jump to navigation

The Real World: Astronomers’ Edition April 30, 2009

Posted by Sarah in politics, science.
Tags: , , ,
trackback

Compared with other branches of science, you could think there’s not much at stake in real world terms in astronomy research. The amounts of money involved are relatively small**, both in terms of investment and potential gain. Also, galaxies (fortunately!) don’t take us to court if we get their redshifts wrong, and young stars don’t die horrible deaths if they don’t get their shots. Both the community itself and the public see astronomers as this beautiful global brotherhood (a few sisters even these days) working together to unlock the secrets of the Universe. Unfortunately the days where our competition is purely intellectual are long gone, and the modern way of astronomy, as with most fields in science, is a bloody fight for resources, regconition and cash. And the battle grounds are the journals that publish our work.

At the core of science research lies the publishing process: we communicate our work to the rest of the community through articles in journals. These articles are reviewed “anonymously” by 2 or 3 peers who are knowledgeable about the subject but weren’t involved in the work itself. Our merit as scientists is judged by what we publish and how often others cite our papers. Once we’ve got publications under our belt that demonstrate our expertise and our standing in the community, we can go on to apply for our own funding and start up new projects.

So it’s clear that authoring savvy is key to being a successful researcher. Yet young researchers frequently have to find their own way through this treacherous forest, littered with abandoned PhDs and unemployed postdocs. In the world of science, it’s publish or pack your bags.

Here’s the thing: astronomers are normal people. Yes, we’re clever, educated, well-travelled, whatever. But on the whole, we’re no nicer than your average electrician or supermarket cashier. We can be petty, selfish, we bear grudges, we have pet loves and pet hates. A strong competitive streak is almost implicit. Combine that with the fact that there aren’t enough jobs and funds for all of us, and it’s easy to see how things might get messy.

I took this poll a couple of weeks ago (full results coming up!) to see what you readers consider to be “misconduct” in research. Every one of the scenarios I listed was considered bad behaviour by at least a few of you. I didn’t make those things up, they all happen to some extent and remain largely unpunished. Who should be handing out the sanctions anyway? The whole system of using publications as indicators of scientific output essentially rewards “creative publishing”: why publish one excellent comprehensive paper if you can squeeze two mediocre ones out of the same
work? Why remove your name from a paper if you didn’t contribute, but the other authors don’t protest? Maybe you’ll return the favour one day. Is there a mistake in the work? Great, maybe you can write a paper to correct it later!

The peer review process for papers and grant proposals offers some quality control but is itself flawed: in a subject as small as astronomy it’s virtually impossible to find two reviewers who are experts in the field yet have no vested interest in or pre-existing opinions about the work. This could be remedied by double-blind refereeing, although I’d argue that experienced researchers know each other’s methods and style well enough to deduce the authors’ identities.

Using indicators of a scientist’s impact rather than bulk productivity through indicators such as the h-index helps bring the focus back to quality over quantity. But the h-index can lie: a logical way to rack up citations is by having a mistake in your paper, that others will rush to point out.

While ethics issues in astronomy may seem trivial to those working in other subjects, we need to do more to promote good practice in research.  The system of assessing scientific output should actually promote scientific rigour. This won’t suddenly create jobs for everyone who wants one. But it will help secure jobs for the researchers who most deserve them. Too many talented scientists don’t make the cut in research because they didn’t assert themselves early on. We teach students to scrutinise every photon, why not teach them to question their peers? Young astronomers should be told about the lack of jobs and puppies and rainbows, and the importance of sticking up for themselves. They’ll still have a dark forest to navigate, but with a knapsack of knowledge to help them through.

**Dear politicians: that doesn’t mean we won’t fight if you take our money away!

Comments»

1. Rob Ivison - April 30, 2009

wonderful post, darling girl.

i’ve seen people try every trick in the book to raise their h index. i’ve tried a few myself too! but i don’t think anyone has deliberately published a mistake to get cited, have they?

one nutty thing about the h index is the lack of correction for age. another interesting thing about it: when you reach 20 or so you begin thinking “if this paper isn’t good enough to get 20+ citations, i might as well not bother”… usually a good thing, but not so good if people don’t publish their non-detections and others spend gazillions repeating the same experiment.

2. Stuart - May 1, 2009

It appears that I’m doomed.

I took my name off several papers produced in a collaboration because I didn’t feel that I had contributed to those papers specifically. This is bad for me but I think it is the right thing to do. As you say, there is an incentive to add authors to your paper so that the favour will be returned. Perhaps this should be discouraged by reducing the weight of a paper by the number of authors.

Given the many ways the systems can be gamed, I find it difficult to take citation indexes, RAE scores, or league tables any more seriously than a Channel 4 Top 100 programme.

3. astropixie - May 1, 2009

i’ve always wondered why the referee process isnt double-blind. i mean, surely some people could deduce who at least one author is, just like i can usually narrow down the range of potential referees of my papers… but i think its a good first step.

why are journals not using the double-blind method now?

4. He worked my ex with the van » Blog Archive » Quick scan of the net - research electrician - May 1, 2009

[…] https://sarahaskew.wordpress.com/2009/04/30/the-real-world-astronomers-edition/But on the whole, we’re no nicer than your average electrician or supermarket cashier. We can be petty, selfish, we bear grudges, we have pet loves and pet hates. A strong competitive streak is almost implicit. … Too many talented scientists don’t make the cut in research because they didn’t assert themselves early on. We teach students to scrutinise every photon, why not teach them to question their peers? Young astronomers should be told about the lack of jobs and … […]

5. Sarah - May 2, 2009

Stuart – “doomed” may be a strong statement. I always felt proud to be conscientious in my work with things like that but was recently told by someone senior with my best interest at heart that maybe I was a little “too nice”. Huh.

Amanda – I have no idea why journals don’t do double-blind refereeing. There’s this editorial on the Nature page with lots of interesting reader comments. On the one hand there seems to be evidence of bias in single-blind peer review, on the other lots of people who think it wouldn’t make a difference anyway…. Had a quick look at ApJ and they have no mention of it in their editorial and ethis policies. I might try to follow up on that.

Rob – firstly, *blush*. Second, no I would hope no one does that! About the age correction, I have a friend who did calculate some kind of age-normalised citation index for all the faculty of our home institute and basically/weirdly found that they were all pretty much identical; and the one positive outlier was by no means the most influential scientist of the department. Should ask him what it was he calculated…

6. andyxl - May 8, 2009

Rob, Sarah – yes, h-index increases with age; many CVs I have seen over the last couple of years therefore use the y=h/T, where T is the number of career-years since PhD. I list h and y on my own web page

Also, many people look for normalised citations : you divide citations of paper by the number of authors. (ADS will give you this.) Thats still not right, but better. A selection panel will often start by counting papers, and then citations, but then to really sort out the sheep from the goats, they look to see if someone has a few single author papers or papers with only two or three authors.

I actually think the citation index is healthier than the way it was at the beginning of my career, when sheer number of papers was what counted most. At least now there is some pressure to produce work that somebody actually reads.

7. Dark Matter Damping « The e-Astronomer - May 8, 2009

[…] Matter Damping There is an interesting discussion over at Sarah’s blog, about astronomers being driven by ambition and greed just like everybody else. I remember this […]

8. Stuart - May 18, 2009

After reading Andy’s comment, I’m even more doomed 😉 That is the trouble of working in very large collaborations where every paper has many authors. Lots of work does not equal a good citation index or career prospects.


Leave a reply to Sarah Cancel reply