of the academic year, he’ll be a retired scientist. His departure from Cornell was prompted by questions raised about some of his research techniques, leading to the retraction of several of his published papers and ongoing scrutiny of others. His comments about the situation are available on his web page.
Interestingly, to us at the IRB anyway, some questions about Dr. Wansink’s work involved questions about his data analysis methods. We here at the IRB tend to get questions about whether a protocol’s data analysis language is subject to IRB oversight. One person even responded to a data analysis contingency with only a “Is this an ethical issue?” comment. Because one of the criteria for research approval is that risks to subjects are minimized by using procedures consistent with sound research down. Data analysis that supports your research question is one aspect of sound research design, and the IRB can’t approve a study without it.