The relationships among fame, impact and research quality


I just read a particularly interesting post by Dr. Becca about life about halfway through the tenure track that got me thinking, particularly one section:

I feel like most of my job right now is to be famous… What I mean by this is that I’m pretty sure a lot of my future success is going to depend on whether people remember my name when they review my grant applications and manuscripts…

What determines your success? How famous you are.

Most famous scientists have a history of excellent research with high impact. And most researchers with a history of excellent research with high impact are famous. (Fame, that is, among scientists.) However, the r2 on this relationship is well below 1. What explains the variance?

What are the factors that makes you more famous, or less famous, than would be merited by your research quality?

Is the impact of your research — how much it influences the work others — closely tied to your fame or are there people who have a high impact but not well recognized – or people who are quite famous but don’t have much impact?

Fame path diagram

A working hypothesis for the relationships among aspects of a scientist’s research program

I posit the figure above only as a suggestion, a working hypothesis that I’m not wholly wedded to. It’s a good template for discussion.

The ceiling of the impact of your research is dictated by how famous you are. Your impact could be (very) crudely measured using impact factor, or by an h-score or some other measure of citations. How much of a difference you make. You might get cited a few times if nobody has heard of you, but essentially you need to be known for your work to make a splash. You can only make a difference if people know who you are, which is exactly the point that Dr. Becca made. Your job, if it is to make scientific progress, is to become famous. Because you can only make a difference if you’re famous.

If asked to name two huge advances in biology from mid-1800s, most of us would pick the same things. One came from a person working in obscurity and another by one who was, among scientists of the day, mighty famous and was in regular communication with other famous scientists. Darwin’s scientific impact was immediate. Mendel’s finding required the fame of Hugo de Vries to create a scientific impact more than thirty years later.

There are many things that contribute to fame. One of these is research quality, but also the institution you came from, your academic pedigree, attractiveness, personality, and also your ethnicity and gender can have an effect.

What’s another thing plays a key role in facilitating, or limiting, your fame? The institution where you work. If you’re not based out of a research institution, there is a hard cap on how famous that you’re allowed to become as a research scientist. However, if you’re at a teaching institution, the school doesn’t really want you to be a research scientist of any fame, anyway. Fame isn’t part of the evaluation process for tenure, and you could be entirely unknown off campus and this shouldn’t (necessarily) negatively reflect your tenure bid. This would be fatal at a research institution, where you’re expected to establish a visible profile in the research community.

Our jobs at teaching campuses do not expect us to be famous and do not require it. This might be a defining contrast between a teaching campus and a research campus. However, there are lots of us in teaching institutions that not only are doing consequential research, but also want this research to have as much impact as it possibly can. However, based on the name of the institution found on our nametag when we present at conferences, this becomes very difficult.

There’s a positive feedback loop connecting one’s pedigree, social network, publication history, favorable reviews of grants and proposals, funding, talent of collaborators and fame. They’re all connected to one another. And if you’re at a teaching campus, you’re at a strategic disadvantage because those positive feedback loops don’t work as tightly.

Leveraging your pedigree, papers, and collaborations is harder to do, because of unacknowledged biases against teaching campuses in the research community. You can’t be famous above a certain level, because those at research institutions assume that you aren’t working at one because you can’t get a job at one. If you’re doing research from a teaching institution, that means that you haven’t had enough success to work at a teaching institution. So the thinking goes. Even in the incredibly tight job market, that line of thinking still prevails. You’re skeptical? Pull up a few journals and look at the mastheads, to find the institutions of the editorial board members and the subject editors.

So, unlike Dr. Becca and those at research institutions, my job isn’t to become famous. Even if I was famous, nobody on campus would even be aware of it anyway. However, if I have ambitions for my research to make a difference, then I need to become famous. This fame is required to activate the positive feedbacks among friendly reviews, funding, invitations, collaborations, and so on.

15 thoughts on “The relationships among fame, impact and research quality

  1. REally interesting post, Terry, on an issue that I have been thinking about a lot lately. Two additional thoughts:

    First, at some teaching institutions the balance towards “fame” (however narrowly defined) changes a bit as you move through tenure and towards promotion to full professor. At my school (and others I know of), promotion is based not so much on your status within the institution, though of course you still have to be doing well in the classroom. Unlike the tenure decision at a teaching institution, the evaluation is based more on the assessment of outside colleagues from your discipline. What is your impact in the discipline as a scholar? For us scientists, this mostly means research, but it can also entail professional service like participation in professional societies, advisory boards, editorial boards, funding panels, or educational consulting. These are not the same type of fame that you are talking about, but they do come into play.

    Second, I think your hypothesis gets it backwards. The arrow from research quality should go towards impact, because that is how you actually get “fame.” How could high research quality lead to fame without garnering impact along the way? The question for me is then: why is it that some high quality research produces very little fame? And even among researchers that achieve decent impact, there is relatively little fame in the form of name recognition? I think that these are the transitions in which biases against teaching institutions come into play. Especially with collaborative work, I’m sure there is a tendency for readers (i.e., other scientists) to discount author contributions based on their institution.

    I would hypothesize that there are R1-based authors with *lower* impact factors than active researchers at teaching institutions (e.g., you!) who get more invitations for talks, review panels, editorial boards, and the other tokens of name-recognition type “fame.” So I guess I am saying that the bias against teaching institutions disrupts the arrow from impact to fame, rather than disrupting some direct link between research quality and fame.

  2. I was being particularly cynical by saying that research impact has to go through fame. You could write the most awesome paper in the world, but it will have no impact unless you have the fame to make it have any impact. This awesome paper might not end up in a good journal because you lack the famous pedigree, or it might be unread because nobody knows who you are, and it doesn’t get cited because people don’t feel as if they need to cite you.

    I guess it’s possible that you get famous from your impact, to some extent. There are a lot more arrows, entering from factors not included in the model. And there is an arrow between research quality and impact, but I think it’s a little one. Overly cynical?

    Right, high quality research that leads to fame, brings impact along for the ride. But is the impact a cause or a result of being famous?

    When I work with undergrads, and new grad students, they don’t understand that the first thing that a scientist sees in a paper in one’s field is the authors. That’s the foremost feature that’s used when evaluating the paper. To students, who did the work is almost secondary, but of course for us, it’s the central feature, and how your name is perceived is huge.

    I think you’re right, if I wasn’t at Podunk State, I would have been invited to serve on an NSF panel by now, and I’d probably be invited to more editorial boards. And there are people who are on panels and some editorial boards who don’t have a CV as strong as mine (not that mine is super kick-ass or anything, but it definitely shows I’m legit.)

    That’s a generous interpretation, though, that at teaching campuses our work does have an impact, but we’re just not as famous for it. Hmm. I’ll have to mull that one over.

  3. Great point about how practicing scientists use author as a key evaluation tool for a paper, and the difference with students.

    In some ways, I think I am trying to resist the same cynicism that you are indulging. I think there is a genuinely meritocratic aspect of scientific impact, though it is far from perfect. After all, we are social apes with a long history of hierarchical status structures. It takes a lot of vigilance and effort to work against that gradient.

  4. Agreed. It’s a lot easier to diagnose a problem than do something about it. It is a partial meritocracy, and it’s up to us to actively keep it that way. I gotta love the positive attitude of you and other folks on the site.

  5. Very nice post. Funny how you refer to Hugo de Vries. I am Dutch and some German colleagues asked me to name some famous Dutch biologists. So of course I mentioned De Vries, and Niko Tinbergen, and they knew neither!!! Not to argue that “thou shalt know famous sicentists”, but I personally perceive this as a terrible omission, in a nationalistic sort of way🙂

  6. I think the point about the author names associated with your work is a very good one as I’ve seen in be very true in academia, but also in other fields. For example My professional muscian friend (now PhD) takes time to mention who he studied under, who he played with. His audio files, his conducting cred do “count”, but it’s his associations that bolster his credentials. Does his work speak for itself – sure. But if that was all that mattered, why bother to mention his educational and professional relationships? What about the relationships you are forming with students? Will interest in your work grow if one of your students suddenly comes upon a newly popular avenue of study? Will your institution suddenly become more popular because some undergraduate’s career path brought them to the White House? Will people look back to see your name in some of their early work, or would they mention their mentor in in explaining the genesis of their thought process which then makes people look at your work in a different light? Which brings me to the second point…. fame, even in academic circles may come from fortuitous, or serendipidous if not perhaps sometimes tragic and popular and relevant, outside events. Not that one should run their career or research interests based upon this equivalent of the lottery, but often it is the person most prepared at a given moment that can use such opportunities to create their own “luck”, Often, like the miracle of compound interest, all these things that happen on a much more modest scale will likely accrue into some sort of overall achievement, impact, fame, whatever you want to call it.

  7. Hey, if my undergrad degree gets more valuable because I went to the same college as Obama, I won’t argue with it. I don’t think Whittier College did much better because Nixon went there.

  8. and how many Dutch people have won the Nobel Prize? For Animal Behavior too!? Tinbergen’s is the giant upon whom all animal behaviorists stand.

Comments are closed.