Friday links: your sample size is too small, tips for the academic nomad, and more

Also this week: NSF preproposal stats, grad school vs. having a baby, and more. Speaking of babies, we also have a video of a baby T. rex throwing a baseball.

From Meg:

NSF BIO-IOS has joined DEB in having a blog! Among their first posts is one on the IOS preproposal system (two be followed by two more on the same topic). ht: proflikesubstance

And, not to be outdone, DEB updated their blog with info on preproposal stats. I believe Emilio Bruna summed it up best when he said that “evolutionary biologists are 5-6% sadder than the rest of us”.

This article considers the mental health of graduate students at Berkeley, with sobering conclusions. From the article: “The survey findings were telling: nearly half of graduate students reported having an ‘emotional or stress-related problem’ within the previous year and almost 60 percent said they knew another student who had an ‘emotional or stress-related problem that significantly affected their well being and/or academic performance.’ These indicators of distress were notably worse for women and international students. The results also showed underutilization and a lack of awareness of existing mental health services by graduate students.” The article also includes this graphic, which says that 50% of self-reported suicide attempts at Berkeley are made by STEM graduate students. Very scary.

On a much, much lighter note: PhD Comics on vacation vs. stress. Spot on, as usual.

And, finally, another post on flipped learning, again from Casting Out Nines. This one addresses the question of whether flipped learning is just self-teaching – that is, is the instructor really doing anything in a flipped classroom? He makes the analogy to a coach interacting with athletes, which I thought was an interesting one.

From Jeremy:

So you say scientists should report estimates of effect sizes along with confidence intervals, rather than just focusing on rejecting the point null hypothesis of zero effect? Well, there are good arguments for that–but if you’re going to go that route, you’re going to need a bigger boat sample size. Like, maybe 100x bigger. (The linked post is from psychology but the point is general).

Don’t give undergraduates the blanket advice that they shouldn’t go to grad school. For many students, going to graduate school is a good life choice, for reasons having nothing to do with the availability or otherwise of tenure track faculty positions (which many graduate students don’t want anyway). Students should go to grad school with their eyes open, of course, and not under any illusions about the odds of obtaining a tenure track position if that’s their goal. But that’s only one among many bits of information that a prospective grad student needs in order to make an informed decision about whether or not to go to grad school.

Ecologist Amy Parachnowitsch had a baby in grad school. Here she talks about how she made it work.

Early in your academic career, you often have to move around a lot. Tenure, She Wrote has some good suggestions for how to meet people and make friends when you move to a new place.

For those of you who are into structural equation modeling and related approaches to causal inference such as instrumental variables: economist Francis Diebold identifies different “tribes” among users of structural equation models and instrumental variables. Also provides links to some key references on these approaches.

Vaguely relatedly: on regression analysis and the “tyranny of average effects.” I have mixed feelings about this piece. On the one hand, sure, regression can be a vehicle for averaging across interesting/important variation that shouldn’t be averaged across. On the other hand, I worry a lot more about overfitting and the dangers of fishing for statistical significance than the author apparently does.

Ok, this has nothing to do with ecology, but it’s fun: quantifying the most overrated and underrated films of all time. Actually, it’s quantifying the films to which critics and audiences had the most divergent reactions (so “underrated” is operationally defined as “hated by critics, loved by audiences”). Surprised The Sound of Music doesn’t make the list, as it famously was hated by critics but loved by audiences. Includes the R code if you want to play around with the data.

And finally, here’s a video of a dinosaur throwing out the first pitch at a baseball game. Don’t say I never did anything for you.🙂

5 thoughts on “Friday links: your sample size is too small, tips for the academic nomad, and more

    • I think the T. rex scuffed the ball with those teeth, that’s how he got so much movement on the pitch. Future paleontologists will someday estimate which dinosaurs were the best pitchers by analysis of tooth wear patterns on baseballs.

  1. While we’re on the themes of baseball and statistics, there are more than a few sabermatricians and wanna-be same who should read that piece by Dorman. My how they are fond of the concepts of “regression to the mean” and “luck”! They clearly worship at the altar of Bill James. Hint: never try to explain the difference between luck and random variation to those people, it won’t go well.

    Nevertheless, I agree with your take on it. The two issues are flip sides of the same coin.

  2. I like the link that was about the response to criticism of flipping as professors avoiding their job by having students teach themeselves: “In a perfect world, the very thing that students are railing against here – teaching yourself a subject – is actually one of the end goals of “higher” (read: “meta-”) education, namely education about educating oneself.”

    • Yeah. In part, debates about classroom flipping clearly function as sort of proxy wars over alternative educational philosophies.

      Having said that, I do want to learn more about the various approaches to flipping out there and the evidence for their effectiveness. Flipping in some form is something we’re considering for our intro biostats course at Calgary. Before I make the leap, I want to be more confident that it will improve learning (as opposed to merely improving enjoyment). And I want to have a better sense of why it works. I wonder if it works at least in part because it’s a device to force students to spend more time on the material. Instead of just attending lecture and skipping the background reading, they’re forced to spend out of class time doing readings (or watching videos or etc.) in addition to the time they spend in class, thereby increasing the total time they spend on the course.

      I’m intrigued by various versions of the approach, as teachers whom I know and respect have gone this route and (anecdotally) have had good results. And like a lot of profs I’m frustrated that no matter how I tweak my traditional approach to teaching intro biostats I can’t seem to push student learning any higher. But before I make a “macromutation” to my approach (and the associated large one-off investment of prep time), I want to be as confident as I can be that it will pay off in a reasonably big way.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s