Hacker News new | past | comments | ask | show | jobs | submit login
What principal investigators taught me about my failure to land tenure (nature.com)
116 points by pseudolus 35 days ago | hide | past | web | favorite | 72 comments

23 years of post-doc with 5 first-authors? That's really all you need to know: that's a terrible publication record for someone looking at tenure-track.

A decent target is an average of one first/last author paper a year. Two one year and zero the next isn't a problem, but one every five years certainly is. This doesn't even get into the 'careerism' of picking safe but dependable projects; anyone making reasonable choices about what to investigate and how to go about it can hit that kind of rate. It takes a lot of work, but that's very much the state of the field.

This guy isn't a PI.

He's not leading a research group, nor does it seem like he's all that interested in doing so. It sounds like he's interested in performing research in someone's lab, a staff scientist.

There's nothing wrong that; it would be quite nice if there were more long-term career opportunities for people like this. They tend to be the people that develop the highest skill in actually performing experiments. They are useful, not just for their own productivity, but for the expertise and continuity they can bring to a lab. They are incredibly valuable for graduate student mentorship. They occupy a wonderful space that is otherwise vacant due to the demands placed on other lab members.

Generally only larger labs can afford such positions, and the competition of cheap post-doc labor means that many are stuck in eternal post-docs where they are underpaid, underappreciated, and given inappropriate expectations.

I don’t know the norm of the author’s specific field but I had the same reaction: His publication record is not even in the ballpark of where it probably should be for tenure.

I feel so sad that the scientific industry is set up such that guys like this can plod on two decades or more past the point they should have realized they weren’t on a tenure trajectory. Now he’s 40-something and has to start over in something else (data science?), probably with very little saved for retirement and possibly having missed out or significantly delayed other aspects of life (marriage, kids, house). Why didn’t anybody tell this guy to start his career in private industry 15 years ago? Every PI in every lab he touched should have seen this coming.

>Every PI in every lab he touched should have seen this coming.

People are surprisingly willing to let you accidentally pour your life in to something that only marginally benefits them, that's why you have to be ever-vigilant against the many opportunities you'll have to do that. Some people are prone to slipping into the mentality of the ideal feudal peasant: "life is bad here, but the lord of the estate isn't going to grant me any freedom, so I'm just going to focus on making the best of it." A mentality like that would be great for someone who had no freedom, and indeed it would make the best of an immutable bad situation, but I see a lot of people slip in to it even when they do actually have the ability to jump to a different life track. Scientists are especially prone to this because they're pre-selected for high levels of tenacity and low levels of impulsiveness.

There is wisdom in this comment, thank you for putting the sentiment into words.

The really bizarre thing about this article is that the headline is about failing to get tenure, but the content is about failing to get a tenure-track job. These are really different things. I would think the author's publication record would maybe put them in the ballpark for a tenure-track job (though maybe not over 12 years, and certainty in a lower-ranked department), but absolutely not for tenure in the sciences.

I might be a bit too cynical about this topic, but it really should be obvious to anyone in the field that you're not on track to become a PI this way.

If you want to pursue the academic career, you need to focus on that from the start of your PhD. And you really need to be on some position on a track towards full professor after one or at most two PostDocs.

I'm quite confused by the timeline here, though. A PostDoc is usually something around 2 years, maybe 3. Four PostDocs in 23 years is unusual, and indicates positions that are closer to something like a staff scientists usually done by people with permanent positions.

Why would it delay marriage? Are people holding out for partners with tenure?

Dating takes time, weddings cost money, and its hard to drag partner/spouse/family around the globe for each new short-term position. I once worked in a lab that joked about a “celibate for science” lifestyle. I think the PI really actually wanted that for his grad students.

I had a coworker in a lab who was married to another scientist. They had been married more than 5 years but had never lived in the same city.

My sister and BIL are both academics, and didn't "live together" for probably around 10 years, 6+ of which were after marriage. They bought a house together in the city my sister taught in, and my BIL commuted on the weekends sometimes, and they spent summers together, but it took the better part of a decade for them to both land jobs in the same city. It's incredibly common.

Haha. Not quite but we did wait.

The logistical stuff is difficult as a postdoc. A traditional wedding is fairly expensive and time-consuming to plan, and many postdocs are cash and time poor. People often like to have the ceremony near family—-isn’t the bride’s hometown traditional?—-but people move, often very far, to find good positions. Traveling back and forth is also tricky if you have experimental time points or lab animals to take care of, as we did.

More psychologically, being a postdoc is a weird transitional state. I felt odd promising life-long commitment and support to one person, while not knowing if my own job was still going to support me next year. (I was on one-year renewable contracts).

Finally, we were a little concerned that being married could make things more difficult for my wife. Although they shouldn’t, a surprising number of profs are reluctant to take on a postdoc who might be ‘distracted’ with a spouse or kids. Furthermore, if you have a kid as a professor, you often get an extension in the tenure clock, help with daycare, etc. As a postdoc, you can maybe get a little more time on these time-limited fellowships, but that’s about it.

We eventually did get married though (and it’s been great) but waiting was not totally irrational.

Because you get paid peanuts while working long hours.

> 23 years of post-doc with 5 first-authors?

No, 12 years of post-doc with 5 first authors: “after working in several postdoctoral positions over the course of 12 years, I started applying for tenure-track and other permanent positions. At this point, I had 19 peer-reviewed publications, including 5 as first author”

> Why didn’t anybody tell this guy to start his career in private industry 15 years ago?

He did start trying that no later than 11 years before you seem to think, since at the same 12 years point referenced is when “[i]n the next 3 years, I sent out more than 100 applications to academic institutions, to the private sector and to government agencies (predominantly in the United States, where I was living at the time).” [emphasis added]

That's still pretty low. In the us at a top tier, which you'll need for most tenure track positions you're expected to have about 5 first authors in five years as a trainee and moving forward into a postdoc a pace of about 1.5 to 2 would be good. Strategic timing by having a really high profile paper come out from grad school around when you're finishing your postdoc probably also helps.

But as I indicate in a direct response I don't think any of that really matters either

Everyone's talking about publication count. Is the quality (influence, number of citations, any other relevant metric) of those publications really so irrelevant?

No, and yes in a way. The quality of the venue matters a lot, and many people seem to obsess over h-index values, which quantify the citation count and journal impact factor. But in my opinion these things are useful but flawed heuristics for the actual importance or impact of research work.

Imagine you are on the professor search committee. You get inundated with 500 applications, many from jokers with no chance from a diploma mill who just spam every search committee. How do you filter so many applications?

I find number of papers published to be a questionable metric for measuring qualification. It feels similar to judging a software engineer by number of commits. Is this really a key measurement used to determine tenure track?

I think peer-reviewed publications measure something closer to "sprint points closed" or "major releases shipped" than commits (because you can break commits nearly arbitrarily small, where a too insignificant paper will fail peer-review).

A primary purpose of an academic researcher is to contribute to the field via research, largely measured by accepted publications where they are the first (did most of the work and writing) or last (group leader/principal investigator) author.

This. I had more than that as a graduate student.

Accept your data. According to one PI, success depends more on the people doing a project than on the topic of study; some people make their project work no matter what the topic is. Others block their own way to success because they require too much evidence. I, for one, have always been excessively critical of my own data. Although I am a meticulous experimenter, whenever my results seemed to confirm my hypothesis, I was afraid to accept them because I knew I was biased (it’s funny, I know). I was spending too much time looking for alternative explanations instead of accepting that I was on the right track and moving on to the next step. We have all heard about the reproducibility crisis — and I am not advocating the reckless publication of unverified data, or recommending being so careful that you end up publishing nothing — but there is a difference between being reckless and shooting yourself in the foot.

And this, fellows, is how you end up with the reproducibility crisis. Incentives are just all messed up.

You are reading that completely differently than how I read it. I personally think that the author is reflecting on the fact that he may have been not focused enough on completing a single task before starting another. It sounds (and this is speculative, I don't want to put words in the author's mouth) like the author would start with a hypothesis, run an experiment that generates data that appears to support that hypothesis, but then start running tangential experiments disproving alternative hypotheses to further strengthening the original hypothesis. Hence his refelection:

> I am not advocating the reckless publication of unverified data, or recommending being so careful that you end up publishing nothing — but there is a difference between being reckless and shooting yourself in the foot.

To me, the author is looking back and now realizes that he may have been too cautious in that he was running too many experiments to publish one result and that there is a spectrum of caution and that he may have missed the sweet spot. As others have pointed out ~19 papers (and only 5 first author) over 23 years seems to be a low publication rate, which I think the author is reflecting on.

I agree that the incentives are messed up and these upside down incentives are a driving force behind the reproducibility crisis, but I wouldn't take this quote from the author and say it is an illustration of the reproducibility crisis.

> As others have pointed out ~19 papers (and only 5 first author) over 23 years seems to be a low publication rate, which I think the author is reflecting on.

Low compared to what? Compared to the expectations set by his fellow scientists in a field shaped by the norms and incentives in question.

There's a circularity here (and in the earlier comment by Obi_Juan_Kenobi): apparently this guy's real problem was his own low productivity, which we know because he published less than his colleagues and competitors -- therefore his case can't tell us anything about the warped incentive structure that favours publication maximization over intellectual integrity.

By the author's own admission, he thinks he could have focused himself a little better. He isn't saying he would have had to cut corners and do poor science, just not "shoot himself in the foot" to use his own phrasing. In reflection it sounds like he thinks that there was an opportunity for him to publish more frequently without doing bad science.

This is why I don't read his self-reflection as necessarily being an illustration of the type of mindset that leads to the reproducibility crisis. It sounds like a reasonable scientist reflecting on things he might have reasonably done better with the benefit of hindsight. No where do I read the implication that he is lamenting his choice to not do shoddy/quick science for the sake of securing a tenure track position.

Moreover this is self reflection so a lot the things he is reflecting on are speculative in that he thinks they might have helped him find a tenured position, but he can't be sure. He even has this disclaimer at the beginning of his piece:

> I’m not claiming to know the formula for how to get tenured (if I had that, I would have used it for myself). Instead, I would like to offer the advice I’ve gathered.

Yeah - I know scientists who do this and it has nothing to do with the reproducibility crisis. It's a desire to have a "complete" story, to have explored every nook and cranny, and have the perfect, flawless magnum opus of a paper.

That way lies madness.

What's good for the field (we've decided) is when we only allow a 5% Type I error rate (95% of published positive results are correct). What's good for the individual is to drive their Type I error rate up just enough so that they can publish as much as possible without getting caught.

Umm... That's not actually what only publishing a result when the p-value is less than 0.05 accomplishes. Such a publishing policy instead guarantees (or would if there were no other problems) that __only 5% of the FALSE ideas that researchers test get published as being "confirmed"__ (i.e., when the null hypothesis is true, only 5% of the time does it get "rejected" at the 0.05 level). It guarantees nothing about what fraction of these supposedly-confirmed ideas are actually true. That fraction could vary from zero (if researchers never think of true ideas) to 100% (if researchers never think of false ideas).

You're right, I was trying to dumb down p-values to make a point about perverse incentives in academic publishing. What's worse is that other factors that increase the Type I error rate (only positive results are published) only compound the effect.

A 0.05 p value very much does not mean there's a 95% likelihood of the result being true. It's a fallacy that is far too common.

Like it or not, being a good scientist (read: getting tenure track and funding) is not just about good science, but also exciting, cutting-edge science.

If a scientist runs a couple experiments and find something truly remarkable, the guy who rushes to publish is going to win out over the guy that says “this needs another year of work before it’s perfect”.

I disagree with all of these insights. I checked a lot of boxes, had great scientific output, managed underlings well, had an excellent pedigree (including top tier undergrad and grad schools and Nobel laureate as my immediate postdoc supervisor), but failed to obtain a tenure track position.

Looking at the peers who made it and didn't (including ones that were slightly less deficient), I realized it's because you need to have one of those PIs go to bat for you. It was something as a leader I did for my underlings (I arranged for both of my interns to have one-on-one lunches with a Nobel laureate and a his buddy that didn't get to win a prize in spite of inventing site directed mutagenesis - an experience that I haven't even had myself). As soon as I realized that my high profile bosses had checked out and would never go to bat for me or anyone really (still love them though) I got out of dodge.

Anyways, that my advice for anyone who wants a tenure track position. 1) Table stakes are a top tier postdoc (ideally a top tier grad school too). 2) Table stakes are 1-1.5 (ideally 2) publications a year. If you don't have these it will be VERY hard. Don't feel guilty for giving up, don't let anyone trick you into thinking it will be easy. Most most most importantly, work for someone in grad school or postdoc who will network for you.

I'm in the tech sector now, and have performed well enough that my CEO seems to think of me as a performant employee, so we'll see where that goes. Maybe I'll get really rich and pay for my own science. (doubtful, but at least I get paid well enough to not worry about my future).

Coming from another field, 23 years as a post-doc seems insane to me. Is this common in some fields? I never thought of post-docs as (semi-) permanent positions or something one would spend a major chunk of a career doing. I always thought of it as a transitional position, something people did while they worked on getting their own funding or transition to industry. Am I wrong or are there different standards in different fields, or is this person just an extreme outlier?

I’m also from another field, but 19 papers and 5 first-author papers over 23 years is a huge red flag. Many graduate students wind up with 3-4 first-author papers by the time they graduate. I don’t know what he was doing with his time, but the publication record seems to be the elephant in the room.

To be fair, even though it seems low, paper output/metrics are very field dependent.

I'm close enough to their field to say that that's a really low number.

Agreed. I don't know if this applies to the article author, but I've met a few "lifestyle" post-docs. They do just enough to keep in the system, but don't rise above that (either capability or motivation).

There's a huge glut of PhDs and postdocs in the life sciences compared to the number of jobs available. 8K new biology PhDs each year compared to maybe 2K in CS in the US; probably 25-30K biology postdocs today compared to 1-1.5K CS postdocs.


Big pharma is cutting R&D left and right, and there aren't enough small companies to provide jobs for all these scientists. There are very few tenure-track positions available, so a career in academia isn't really an option either. Compare this to CS where by the sounds of it there are far more open positions than qualified candidates (not my field, but i hear this all the time)

If you are a smart software engineer who just wants to write good code and have good work / life balance, you can make low to mid six figures at a big company. If you are a smart biologist who just wants to do good science and have a good work / life balance....you don't really have many options. You can be an eternal postdoc, but the work life balance isn't that great, and you're making $40-60K, even in your 30s

Why is big pharma cutting R&D?

The past decade of commercial research has failed to deliver the market successes needed to justify the premium. If you look at the biggest winners for pharmaceutical companies over the last decade (immunotherapy is the poster child), most of the research happened in academic labs. Cheaper to license the winners than do basic research yourself.

That's partially true (big pharma r&d has failed to deliver returns) but it is not true that most of the research happened in academic labs. Most of the research and early development happens in small companies. Most of the later stage development happens at big pharma companies

There's also a melding of those two - there's a big push for university researchers to do commercialization and start startups. Those small companies bear a lot of the risk of early stage development, and then get bought out when they get something promising.

Obviously not everyone can promote all the way up, and not everyone wants to. There’s nothing wrong with someone doing good competent work at the level of a postdoc for as long as they want. We have a postdoc in his 40s in the lab I used to be in and he does excellent work as part of a group. I don’t think he wants to run his own lab and why must he?

So a couple issues:

1) As postdocs get more experienced, their pay usually goes up.

2) As mentioned in the article, it's possible to "age out" of a lot of fellowships, which have limits like "5 years since your last degree" requirements.

Which means the burden then falls on the PI of the lab to support them as essentially a staff scientist.

If I pay a postdoc $60,000 a year, at 10% pay lines (which is not unreasonable for the NIH), I have to write about $600,000 in grants in expectation to be able to afford them. That's a lot of work, and unless they are significantly better than a new postdoc, potentially not worth it. They're also not a good speculative investment - they're not going to go on to be a PI of their own lab, which has benefits for their PI down the line.

The NIH scale doesn't rise that fast. A brand-new postdoc costs about $50k on NIH pay scales.

There is often also an assumption that more senior people will also bring on money, either completely by themselves or by enabling more projects. The (completely idiotic, IMHO) 2 or 5-year cutoffs on some fellowships makes this somewhat more difficult.

I also think experience is weirdly undervalued in academia. I would bet that, in many fields, three random students at $25k/ea are not appreciably better than one highly skilled staff scientist at $75k, especially if the lab actually paid full freight for the students instead of having them subsidized by training grants and time-limited fellowships. This is tricky to study though....

1 skilled research staff will out produce even 5 early PhD students, no doubt.

But I would be inclined to hire the students. There are more funding opportunities for them (NSF prioritizes students and departments fund TAs) and I’m expected to graduate students. In my field (CS), research staff and postdocs disappear too quickly anyway.

That's another aspect I didn't touch on, especially for the NSF - they're not terribly interested in funding staff scientists.

A few things...

1) A brand new postdoc costs $48,432 on the NIH scale in 2018. This author of this article costs $59,736. Their benefits are also about $3800 more. It's not that fast.

2) I agree that experience is undervalued - and said staff scientist will almost certainly out produce them. But the burden often falls in different areas - I'm evaluated based on mentoring students for example, not staff scientists. Training grants exist. Etc. A senior staff scientist is always full freight born by the lab (barring like, really nice retention/recruitment packages).

But I think the real problem is the author isn't some sort of rockstar senior staff scientist (I've known some). Their productivity just isn't there enough to assume that they're 15 grand better than trying a new draw from the postdoc pool.

Idk he might eventually grow tired of living off tuition and taxpayer money.

Would becoming tenured change that? I would have thought both postdocs and tenured researchers would usually be funded mostly by research grants.

They are.

Apart from the 'duh' factor about someone in their 40's with a whopping 5 first author papers not getting a tenure track position, and not just any, because she refused one in Hungary because 'it didn't fit in her life' (but then 5 years later she does move to Europe? - but anyway - ) ; the rest of the advice is actually quite good. Don't be a passive 'just doing what I'm told' graf student/underdoc, have a plan and make sure you drive your progress towards that plan instead of hoping someone else will carry you.

I've been on search committees, and the number one thing that will kill someone is "They seem like a technician." Being good at implementing someone else's ideas is not the way to be a PI.

Self-direction is essential.

Lots of harsh comments here, some seem a little short-sighted. I wonder how many are actually from Tenure Track Lecturers? Every field is very different. Some fields value industry experience highly, some care only for publications. Some value teaching experience more than others.

My experience of getting a tenure-track position is that you’ve got to be the right person in the right place at the right time. My academic record is patchier than the OP’s, but it is offset my several years at fairly senior level in industry, which my university values happens to value. And I entered the picture just at the right moment.

Wow, lots of harsh comments on this page.

Academia is kind of a pyramid scheme and there is a huge mismatch between the number of graduates and academic jobs. Moreover, universities are turning increasingly to adjunct faculty rather than adding tenure track positions.

Fortunately in the sciences and engineering you can work in industry or government labs; it's a lot harder for people in the humanities and social sciences.

I was also surprised by how harsh the comments are, but I suppose it mirrors the harshness of the academic world.

yeah,,, there's something about some comments on here that seems... emotionally unhealthy....

From my interactions with students and postdocs in biology (im not a scientist myself but work in biotech), there is a culture in many research institutions that views working in industry as "inferior" and glorifies the tenure track as the only respectable option. Though the odds are terrible of getting tenure even for scientists with top notch publication records, many scientists pursue that path, just as many entrepreneurs think they will start a billion dollar company even when the odds nearly guarantee they wont

I know scientists with PhDs from top-5 universities doing postdocs in highly regarded labs who have only seen a handful of job postings in their 15 years in academia. Many scientists focus solely on the tenure track, and only reconsider after a few years in a postdoc shows them how hard it is

"At the age of 41, after working in several postdoctoral positions over the course of 12 years, I started applying for tenure-track and other permanent positions."

Why would you wait 12 years after your terminal degree to START applying for permanent jobs?

Author appears to be a cell biologist, where this is not atypical. There are very few permanent jobs, there are hundreds of applicants for every one, and the hiring committees would generally rather not hire anyone than accept less than the very top applicants.

Author would surely have been well served by leaving science sooner. As another commenter noted, his output wasn't impressive, and it should have been obvious he would never be hired as an independent researcher. But that can be psychologically difficult for a 40-year-old whose project of becoming an independent scientist started when he was literally a teenager.

For many in the third world countries, going to America for Ph.D or for postdoc is a way to get out of their native countries. However, if you are an American citizen or a citizen of any developed country, be careful about getting on that post doctoral treadmill, which Prof Katz calls this phenomenon.


This is so true... even the doctoral treadmill isn't necessarily worth it. I'm a little disappointed that he claims that American students who pursue PhDs are weak students. This is not necessarily true.

For many students in Iran, getting a student visa for MS is difficult. They join a Ph.D program in flyover country.

They guy may have been a low performer or not. But the whole system is very very broken.

"My take-away from the careers aspect of the book is that if you (1) love competition, (2) have a huge appetite for risk, (3) don’t mind working long hours for a minimum of 27 years until getting that first grant, and (4) are mostly indifferent to money, pursuing a physics PhD and an academic job might be a reasonable plan. You’ll get to work with a lot of smart people, for sure, but, as the book notes, quite of few of them may be planning to stab you in the back when it comes time to assign credit for a Nobel-worthy discovery. It is not like most other fields of human endeavor where there is room for everyone to excel in his or her own way."


A few points many are missing :

Publications/year depends on field. In parts of biology, you just need one every 3 years to be fine. The journal matters too. Even one every 4, if it's in Nature, Science etc., and you'd secure a job. And field: Even papers not in Nature if they're in a hot field would be enough. And PI: even papers with a famous PI, without being a hit field or high impact, might be ok.

In this case the author appears to be a biochemist without any high impact publications or from any famous labs. So yeah, no dice. (source: just Google him)

I think it’s really valuable to share more failure stories. Lessons learned are common from successes but I have a hard time seeing earnest, true failure stories that help me learn. I mean true failure as ones that don’t involve a super successful person reflecting on a failure that had a silver lining (eg, Odeo->Twitter years into Twitter’s success; countless TED talks) not because they aren’t valuable, they are, but because they are safe and are musings that all have survivor bias because the really bad failures are weeded out.

That being said, this format is weird because it’s a numbered list of advice that is difficult to know how valuable it is for landing tenure. When I re-read it as life coping advice or how to research well, it helps. But phrased as tips for tenure I don’t find it very helpful. Although I’m not interested in tenure and not in that world, so I may not appreciate some of the content. I am really interested in organizational innovation, culture change management, translation, and research and think this is helpful.

Landing a faculty spot is about one thing: how much money will you bring in for the department you’re applying to? All these other metrics (paper count, references) are simply indicators of your potential to do that.

Obtaining funding is a function of your existing publication record, network (read: relationships with program directors), but also your ability to sell new ideas particularly as a member of an interdisciplinary team. You can be the best in your area but if you don’t offer synergy with a dept’s other PIs it’s probably a no go.

Good nugget in the middle of the article:

> Of course, “be lucky” is not a very useful piece of advice. But some of their advice might help you to recognize and take advantage of your luck.

Many years ago I was told the new-age aphorism “luck happens to those who let it” and at least this unfortunate writer was able to see this.

My wife and I used to quote the statement to each other quite a bit as it happens. It helped make us seemingly more lucky than most.

Or as Louis Pasteur put it: “chance favours the prepared mind”.

All great advice even outside of academia. It is also important to look at other issues such as demographics in the field itself. Computer Science is now on a tear because so many kids wants to study CS and deep learning is the new hotness. Structural biochemistry is also probably about to explode due to deep learning and recent novel techniques for fast structure determinations. There's probably a few startups in this space looking for someone like the author.

My university has a 6 year limit on postdocs.

TLDR if you start thinking about tenure @41 you're a decade late. No one in her 50 advisors told her that?

In biology this is not uncommon.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact