Friday links: visualizing sampling error, Ben Bolker vs. statistical machismo, why be wrong, and more

Also this week: big news about peer review at Am Nat, allometry vs. monsters, zombie ideas ideas about zombies, administrators > faculty, pointless (?) scientific prizes, and more. Also, the delicious, starchy future of crowdfunding. 🙂

From Brian:

On our ever popular theme of not needing to work 80 hours a week, a post following up a theme that emerged in the comments on our post – its not how long you work, its how smart you work.

And time is quickly running out to sign up for the Gordon Conference on Unifying Ecology Across Scales – its my favorite conference (this will be my 6th straight attendance). The Gordon format really encourages great interactions and the attendees at this one are from seemingly very diverse fields (stoichiometry, macroecology, physiology) but are genuinely trying to build bridges between the fields. If your looking for that last great conference to fill up your summer, check it out.

And lastly, a great post on why it is important to be wrong, how it advances the field, and why humans never admit it.

From Meg:

Big news from AmNat:

Very interesting! I will be really interested in seeing how this goes. And here’s a related article that addresses some of the commonly cited concerns about double-blind review (including that reviewers will easily guess author’s identities). (subscription required)

Here’s a piece from the Chronicle of Higher Education on finding meaning after leaving academia. It focuses on the downsides of academia’s focus on doing what you love and being able to make your own schedule, and talks about the things she enjoys in her new job at a PR agency.

A new study found that having students take notes by hand led to greater retention of material as compared to those who took notes on laptops. Those laptops were disconnected from the internet, so presumably the effect in the average classroom would be even stronger.

I was recently discussing this paper describing the Dead Grandmother Hypothesis with a colleague, and found this version of it online. It’s a classic. It’s a deadpan take on a very important pattern – “a student’s grandmother is far more likely to die suddenly just before the student takes an exam, than at any other time of year” – and proposes hypotheses for why this might happen and how to avoid the problem.

How to use allometry to get your child to overcome his/her fear of monsters. (ht: Morgan Ernest)

And, at the risk of killing everyone’s productivity, bear & salmon cam is back!

From Jeremy:

Here are Ben Bolker’s slides from his recent plenary talk on statistical machismo at the International Statistical Ecology Conference.

The NY Times Upshot blog on big data, prediction, and explanation, using a silly but fun example (predicting which tweets will get retweeted). Could be good fodder for undergrad stats courses. (ht Marginal Revolution)

Speaking of The Upshot and fodder for undergrad stats courses, this post includes a great animated graph illustrating sampling error. And it’s an important real world example (the US government’s monthly jobs report). A terrific way to dramatize to students our tendency to see a “signal” where there’s only noise. Intro stats instructors will be tripping over themselves to incorporate this into their classes. (ht Andrew Gelman)

Administrators are taking over academia. US higher education enrollment increased 150% from 1975-2013. Meanwhile, full time non-faculty professionals rose 369% and full-time executives rose 141%. Full time tenure track faculty rose 23% (part-time and non-tenure track faculty rose much more). Over roughly the same period, pay raises for faculty were well below those for senior administrators.

Here’s a good interview with my former Calgary colleague, ecologist Andrea Kirkwood, about her experiences as a woman in science in Canada. The first in what will apparently be a series of interviews on this topic.

Using zombies and vampires to teach intro economics. Perhaps the same could be done for ecology? Perhaps with Bigfoot, the Loch Ness Monster, etc. thrown in? (ht Marginal Revolution)

What’s the point of scientific prizes? Well, the point of prizes for solving a particular problem is obvious–they create an incentive for people to try to solve that problem. But what about other prizes?

Paleoecologist and blogger Jacquelyn Gill visits the La Brea tarpits.

An ingenious way to identify books that are bought, but not read all the way through. Wonder what you’d find if you applied this method to academic books?

And finally, here’s the future of crowdfunding. It involves mayo. Maybe. 🙂

11 thoughts on “Friday links: visualizing sampling error, Ben Bolker vs. statistical machismo, why be wrong, and more

  1. Brian, is the Gordon Conference on Unifying Ecology Across Scales always in Maine? I’ve got a newborn this year, but will put it on my 2015 calendar if so.

  2. According to my calculations, 80 hrs on 7 days of work per week means that the scientist is working ~ 11.5 hrs a day. Using the immortal words of John McEnroe: “You cannot be serious”. Not possible to do that for long periods of time. Then, it is quite difficult to tease apart work and “I am not exactly working, but I am still thinking about that grant” in Science, but 80 hours a week of real work for maybe more than a few weeks, no way in my opinion. On the other hand, is the lumberjack sharpening the blade working or not?

  3. Thanks for the signal boost, but wanted to make a quick correction: I am a paleoecologist, not a paleontologist. I do some neo-ecology and biogeography, too. A minor quibble, but an important one.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.