The High Society

Nathan T. Carter, University of Georgia1

I cannot believe I am saying this, but I will be taking over Paul Muchinsky’s much-loved (and probably much-hated) role as the I-O RoastMaster General. Those who remember Paul’s column will remember that it offered a humorous—and often biting—take on issues in the field. Oddly enough, in 2006 I wrote to him while I was a graduate student in the master’s program at Western Kentucky University² to tell him that I really enjoyed his articles and told him why I enjoyed a particular one. In his gracious response, he said to me3: “I caution you to be discrete in sharing such opinions openly as a graduate student. In the course of writing my articles I have ‘offended’ many leading figures of the establishment, who would welcome my departure from TIP. At this stage of my career I can afford such professional rebuke. You cannot. After you graduate I would be most willing to pass the torch to you if you care to take up the cause. I guarantee you will not be voted ‘most popular’ in SIOP if you choose to do so” (P. Muchinsky, personal communication, February 8, 2006).

So, it is with a great deal of hesitation that I submit to you my first attempt at being funny about I-O psychology. In reading, try not to take me—or yourself—too seriously. After all, we are only tiny specks on this tiny planet doing things that will not be remembered when our sphere erupts into a ball of fire. Also, do remain aware that often I am writing about things where I am just as much the butt of the joke as you.

I have missed Paul’s “High Society” column greatly and am a bit afraid to screw it up. But, here goes nothing. As was Paul’s tradition, fan mail or hate mail can be sent to pmmuchin@uncg.edu (jk, jk… send to carternt@yahoo.com³).

Is Your Data “Big” Enough?

This year’s SIOP conference in Chicago, and the previous few years, have been replete with hashtags (known previously as “pound signs”) and memes touting the coming of “big data.” However, it leaves some of us with an insecurity complex, asking shyly: Is my data “big” enough? Here is a nice little checklist for you to use to decide whether you measure up! Warning: Double-, triple-, and quadruple-barreled items ahead!

1. Does writing out your N require at least one comma, or preferably more? _____
2. Have you sufficiently confused your colleagues on why your study should have been done in the first place? _____
3. Can safely say you are explaining 85% of your variance without ever bringing up terms like “overfitting” or “saturated?” _____
4. Does your model use at least 20 variables? _____
5. Do you get to say fancy words like “machine” or “genetic” or “AI” when describing your model? _____
6. Do you get to say things that sound like a horror movie about a nightmarish logging camp, like “random forests” or “regression trees?” ____
7. Do you get to use a model with a fancy diagram and pretend it’s not just basically stepwise regression? ____
8. Did you spy on people, or otherwise collect data without their knowledge and/or consent? _____
9. Can you answer questions about your project with an air of superiority because the         questioner “doesn’t understand math?” ____
10. Is your idea of a fun Saturday night getting lost in an R Stack Exchange? ____
11. Did you get a little upset that I said “IS your data big enough” instead of “ARE your data big enough?” ____
12. Have you not laughed once at this list? ____

If you checked at least two of these, then rest assured, your data ARE probably big enough! Now you can go forth to laugh at those with only puny datasets, a simple set of predictors, and a AUC or R² that is .001 less than yours—especially those qualitative researchers! #amirite?! You are a true “data scientist,” whatever that means.

If you did NOT check enough boxes, don’t worry! You can talk about big data without knowing about it! Just log on to Twitter! #bigdata #dataanalytics

Notes

¹The opinions expressed in this column are that of Nathan Carter, and not his colleagues – whom refer to him as “OpinioNate” (K. Shockley, personal communication, circa 2017) – nor his graduate students, who are much better people than he.

² This would be the FIRST Bowling Green Massacre I would endure. The second would take place in pursuing my PhD at Bowling Green State University (Bowling Green, Ohio). This article is dedicated to the victims of those fictional—er, alternate-factual—tragedies.

³ Found in the depths of my surprisingly still-active Yahoo! e-mail account, which now is home to a variety of enticing offers from several loan agencies and your thoughts about my “High Society” articles.