Tomorrow I'm giving a lecture to residents and fellows at St. Chris Children's Hospital about collaborating with a statistician.

When I was asked to do this, I was invited to give a 1 hour talk about how to analyze data. This is a common request, and while I appreciate the sentiment, I spent several years learning to analyze data (and am still learning) and it can’t be summarized effectively in 1 hour.

As we drilled down to what I could speak about in an hour, I brought up several single topics: getting started with data analysis; how to approach power/sample size; effective collaborations.

As a longtime collaborative statistician, talking about effective collaborations is one of my favorite topics to speak to clinicians about. Especially to early career clinicians.

If we can teach early career scientists about the need for and process of team science early on, we can have better science more quickly.

I always feel like I’m walking a fine line when I give this talk – many clinicians are very statistically savvy, and I don’t intend to or want to offend anyone’s skills.

However, just as clinicians spend a long time learning to treat patients, statisticians spend a long time learning to analyze data.

By working together, we can have a much better product, since the clinician can really focus on the clinical/health aspects of the research, and the statistician can really focus on the analytic aspects.

Ultimately, it is a win-win!