A medium sized laboratory

September 3, 2015

How many staff members (mix of techs, undergrads, graduate students, postdocs, staff sci, PI) constitute a “medium sized laboratory” in your opinion? 

36 Responses to “A medium sized laboratory”

  1. docack Says:

    Bigger than 5, smaller than 15 = mid sized.


  2. Joe Says:

    8 to 10


  3. baltogirl Says:

    I think mid-sized is smaller, maybe 7-8. One or two techs, two undergrads, two grad students, two postdocs, maybe one more or less of each. I would argue a mid-sized lab could be supported by 2 R01s. A small lab- one R01.


  4. drugmonkey Says:

    Are you talking modular R01s? Or ones that come close to the $500K firm cap? Or sky’s-the-limit lucky duck R01s?


  5. I wouldn’t count undergrads – they generally come and go frequently, and are often cheap or free for the PI. I’d say a “medium” lab is 8-15 core members, where core members are grad students, PDs, and other techs/staff scientists.


  6. drugmonkey Says:

    Undergrads still add a burn rate to the supplies budget though.


  7. Philapodia Says:

    Depends on your level of BSD-ness. Many of us riff-raff would consider 6-8 a medium size lab, whereas those in the vertically ascending club probably think that less than 12 is a small, ineffective lab. Also depends on how students are supported. At my Uni students in my department have to be supported as GRAs or fellowships on grants, whereas in another basic science department with a large teaching commitment students are almost always supported by departmental GTA-ships. That department has about 5-6 times the incoming students we have every year and the labs tend to be quite a bit bigger.


  8. SPeyton Says:

    Agreed with Philapodia. This depends on the university and field. Where I am a “big” lab is 10+ grad students/staff/post-docs. I think that would be considered a small lab at my post-doc institution. Mainly this is due to how much institutional support there is for students. My department has 0 TA lines, so no free students unless they win an extramural fellowship. This heavily limits how big a lab can get.


  9. drugmonkey Says:

    I’d put that into the discussion of hard/soft money salary percentages for PIs. It’s an institutional commitment to use supposed TA dollars to support graduate student research activities.


  10. MF Says:

    The TA support may depend not only on the institution’s commitment to support research but also on the need for teaching. We have hard money salary support for faculty but very few undergrad courses so there is no TA funding (just a few institutional GA stipends).

    For us, a lab with 8-10 people is considered large.


  11. Anka Says:

    I’m in the life sciences at a major research university. I’d say a small (and probably struggling) lab is 2-5 people. Medium is 5-10. More than 10 is large. More than 20 is HUGE. I came from a medium graduate lab (8 people) to a HUGE postdoc lab (26 people).


  12. drugmonkey Says:

    Interesting point Anka. Very good question about how lab size relates to medium to long term viability. Why do you think 5 or fewer = “struggling”? Is this a marker or a cause?


  13. Bashir Says:

    Lets say 2 grads, 4 RAs and 1 lab manager or postdoc. 7 seems medium to me (not including PI).


  14. baltogirl Says:

    In my experience larger labs are considerably LESS productive per capita than smaller ones. I never want to have 11 again. Six is great, for me.
    More than that and you need an intermediate management level that I don’t have (i.e. stable techs or staff- very difficult to assure continued funding for, these days!).


  15. drugmonkey Says:

    Does that take into account productivity across the long haul and including small labs that fizzle out? Easy to compare the small labs *that survive* against all large labs but it ignores viability.


  16. Noncoding Arenay Says:

    Counting a maximum of 3 undergrads who are in the lab for at least 2-3 h a day, I would consider a medium sized lab to be 8-10 people.


  17. zipper Says:

    I also found the standards also differed on location: where I was in the US a med. sized lab is ~10 people; where I am overseas a med. sized lab is ~6 people (not incl. undergrads). Smallest lab I’ve ever been in was 4 (PI, tech, postdoc, student). It didn’t strike me that we were less productive than any of the other bigger labs I’ve seen.


  18. Comradde PhysioProffe Says:

    What length and girth constitutes a “medium-sized cock”?


  19. Namesaste_Ish Says:

    This is all bullshitte unless you are talking about what field you are working in. Stacking folks up on rig’s or when non-human primates are involved is going to be resource limiting.

    There is also a point where you just hire someone who oversees your operation. I intellectually max out with 6 FT trainees + their undergrad minions.

    But then again, I also intellectually max out most days anywho.

    Now how do I put my damn .gif in this comment area?


  20. Namesaste_Ish Says:

    Fuckke….of course CPP would post right above me. DAMMIT.


  21. dr24hours Says:

    Namaste_Ish is right. Hugely field dependent. In epidemiology, a “lab” might be prof, postdoc, GS. If I had a “lab”, I’d never want more than 5 total, including me. At my gradschool, labs never included postdocs, only GSs. Rare to have more than three at a time, usually not that many. Even for the “big” experimental robotics stuff.


  22. Dave Says:

    What length and girth constitutes a “medium-sized cock”?

    Length is less important. Everyone knows that.

    My place, a BIG lab is probably 2 R01s worth, so probably 2 post-docs and a couple of grad students and a tech or two. You are killing it here if you have this set-up. Most labs have a single grant (if at all) and are getting by with 2 – 3 people. Are they struggling? Yes. Will they always be struggling? Probably.


  23. L Kiswa Says:

    1-2 full time trainees (PD, GS) = small; 3-5 = medium; > 5 = big.

    but very field dependent…in physical science/engineering, anything less than 6 grad students is small (madness!!!).


  24. drugmonkey Says:

    People are the majority cost in the vast majority of labs. Variations in other costs are comparatively minor and to the tune of 1-2 salaries. Save in really extremely distinct cases.

    This is close enough for broad discussion purposes about how research support should be allocated. And particularly for the disconnect between these discussions and concepts of average, expected, desired (necessary?) lab size.


  25. Noncoding Arenay Says:

    @CPP – well, like some of the other answers here that would depend on the species of poultry and vary from one type to another.


  26. Newbie PI Says:

    My lab is 8 people (including me) and I think we are small. 3 grad students, 3 undergrads, 1 staff scientist. One grad student has a fellowship and one is paid by her program through TA fellowships. The undergrads either work for credit, get research scholarships, or volunteer. That leaves me to cover part of my salary and the salary for one student and the staff scientist, and to buy lab supplies. Totally do-able with 1 Ro1.


  27. jmz4gtu Says:

    Wasn’t there a study that showed that, in terms of papers published*IF, 2-3 R01 labs were the most productive/dollar?
    That seems to square with what most people here are calling a medium lab: about 500-750k in directs minus the PI salary, which translates to about 6-10 people. I’m curious as to where the added boost in productivity comes in. My guess would be long-term technical support.

    My ideal hypothetical lab would be 2 PDs, 2GS, 2 technicians, and a lab manager/super tech. If I figure 70k/PD, 45k/GS-Tech, and 90k for the lab manager, that’s about 400k in salary costs. Toss in 75k in cage costs, probably 10k all told for worm, and 50k for cell culture stuff and that’s about 150k for reagents. So I’d need 3 concurrent modular R01s which….doesn’t seem likely. Looks like I’ll have to scale back my ambitions (also, start a fantasy lab league, maybe under the guise of a trainee program).

    Related random question, does anyone know if undergrads working in the lab qualifies for work/study financial aid subsidies?


  28. Rheophile Says:

    To me, small = 1-2 grad students + maybe a postdoc, medium = 5-6 trainees total, large = 10+. This may be because in my (computational) field, one grad student + PI can complete a reasonable project on their own, so the cutoff size where a group starts to struggle is very different.

    jmz: apparently this varies from university to university, there was a discussion on Small Pond Science a while back: http://smallpondscience.com/2014/10/15/does-your-campus-allow-federal-work-study-awards-for-undergraduate-research/


  29. David Says:

    @jmz4gtu – All my workstudy dollars as an undergrad were in research labs (8 semesters worth). Was very educational; much better than working in the library and just doing homework.

    My two cents, coming from an engineering background, a large lab has multiple PIs (i.e. tenure professors), each PI has multiple students, plus full time tech support.

    Medium lab is 4-6 graduate students/post docs (granted post docs are fairly rare in engineering IME).

    Small lab is 2-3 graduate students


  30. Newbie Says:

    Field limitation, instrument limitation (NRM, LC/GC-MS etc. is worth a lot of postdocs) , small is PI and up to 3 others, mid is anything up to PI + 10 members, Big is 10+, BSD is 10+ lab + satellite/proxy/umbrella labs.

    My personal favorite medium sized lab idea is to have 2 PIs run a lab together. I have seen this rarely done, but when done is often productive and a nice place to work. Would be a nice NIH experiment to try to fund this kind of endeavor?


  31. drugmonkey Says:

    Why does NIH have to be involved in that? That’s local U beeswax


  32. Dave Says:

    My personal favorite medium sized lab idea is to have 2 PIs run a lab together. I have seen this rarely done, but when done is often productive and a nice place to work. Would be a nice NIH experiment to try to fund this kind of endeavor?



  33. Newbie Says:

    ’cause if the NIH doesn’t specifically support, Dave’s comment kills any other attempt unless already established (see Brown/Goldstein)

    Other people’s thoughts:


  34. qaz Says:

    The most successful really big labs I know of are the ones run by married couples. (Think the Mosers, who run a very big lab.) There are several such cases in my field, where a married couple each has their own R01s, but they pool lab space and often co-advise students. This allows them to run a big lab (4 R01) with lots of GS/PD (12+) and shared techs.

    It also allows them to play good cop/bad cop…. 🙂


  35. AcademicLurker Says:

    I went from a small lab (2 postdocs, 2 students) as a grad student to a huge lab (21 postdocs, 4 students) as a postdoc. My own lab has always been small, and I prefer it that way.


  36. Anon Says:

    I prefer a functional definition. Small labs have PIs that still do benchwork. Medium labs are where the PD/GS receive most of their training/advising directly from the PI (no RAP or SS in between). Large labs have some sort of management hierarchy.

    This excludes labs where the PI is mostly absent (and has another job in administration, etc.) and has a RAP appointed to run his/her lab. That is a small/medium sized lab in disguise…


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: