I previously referenced Donald Sharpe’s idea of a statistics maven: people with one foot in a science field and one foot in statistics, who frequently act as a conduit for new quantitative innovations. Afterward I had an email exchange with someone who wanted to know how to become a maven, and I had to pass along the news that he probably already was. As a public service to others with similar concerns, I thought I should gather together the most probable symptoms (pending a comprehensive program of construct validation research, of course). Here at the top ten signs that you are a statistics maven:
10. You have installed R packages just to see what they do.
9. Your biggest regret from undergrad is a tossup between that person you never asked out and not taking more math.
8. You call the statistics you learned in grad school “frequentist statistics” and not just “statistics.”
7. People who are not quantitative psychologists call you a quantitative psychologist.
6. But you would be embarrassed if an actual quantitative psychologist overheard them.
5. You have a dead-tree subscription to Psychological Methods delivered to home so you can read it in bed.
4. You are thanked in the acknowledgements sections of your entire cohort’s dissertations.
3. You have a Keep Calm and Read Meehl poster in your office.
2. You once ran an entire study just to have the right kind of data for an analysis you wanted to try.
1. You have strong opinions about bar graphs and you are not afraid to share them.
(p.s. Shoutout to Aaron Weidman for #5.)
The Hardest Science 
Nice list! Leona Aiken and Steve West talk about twofers — researchers who have active substantive as well as quantitative research programs. Often it takes an active substantive research program to spark the translation or adaptation of novel quantitative tools. Having a burning research question can be a powerful motivation.
My biggest regret as an undergrad was not knowing what math and statistics courses to take. I am not sure what is special about bar graphs. I have strong opinions about most of the figures and graphs I see…
For me the rubicon was uninstalling SPSS and not looking back. I now use R in undergraduate and graduate classes. Teach to the future, not to the past. Nothing like some data to motivate students. See Trends in Computing Software
That’s a great point about “twofers.” I think you can get there two ways: some people start out in a substantive field and become mavens, other people start out as statisticians or quantitative psychologists and latch onto a problem or phenomenon that they really want to explain.
Coincidentally, while preparing for my research methods class today I came across this very relevant quote from Edwin Boring (1919): “Statistical ability, divorced from a scientific intimacy with the fundamental observations, leads nowhere.” So Boring was a fan of twofers too!
Boring’s quote is spot on and applicable in many circumstances. I recall a colleague in graduate school who was a former engineering student. She had taken linear algebra previously and that was a boring (no pun intended) intellectual exercise that she saw little point in. However, later when she took multivariate statistics that its value clicked for her and it all came together. Some serious aha! moments and excitement. It was then that she really learned the material.
My own experience with structural equation modeling, multilevel modeling, etc., has been that I really learn when I have my own data and questions to apply those modeling approaches to. There is a reason quantitative psychology exists, as does biostatistics, epidemiology, econometrics, etc. Ken Bollen’s SEM course in years past starts with a single hour where he develops the full linear model using linear algebra. In other words, the full multiple regression framework in a single hour. All theory. Fully complete. However, to apply that without context or additional training would be nearly impossible if that was your only exposure to multiple regression. We need a context to ground it and then learn how to interpret and extract knowledge from the model.