Wednesday, October 19, 2016

What is the best possible grade?

What is the best possible grade a student can possibly get in a course? The answer seems to be obvious: it is an "A", or maybe "A+", right?

But imagine a student with 4.0 GPA. Would not it mean that this student did not challenge themselves enough? That they took courses that were too simple for them, like CalcI when they ought to have taken CalcII? Would not it mean that they never struggled with the material? Arguably, if you are smart, the easiest way to get a 4.0 GPA is to always pick courses a notch lower than your current level. Then you will surely shine, like a superhero among normals.

Which is curious because from behavioral studies in animals and humans we know that we learn best when we fail in about 50% of the cases. It maximizes information transfer, and so maximizes learning. It is surely very uncomfortable, even humiliating, and it would surely make you question your place in science if you fail on every other attempt, but curiously, all other aspects being equal, that's when you would have learned best.

I obviously don't suggest that we make students fail in every other assignment (it's not middle ages anymore, and we just don't have the mental and emotional preparedness for it), but to learn they should fail at least every now and then. Which typically, for an honest and hard-working student, corresponds to a grade of A minus. Maybe even B plus.

Does not it suggest that grades are useless though?

Anecdotally, it seems to be the case. When I grade objectively, on a rubric with fixed thresholds, I see that non-specialists (students of different majors) and prodigies (students who take senior-level classes in their sophomore year) typically get about half a grade lower than similarly hard-working majors and seniors respectively. But is not it silly? They surely learn more, and in a way the very fact that they take harder courses than they are expected to speaks of their resilience, enthusiasm, and brilliance. But it's not reflected in the grade (although I can comment on it in a recommendation letter).

And if it is silly, what should I do? Just give all sophomores a boost of half a grade? This would not seem fair. Grading on "effort"? I don't think it is possible to grade the effort objectively; some people would just suffer silently, and also it would send a wrong message to students. I have no good solution here, but sure it is an interesting question.

And at least at the personal level I can tell that if I needed to hire an assistant, I would probably always prefer an A minus student to a straight-A student, as A minuses just seem to be more persistent and / or adventurous.

Friday, September 16, 2016

Research / teaching balance

This semester I meant to keep Fridays (one day a week I don't teach) for research exclusively, and resist the urge to catch up with teaching prep work on Fridays. But lo and behold, it took me exactly 3 weeks to relapse. The first week went well: I was writing a research paper. The second week was fine as well, but I had to come in on Saturday for a few hours to catch up with other work. But the third week came, and I'm defeated, at least temporarily. I need to rework a lecture that failed last year (the one about normal distribution - hard topic to conceptualize), write some lab assignments, and so on.

Don't get me wrong, it obviously get easier with time: it seems that had I stuck exactly to my previous year lesson plans, I could have saved about half a day, maybe a day worth of time every week. But I am trying to rework both courses, to make them better: to introduce more group work and primary literature in my intro, and to move the emphasis away from probability theory and towards data presentation in my biostats class. And it means prep work, and weekly firefighting.

Now, here's an interesting blog entry (from 2011, but relevant and very well written) about what it takes to get a tenure in a major research university:

http://blogs.discovermagazine.com/cosmicvariance/2011/03/30/how-to-get-tenure-at-a-major-research-university/#.V9lN3vkrLIU

It may seem like a non-sequitur, but actually it's intimately related to the existential threat of research Fridays. The question is: how should I balance research and teaching, in an ideal world? Is research only for vacations and weekends, or is it possible to do it during the week? And also, should it be possible, from the administration point of view, thinking in their shoes? Should we (the people, the faculty) encourage a more even split between teaching and research? We are a teaching college, but an aspiring one: we are a SLAC, as in "Small Liberal Arts College", but we want to become a SLAC as in "Selective Liberal Arts". We are trying to boost our profile, and it means that while teaching takes most of our time and effort, surprisingly, it is research that mostly comes up during tenure evaluations. I mean, if you are bad at teaching, you are fired. But once you are good, or at least decent, everybody just shift to weighing and assessing your research. Is it sustainable? But is it ambitious enough? Are you stretching too thin? Or maybe too narrow? Too many collaborations? Too few? Too little work with students? Too much student work? There are many dimensions to assess, and many considerations to balance.

In a way, it came to me as a surprise that our tenure discussions are actually not that far from that in a major research university, at least in spirit. Granted, we can collaborate with our former advisers, we can be third authors, and the expectations for productivity are much lower; perhaps as much as 5-10 lower (depending on what weight you ascribe to collaborative papers). But the criteria themselves become more and more research-oriented.

There are aspects of this shift that are worrying. For example, I don't quite like the shyness with which the older tenured folks refuse to set clear criteria for the publication record. The reasons for this shyness are actually good and valid: in a small college the same group of people has to discuss publication records of a computer scientist (all conferences), theoretical physicist (all arxive), molecular biologists (typical paper has 30 pages and 12 figures), and synthetic chemist (typical paper has 2 pages and 2 figures). It's hard to come up with clear criteria when every single case is so unique. Yet it is a bit annoying, as in theory this flexibility can be used both to save a case, and to sink it.

But at the same time, there are upsides here as well, and not just because I personally like research. Perhaps the most curious one is that with research emphasized so strongly, our tenure goals are now not that far, in terms of CV building, from job search goals for a person who suddenly decides to leave for another institution. So in way now we can try to just "be successful" as potential job candidates. If we are successful, we'll probably get tenure as a collateral, but if for some reason we won't, we'll still have some decent chances of finding another job. It feels that in a teaching-only college there would be a stronger fork here, a bigger difference between tenure goals and job search goals. In our case it's not that bad, which makes the situation less risky.

And in practice it means: publications, publications, and some more publications. No popular books no textbooks, minimal service. Teaching should be good, but pedagogy related publications, conferences, grants and projects are more important, as they are more objective and more visible to outsiders.

That's the plan.

Wednesday, September 14, 2016

Thoughts about tenure evaluations


It is the season of pre-tenure and tenure evaluations in my college, and all faculty are encouraged to write "testimonies" for their colleagues who are up to evaluation. These testimonies are supposed to be used for the tenure and reappointment discussions, one way or another. I wrote a few as well, and now I'm wondering whether I should also send them directly to the people in question; those who are about to be evaluated.

There are some strong arguments in favor of sharing the evaluations openly and directly. Most importantly, my evaluations are actually very positive, and I think that we humans generally don't get nearly enough praise in life. It's all competition, benchmarking and impostor syndrome all the time. So maybe it would be nice for them to read something good about their teaching and research, for a change. Especially in this relatively stressful time when the meetings are about to happen that will (supposedly) decide their fate for nearest few years, and that they won't be able to attend. Also arguably it is useful to receive some real open feedback every now and then. Of course, they will receive the "evaluator's report" a few months later, but most probably not a single row of my original testimony will be quoted in this final report, or maybe half a sentence at most. Supposedly, testimonies are somehow "integrated" and "summarized" in the evaluation document by the evaluator, but not more than that.

On the other hand, one could argue that if you send nice letters directly to people, you forever wave a possibility of writing a negative letter. Or actually not writing a letter when you are torn or indifferent. Because you would not probably share a negative letter, yet if you are known as a "sharer", but don't share anything next time, the person would infer that the letter was probably negative. That's the whole reason people use secret ballot voting to begin with. Also, I am kind of concerned that some of my praise may be not to the point, as I don't quite understand some of the aspects of other people's scholar work. What if I'm praising them for things that are not actually relevant in their own eyes? Who knows, different disciplines are different... At a risk of sounding paranoid, is it possible to inadvertently "damn by praise" - not even because it is faint, but because it is somehow idiosyncratically not to the point?

For now I don't quite know what to do. Maybe I'll toss a coin really. I really like the idea of transparency and clarity, but at the same time there is a good reason tenure votes are always done by a secret ballot. I am not sure there is an ideal solution, but I am wondering what an optimal solution could be.

Tuesday, August 23, 2016

Advice to computational postdocs: apply to math and CS jobs

If you are a computational neuroscientist, and would like to teach, consider marketing yourself not only to neuro and psych departments, but to math and computer science as well.

Why? Because I'm looking at our place, and how we totally struggle to get good candidates in both computer science and applied math. I guess the cynical way to put it is that both fields are so incredibly useful these days that any person who is skillful in them, and who can also teach (which implies good management and interpersonal skills), can probably find jobs in the industry with much higher salaries. And with similar levels of enjoyment. Either way, the fact seems to be that applied math and computer science are understaffed, despite the high demand from the students. During job searches, for each decent job application we get in computer science, we get 10 applications in psychology, even when the research topics are actually quite comparable.

In practice it means that a good postdoc or grad student in computational neuroscience can at least triple their chances of landing a great TT job if they create two more sets of application documents: one tailored for applied math jobs, and another - for computer science. And while it may seem scary, it's actually pretty easy to do.

Let's give it a close look. In a SLAC, faculty typically teach 4 types of courses:
  1. Intro courses (something every major needs to take in lower college)
  2. Core courses (something every major needs to take in upper college)
  3. Fancy stuff (electives of various kinds)
  4. Crazy fun (like math for lit majors, or computer science for historians)
Basically, if you apply to math or CS dept as a neuroscientist, you need to make them know that you can teach all types of courses from this list, plus establish some "street credibility", so to say. Type (1) is never a problem: it would be "calculus I, II" in math (every computational person can do it), or intro to object-oriented programming in CS. You can do it. Type (3) is also easy: it would be what you do for a living, as a researcher, or maybe some one-two fields nearby; something like modeling, numerical computation, big data analysis, dynamical systems, machine learning, methods in Bayesian statistics, or something like that.

Which means that basically you just need to invent one crazy fun course (which should be relatively easy; just draw inspiration from your hobbies and side interests), and to convince the committee that you can teach core courses: something like linear algebra, differential equations or vector calculus in math; or data structures, algorithms, and discrete math in CS. That is a bit harder, but once you cover some of these courses (one may be enough), you are fine!

Now just reword your research statements accordingly, to compensate for the relative lack of "appropriate" education in these fields, and you are golden. You can apply to 3 times more positions than a straight neuro person would apply, and you would compete in a market with a much higher demand and lower supply, boosting your success rates.

Monday, August 22, 2016

Best way to create custom color palettes for visualization

Colorbrewer is awesome, but quite restrictive. After browsing the web for some time, here's the best too I found, with tools to create very nice-looking, yet usable and informative custom color scales in any aesthetics you want. It's called the "chroma scale helper":
http://gka.github.io/palettes/#colors=lightyellow,gray,teal,indigo|steps=5|bez=1|coL=1

Here's the description of how it works (it's very clever, and worth the read on its own, even if you never use the actual scale helper"
https://vis4.net/blog/posts/mastering-multi-hued-color-scales/

Here's a table of color names it uses (you may have to browse for the color you like, but it's very doable)
http://cng.seas.rochester.edu/CNG/docs/x11color.html

And finally, the source of these links (with some more advice on the matter of colors):
http://lisacharlotterost.github.io/2016/04/22/Colors-for-DataVis/

Wednesday, June 29, 2016

Teaching scientific critique

A very nice text on teaching how to critique scientific literature:

Main idea of the text: too many teaching assignments we use essentially encourage students to "bullshit"; to generate some plausible-looking, but empty rambling about the topic, or post-hoc interpretations of their results. It's hard to grade, it does not teach students real scientific thinking, it's just generally bad. The author then gives some good pieces of advice about how not to fall into this trap:
  1. Be more specific: offer a critique yourself, evaluate the paper, and, potentially, vindicate it. Send a clear message that our goal is not to find a flaw, but to be able to asses whether there's a flaw in the study. 
  2. Clearly separate critique of methods from critique of results. I fully agree here; students tend to conflate hypothesis-building, experiment design, and results interpretation; they somehow combine it all into one horrible bezoar ball in their heads, and then try to describe it all at once. For example, they tend to perceive negative results as failed studies. Being very clear about what aspects of the study we are actually trying to critique should help here.
  3. My favorite: instead of discussing papers, talk about pop science (post-press release articles that appear in the press). I think that's the most productive idea of all.

Wednesday, May 25, 2016

The Slow Professor (book review)

"The Slow Professor" by Maggie Berg and Barbara Seeber is a manifesto-like book about some important problems in modern academia. It was published a few weeks ago (I actually pre-ordered it), and if you have anything to do with academia, I do totally recommend that you read it. It's also rather short, which means that you can read it quickly (I hoped it would be a bit longer). Let me summarize what I liked and what I did not like about it in two lists below:

What I liked about "The Slow Professor":

  • It tackles one of the most important problems in modern academia: everybody are perpetually busy (applying for grants, publishing, working on committees), and nobody has time to think. People are ashamed to think (it does not feel like working); moreover, people are ashamed to read (in modern culture it does not feel like working either). And that's bad. The chapter about "what is bad" is the most relatable and passionate part of the book; the description is perfect, and to the point.
  • The book makes you think; it is definitely thought-provoking. It is also written a bit like a manifesto, so I felt energized after reading it. I wanted to change something! This feeling wears off in a few days, as it usually happens with manifestos, but it is definitely not a depressing book, which is really a feat for a book that in its core describes some important problems. Well done!
  • It is short, so you can read it quickly.
  • It actually offers some meaningful solutions, or at least points at some possible directions where these solutions may be.
  • It offers a nice slogan ("The slow professor" is a nice slogan!).

What I didn't like:

  • It is woefully short, and the solutions it offers are very limited. I guess it's the inevitable tradeoff, and I'd really rather read a short passionate book now, than a long thoughtful book in five years. It may be too late in five years! But it is really more of a manifesto than a guide; a pamphlet that names the issues and sets the goals. It is not a self-help book that would guide you through a series of exercises. You need to find the solution yourself. It invites you to be a part of a community though, which is really nice!
  • The book is relatively full of really bad neuroscience and psychology. It mentions serotonin, dopamine, oxytocin and neural plasticity - all incorrectly, and in ways that are totally irrelevant for the topic and the message of the book. As a neuroscientist, I don't usually read pop-science pieces about the brain, because it hurts, so I was not quite aware that the pop-science surrounding the mystery of the brain got that bad over the years. When you buy this book, please just ignore everything it says about how neuroscience "proves" which teaching and research methods work, and which don't. Just skip it without reading, it's all a bunch of nonsense. Also it cites a bunch of retracted and non-replicated (but famous) studies in psychology, so take all psychological claims with a spoonful of salt.
  • Finally, I find it annoying that when professional academics try to write a popular book they still default to academese, or at least half-academese. If feels that every sentence in this book is half-way between the world of the living and the world of the dead; even though sentences are readable and clear, they still have a strong smell of dusty, deathly, cryptic, mummified academese. It feels that the authors fought this tendency to the end, but still could not quite shake of the suffocating embrace of academic writing.
A great book though; I really recommend it. After reading the first half I felt that I need to buy a copy for every person in my department. After finishing it I felt a bit less passionate, but still told everybody about it and encouraged them to buy it. It's a very worthy read!

And also, on a personal note, I am so happy that teaching colleges, and Bard in particular, and maybe even Biology program in particular, are in a relatively good shape, as far as the problems described in the "Slow Professor" go. We actually do talk to each other, and it feels like we have a bit of time to think. We have teaching and grading in place of grant writing, so there is still a monster of "busyness" to fight, but it seems that we are actually fighting this battle already; driven by a slightly different motivation (trying to become better teachers), but still fighting. And there is definitely lots of space for improvement!

If there were a pin with a snail (from the cover), I'd totally buy it. The "slow professorial movement" is something I'd love to belong to!