A record of my own undoing.

July 24, 2011 1:07 PM

How do I cope with persistent project failure at work?

I am a full-time assistant at a research center. I've worked here for more than three years, mostly on the same small handful of projects. During this time, we have accomplished very little of value. I find this frustrating and stressful, and it makes me question whether I should stay in research.

Some examples of what I consider to be "less success than we hoped for." We've spent the last 6-8 months developing a cognitive test that we want to use instead of an established one. After painstaking piloting, norming, and scoring, it turned out that the test we developed is almost completely orthogonal to the one we're emulating. A little while ago we wrapped up a year-long brain imaging study that failed to produce any results whatsoever. We were unable to confirm even the most rudimentary predictions. This resulted in the loss of funding for that project. The past three years have been more or less similar.

I've considered asking for a transfer to another project or research team, but I'm afraid of how that would be interpreted. Whenever there is a possibility that I might get reassigned to assist another researcher, the researchers I support now petition to keep me on their project because of my specialized skills. And it's true, their projects are the ones that I'm best equipped to help, but I'm very tired of them. I've been developing contacts at other labs on our university campus, but have been unsuccessful at finding another placement. I'm loath to leave the campus entirely because I'm getting free tuition on the classes I'm taking here.

My impression is that unsuccessful research on this scale is unusual and symptomatic of bigger underlying problems. I am, at least in part, working here to develop research and technology skills. If the training I'm getting is substandard, then I'm hurting myself the longer I stay here. Also, each time something goes wrong — data analysis, whatever — we start doubting the correctness of each of the steps we took to get there, panic, double-check and rerun all the scripts and analyses, etc. This wastes time and makes me doubt everything I do, not to mention how unpleasant it is to justify everything I do to the researchers I support. It never gets to outright finger-pointing, but the collective loss of trust is very stressful.

1. Am I working at a dysfunctional lab?
2. Is this likely to hurt my chances of joining a good graduate program in the future, should I decide to apply?
3. How can I cope with long-term projects failing to get results?
4. How do I cope with the loss of confidence? I mean, it's always possible that it's all my fault, isn't it?

Thanks.

posted by Nomyte to work & money (8 answers total)

It can't be all your fault unless you're the PI. They shouldn't be letting stuff get out of hand like this. I'd take action because you don't want to be associated with this sort of thing.

I think you should look for another job somewhere else entirely. I think most workplaces should let you study for free.
posted by tel3path at 2:15 PM on July 24, 2011

If it pays for your PhD (and better still, gives you something about which to write your dissertation), and you get the skills you want out of the process, does it really matter that the group itself appears useless?

A friend did his PhD research on laxitives, for a complete Mussolini of an adviser. You could have it worse. :)
posted by pla at 2:17 PM on July 24, 2011

I know next to nothing about the inner workings of research labs. But, I do know it is not all your fault. At the very least because it anyone set the lab up in such a way that could even happen, I'd pin the fault on them. (And I'm sure there are better reasons). I worked in a place for a while where it seemed like all our designs just got shelved. It does give you a bit of hit. My solution was to undertake and succeed in outside of work projects.
posted by meinvt at 2:27 PM on July 24, 2011 [1 favorite]

Non-science type here, but:

- it wouldn't seem that your work there as an RA would be viewed all that critically, if you're a junior member of the team. The PI's reputation is the one that might be getting dinged here, BUT...

- as a non-science type, I'm always being told about science that there's a lot of dead ends. Or is it just a high-flown myth that one should be unafraid to pursue a hypothesis, even if the work disproves it? I realize that some hypotheses (hypothesis's? hypothesi? It still sounds right...) are less likely than others. I mean, if I got a grant to pursue the idea that you could cure cancer by sprinkling baking soda on it, one might suspect that I'm either an idiot or that my funding from Arm & Hammer is distorting my judgement, but you know what I mean). The "developing a test that turned out to be no good" caper sounds suspect, but to have done a brain imaging study that didn't confirm your predictions - isn't that why you run the experiments/tests/models/whatever - to see if the predictions were confirmable? So I'm not sure that your time there is the total loss you think, nor is it necessarily due to incompetence on the part of your team.

- are the others in your team disenchanted/concerned? Are tongues wagging? I know from what you said there is some stress, but is your reputation being damaged? If not, I'd stay the course.
posted by randomkeystrike at 2:47 PM on July 24, 2011

Nomyte said: "How do I cope with the loss of confidence? I mean, it's always possible that it's all my fault, isn't it?"

Unless you're the sole person doing all of the work, no. If you don't have control over something, at the very best you can only influence it. I'm not familiar with research labs, but I'm guessing that there are a lot of variables that are completely out of your control.

Consider hitting a nail with a hammer. Sounds simple enough, right? Just hammer the nail into the wood. You ave control over how much pressure you apply. You have no control over how well the hammer was made or how strong the nail is. The hammer head might come off. The nail might bend. The person who is holding the piece of wood steady might move it at a critical moment. Any of a number of things can go wrong in a reasonably simple task. Extrapolate that out to the sorts of things you're trying to achieve in your research, and it gets very complicated very quickly.

Nobody gets it right all of the time. Ever. People make mistakes. Making mistakes and learning from them is what makes us grow as human beings. Consider that checking your work for inconsistencies is teaching you better habits for the future. If something goes wrong during some other project, you'll have more skills to handle that and work out what is going wrong than someone who always got it right.
posted by Solomon at 2:52 PM on July 24, 2011

I'm a researcher in a field very similar to yours. Been through grad school and all that. I hate to say it, but failure in science is very, very common. You can spend months (or even years) on a project without getting much out of it. Partly this is bcs there are many modes of failure in science -- technical problems, poor study design, results didn't fit your hypothesis, etc. Try to figure out what went wrong in each case. If it's a matter of results not fitting your hypothesis, well that happens. And it can mean years of work down the drain. If it's a matter of poor study design or other controllable things, and if this is a persistent problem in your lab, then try to find another research team to work with ... although I know you mentioned that this is difficult.

To get into grad school, most likely a couple unproductive years won't hurt you. Grad schools will care more about your training, your expertise, and your recommendations. You mentioned that your colleagues want to keep you on the team -- this is a good indication that you'll get great recommendations; it sounds like they have confidence in you. This also suggests that they don't think the failures are your fault; otherwise they'd be happy to see you leave.

On the other hand, if you continue with this lab for a few more years and still don't get your name on a paper, this might raise eyebrows when you start applying to graduate programs. In science, productivity trumps everything else. Nobody gets an A for effort. They'll wonder why you worked for so many years without getting a paper (yes, even though they know how hard research is).

How to cope ... well, I'm afraid you just have to get used to it. Science is like this. Unless you're very, very lucky, you'll encounter a whole lot of failure in your career. If you're like the vast majority of science students, you'll question again and again whether you're good enough, smart enough, and hardworking enough to "make it," and your confidence will diminish more than you ever thought possible. It's not easy. You have to really, REALLY love research to get through all the crap.

Sorry to be such a downer, but that's the reality of it. Best of luck to you!
posted by phoenix_rising at 5:14 PM on July 24, 2011 [1 favorite]

*laughs bitterly* Ha ha ha, welcome to my life. I'm a research assistant (molecular biology - so, the same, but different). I've been doing this for 7 years, two in academia, five in industry. So excuse me if I ramble, but I have a lot of opinions on this.

1. Nothing you've described suggests that you are working in a particularly dysfunctional lab, although losing a grant sucks. But, as you say, "...each time something goes wrong — data analysis, whatever — we start doubting the correctness of each of the steps we took to get there, panic, double-check and rerun all the scripts and analyses, etc. This wastes time and makes me doubt everything I do, not to mention how unpleasant it is to justify everything I do to the researchers I support. It never gets to outright finger-pointing, but the collective loss of trust is very stressful. " That is how science is supposed to work. Well, without the panic. You should be able to justify your choices. And, while I obviously don't know the tone they are using, your coworkers aren't necessarily trying to undermine your work. Sometimes you want to know exactly how something was done in order to see exactly what happened. There's sometimes good news hidden in with the bad - and, y'know, understanding your mistakes is how you learn what to do next time. If you really feel that there is a lack of trust... that's something you need to address.

2. Well, as noted, I've been working for 7 years and have no papers and no patents. I managed to get in to a grad program (which would have started this fall, though I chose not to attend). Since you're not the PI designing the project, I don't see why it would hurt your grad school chances at all. That said... I would certainly not express too much frustration at the failures while applying to schools. As p_r noted above, there is a LOT of failing in science. (I also failed to even get interviewed at another 4 programs and was not at the top of the list for the one I did get in to, so maybe that 'why are you still doing this?' that p_r mentioned is valid. ^_^)

3. This is the question I haven't really got an answer to. It's really frustrating. I've been working on a project for over a year that I'm pretty certain will never work (at least not the way we're approaching it) and I hope to high heaven that we switch soon. And, FWIW, you've haven't known real frustration until you've been taken off a project that was working OMFGKliuDSFKLjhskdfjkjsdfhjkfslkwggqwy. Ahem. The best coping method is to enjoy what you do and who you work with day to day. If you are only continuing at this institution to take free classes, you need to rethink your priorities. (I'm not being a smartass or idealist. If the frustration isn't worth the free classes, you need to look outside this research institution. That would be why I left the academic institution I worked at, despite the best benefits ever and a pay cut. You can't be miserable 8 hours a day, 5 days a week.)

4. Yup. It's always possible that it's all your fault. But first of all, it's not terribly likely that it's ALL your fault, so don't panic. Second, run every bloody control you can whenever you have doubts. Third, and most importantly, learn to understand everything that can go wrong. Seriously. If you understand how all of the controls and variables in your experiment should work, you have a much better place to start debugging. And finally, keep in mind that is also possible that it isn't your fault at all. Sometimes you get the wrong part from a supplier, sometimes parts are backordered, sometimes you were told to do the wrong thing in the first place. Breathe.

So, aside from those numbered questions, my only advice after a few extra years of assistantry is this - if you do want to apply for a transfer, you need to couch the reasoning in terms of training. You would like to learn a new technique or field. But be prepared to fail just as often in the next one. Sometimes nature simply does not do what you want it to, no matter how good or informed an idea you had about it.

Sorry for the babbling.
posted by maryr at 7:27 PM on July 24, 2011

This may be naive, but is it possible to make lemonade here? When you set out to prove something with science and it doesn't work isn't there some value in being able to say that "x doesn't work". Naturally it is more interesting to have a supported hypothesis, but it seems that there is even a good story or lesson in perceived failures.
posted by dgran at 6:46 AM on July 25, 2011