Brian Healy, PhD, freely admits he was born for statistics. It’s the first thing he tells his biostatistics class. Then he sets out to convince them that they too will love and embrace statistics just as much as he does once they realize what it can do for their research.

If that all seems a bit evangelical, that’s because it is, statistically speaking. Healy wants to convert you to a statistics devotee. He wants you to know when the p-value shows its true value. To confidently comprehend confidence intervals. To progress through regression analyses sans aggression. To be able to read the literature and actually understand why the authors chose the methodology they did.

Healy developed the Applied Biostatistics course to do all that and more. A K-award staple, the course introduces commonly used statistical approaches in a self-paced learning platform, with a focus on applications to everyday research questions. He also teaches a shorter, introductory Essentials of Biostatistics mini-course, which is available via Harvard Catalyst On Demand.

Healy is faculty for Harvard Catalyst Education, associate professor of neurology at Harvard Medical School (HMS), and associate professor in biostatistics at the Harvard T.H. Chan School of Public Health.

What do you feel is the biggest misperception about biostatistics among researchers?

“There are people who think statistics is a black box. To me, the best thing that people learn from taking a class with me is that most of the time the complications can be understood.”

That it’s hard and they can’t do it. The best thing about statistics is that most of the time, at the end of the day, it allows you to understand what question you’re asking and what data you need to answer it, and then, importantly, how the data inform the next steps in your research.

There are people who think statistics is a black box. To me, the best thing that people learn from taking a class with me is that most of the time the complications can be understood. And although the vocabulary can be difficult, and some steps may be challenging, there are reasons for that. Once you understand the reasons, you see how it all connects.

The other issue is that sometimes people have a set of data and a question that don’t match up. We may hear: “Can’t you do your statistical magic?” Well, there’s no such thing as statistical magic. It’s about making sure you have the right question and the right data collected in the right manner to allow yourself to answer that question.

What one thing do you most want early investigators to know about biostatistics?

I think the main thing is that an understanding of statistics is going to greatly improve your research. It can clarify your scientific question, help you understand the model or the tools you are using, and enable you to read the literature more carefully, because you know which assumptions the authors are making based on the methods that they used.

I’m convinced that statistics education would benefit most researchers. Once they start learning about statistics, they might be surprised at how exciting and fun they find it, because of the value they get out of it.

When I first developed the Applied Biostatistics course in 2011-2012, most of the classes available focused on teaching the topic, not how to do it. The goal of our program from the beginning has been to ensure that people understand both the topic and how to apply it to their research question using current statistical software.

“I truly think that statistics and statistical thinking enable people to understand the medical literature and their own research better.”

How does the Applied Biostatistics course relate to Essentials of Biostatistics, the on-demand primer you’ve recorded?

Essentials of Biostatistics is more of an intro to biostatistics geared toward a medical audience. It’s for people who would like information about the general ideas of statistics. It covers a little bit on the p-value, a little bit on confidence intervals, a little bit on regression analysis, that kind of stuff. The Applied Biostatistics course is longer, more comprehensive, and includes practice with the software. It also goes deeper.

Personally, I love statistics. I think everybody should learn about statistics. It helps you understand the world in a way that allows you to kind of put things together. Having a working knowledge of statistics is great even for the layperson, because we encounter so much data in our daily lives, including consumer gadgets that record and store information.

What drew you to the field?

In every class I teach, the first thing I tell people is that I was born to do statistics. My parents actually met in graduate school for statistics. To be honest, growing up, I didn’t really know they were both statisticians. My mom taught statistics as a professor at the community college, but I didn’t know what my dad did until much later, when I was almost 25 and started taking biostatistics in graduate school. Now I have these really fun conversations with my parents about random things. I call my mom if I have a question about how to teach something, and I call my dad if I have a question about survival analysis, because that’s what he does.

But what I’m most excited about is this: I truly think that statistics and statistical thinking enable people to understand the medical literature and their own research better. Statistics helps people clarify their scientific questions and be specific about what they’re trying to estimate, which allows them to better understand their results.

I hope I’m able to play even a small role in helping demystify the parts of statistics that people generally are intimidated by, or don’t want to learn because they had a bad experience in the past. Once people get past that wall that somehow built up in them that says, “I was never good at math, or “I don’t understand numbers” or “It’s too complicated,” most of the time they find that it’s not that complicated.

In more than a decade teaching biostatistics to mostly early investigators, what stands out?

Two things stand out. The first is that everybody can learn more about statistics. Even people who think they know a lot about statistics end up seeing something they didn’t see before. Little things can allow you to have a deeper understanding and see more connections in the data.

The second thing is that when I started teaching statistics in 2007, you had to convince people that it was important. You don’t have to do that anymore. So much data is publicly available, and in many scenarios, the analytic approaches are so sophisticated that pretty much everybody recognizes the value in having some statistics education. They come wanting the material.

For people teaching statistics, that’s been a wonderful shift. It’s made my life easier, honestly. Now I only have to teach it in a way that helps them see the joy.