Valuing all Kinds of Astronomy Smarts

by Kelle on November 21, 2011

A couple weeks ago, Sean over at Cosmic Variance brought up the topic of “smart” with respect to Steve Jobs and I want to use that post to springboard to a slightly different opinion piece: Astronomy (and Physics) will not be as scientifically productive as it could be and will not make meaningful progress towards equity, diversity, and inclusivity until we change the value system currently in place to assess “smartness.” Building instruments, reducing data, writing code, and the like need to be just as highly valued as doing problem sets and thinking like a theorist. If you can do science, you are smart.

What needs to happen is a complete overhaul for how we asses and reward students who are good at the various things that are crucial for Astronomy to move forward. (Requiring theory-inclined students to take a lab course is not anywhere close to what I’m talking about.) The current hierarchy is bad for Science and bad for scientists. This is a fundamental problem with our field and we need to all be working towards fixing it in whatever ways we can.

What can you do? First and foremost, everyone should be careful with their own language and call out our colleague’s language. In our everyday conversations, formal and informal, we should be very careful about perpetuating the current out-of-wack system.

Second, there are the bigger issues of getting rid of the Physics GRE requirement for admission and changing the qualifying exams for PhD candidacy, especially in joint Physics and Astro departments. But these are tough and need to be dealt with on a department-by-department basis. Although maybe the AAS, in conjunction with the AIP Statistics folks, can make some statement about the lack of correlation between GRE and Qual success and future meaningful contributions to Astro. If you know of some good references for these stats, please post in the comments.

Third, community-wide, the Roman Technology Fellowship program is a huge step in the right direction towards giving instrument builders the early-career support they need and deserve. But this program is decades late. We can’t afford to wait that long to start valuing our astronomers who are coding geniuses. Do you think there should be a fellowship program dedicated to supporting our colleagues who write data-reduction pipelines or other fundamental pieces of code?

What else do you think we can be doing to properly define what it means to be a “smart” Astronomer in our hearts, minds, prizes, admissions criteria, degree-granting requirements, and funding opportunities?

{ 9 comments… read them below or add one }

1 Matt November 21, 2011 at 10:21 am

A good first step would be to avoid the construction “do science.” It is nebulous and unfortunate; often I have seen it used as an excuse to avoid things like building robust models, understanding computational science, providing open access to data, teaching younger students, putting effort into coursework, increasing participation of underrepresented populations, public outreach, and even interdisciplinary collaborations. It’s content-free and usually only used in support of neglecting the other types of “smart” you list.


2 Kelle November 21, 2011 at 10:39 am

hm, I see your point. But maybe what I want to do is re-define what “doing science” means just like I want to re-define what “smart” means. And to make it generally accepted that “doing science” does indeed mean different things to different people and it’s all good. (ok, not all, but a lot of it.) Now, whether or not this is an achievable goal or not is debatable….

3 Eric November 21, 2011 at 11:14 am

I agree with you, Kelle, up to a point. I don’t think we can get rid of the GRE physics entirely though, because we need to realize that the data and programming side of science are useless without the theory, and also vice versa. It’s a symbiotic relationship, and needs to be treated as such. I fully agree that the various branches of our field need to be realized as co-equal … but it is more than that, because realizing they are symbiotic gets back to the fact that each needs the others and cannot exist without the others. Well, ok, except … theory as applied math can exist without data or code, but then we have to be able to be brave and call it math when it’s just that.


4 Kelle November 21, 2011 at 12:51 pm

I do think that we need to get rid of the GRE entirely as an admissions requirement. Yes, the fundamentals are absolutely important, but the GRE doesn’t actually measure the students grasp of the theory nor does it address the “vice-versa” part of the symbiotic relationship. The only thing the Physics GRE measures is how well the student performs on the Physics GRE…which is irrelevant to understanding the theory, being a successful grad student, or a productive scientist. The information that’s being grasped for by the admissions committee should be reflected in the student’s physics course grades. Maybe an even better gauge to consider would be to require at least one recommendation letter from the student’s upper lever Physics professors specifically addressing the student’s preparation and understanding of the fundamentals.

5 Walter November 29, 2011 at 10:10 pm

Not requiring the GRE would be completely unfair for the foreigners. The grading systems are different, and what happens if you come from a completely unknown school?. I am also sure that in every reasonable school they look at more than the GRE scores. You also have to account for cultural differences, it was really hard for me to realize that my statement of purpose was as important as my grades and GRE scores. In my culture we are not use to talk about how awesome we are, that is probably a US thing. 30 minute exam with a random question? At least the GRE is the same for everyone.

6 Ann Onymous November 21, 2011 at 1:17 pm

I got my PhD in a country where the GRE and the quals simply do not exist. To enter a PhD program the panel first made a broad “short list” based on motivation letters, grades, letters of reference, and research experience (as it shows how motivated/interested the student really is). Then we had an interview. We had a couple of exercises to do on a blackboard which were actually more of a discussion with panel members. They did not really care if we did the exercises right (they are not hard anyway to begin with), but they wanted to see how we thought to solve the problem.

There were no quals either, at the end of the master we were ranked according to our grade and a PhD grant was awarded to the n higher ranked students. And at the end of the master professors knew us well enough to know who was most likely to be able to be a good PhD student and who was not. In the end I think the whole thing worked reasonably well. At least I do not see what the GRE or quals would have improved.


7 Eilat November 21, 2011 at 4:41 pm

Regarding the Cosmic Variance article (and the article it was referencing) I know that I would much rather be labeled “genius” than “smart” if given the choice…

I was reminded of another recent NYT article:
I largely disagreed with this author, along the same argument that you make here. He also makes certain blanket assumptions about physics PhDs that don’t represent the reality that I see.

I do know of one fellowship program that does focus on coding and technological development: The NSF OCI Fellowship. I know of at least one astronomer who is a recipient.


8 Polytrope wants a cracker November 26, 2011 at 5:19 pm

I also got my PhD from a country that does not have quals nor uses the GRE in admittance decisions and I agree that their presence would not have improved the system. What did seem to be important was how well a student did on their fourth year research project. One would get a grade for the overall undergraduate degree and depending on the university, the research component would be worth 40 to 70 percent of the overall assessment with the coursework component heavily weighted by the more advanced courses.

If I were to change things in this country, I would give the student applying for astronomy graduate school the option of submitting a substantial research project report instead of the physics GRE. Instead of having quals, the student would be required to give a talk about their thesis say six months to a year since starting. By then the student should be able to talk about the project, what they want to do, why it is important, how it fits in the scheme of things and give a rough timetable. A student who did poorly, may be given a second chance to give the talk.


9 Christina November 29, 2011 at 10:42 pm

I’m a little confused about your last statement. In most universities in the US, I believe it is policy to have both a qualifying written exam AND a preliminary oral exam on your research. I don’t think the author was against the prelim, just the written exams like the qual and the GRE. Correct me if I’m wrong about any particular university, anyone….

Leave a Comment

Previous post:

Next post: