When two heads are better than one: strategies for successful collaborations

Today we had a panel with three professors at Penn State University (Dr. S, Dr. D and Dr. R) who talked to us for two hours about successful collaborations, driven by questions from the audience. All three professors have appointments in kinesiology, with Dr. D focusing on obstetrics and gynecology, Dr. R focusing on gait and skeletal muscle function and Dr. S investigating neural mechanisms of multi-joint movements.

“Alone we can do so little; together we can do so much” – Helen Keller

“L;enfer c’est les autres (Hell is other people)” – Jean-Paul Sartre

“Tell us about a collaboration you’ve entered into – either one that’s gone really well or one that’s gone south”
Dr D: I’ve never had one that’s failed. However, some have pushed the limits. My training is in exercise & sport psychology, and clinical psychology and I was also a college athlete.  My research focuses on specialized populations, specifically pregnant and post-partum women. From day 1, if I was going to be successful I had to know what I was looking at as a researcher, working together with clinical faculty – all of my work is collaborative in nature. With this, we have to appreciate and understand different expectations and work load between academic and clinical positions.

I would also say, don’t just work with those you need to work with, work with those you like! Sometimes there might be professional differences which can make working out agreements challenging and force you to deal with different mindsets. This pushes me to be more appreciative of a boarder perspective and taught me patience. I think by working with peers, colleagues and other to form mentors, you get access to broad populations to work with. Use your connections in your field, you never know when they might recirculate – don’t burn bridges. That’s key in building effective collaborations. Pulling the plug may be permanent –  understand the consequences.”

Dr. S: “You can classify collaborations in different ways. Someone on campus, which is more contractual. That’s not really what we’re talking about. Collaborationss are those things are 2 PIs working together on big funded projects. I’ve had one that’s been going on for a decade, we went through a number of cycles of R1 and other awards originating at each place of work. She has a different perspective; we did most data collection at her location as we had easier access to specific populations.

Collaborations are necessary to do some things. There is nothing intrinsic about a collaboration that’s desirable just because it’s a collaboration – it’s an inefficient way to do things! You have to make up a bunch of rules – some are concrete, how are you going to deal with publications, who’s going to be last author etc. You need to communicate; we would have co-lab meetings. Sometimes a lot of that communication isn’t necessary if you’re not separate.

My view is collaborations allow you to do things you couldn’t do without them, without specialized skills and knowledge. If you enter into a collaboration – it costs more in time and in effort. There are personalities, no matter how well you get on with someone, there can still be issues. You spend a lot of time doing mundane things – how you are going to transfer data is a common one. But they can be important if you need the collaboration. Before you enter into it, ask yourself if you can do it alone? If you can – do it! Otherwise, plan everything out in advance, make sure you get along. Arguing isn’t necessarily a bad thing.”


“If a junior faculty member has a good idea, and they could do it themselves, should they do it or enter into a collaboration?”
Dr. S: “DO IT YOURSELF! If you can get publications, lead a project and so you, you then have something to offer a collaborator. If you’re just starting out and collaborate, you can get pulled in lots of different directions. Little contract things are acceptable, but careful with multiple PI situations.”
Dr. D: “I think it’s a multi-pronged approach. As grad students, life is inherently collaborative – you have a mentor who’s published with you. When you move, the data you’re moving with is owned by your graduate university. For the first few years in the new job, you’re still allowing mentorship relationship to be productive but start to separate yourself and establish yourself as an independent researcher. While maintaining that mentorship relationship, you also need another pocket of brand new independent work at your new location. You need a multi-pronged approach, understand what you bring to a collaboration, what can flourish in initial collaborations & what your new line of research is in the transitional period. I was building a lab, own students, publishing PhD work, and building own work. You need to be multi-faceted – your own work & collaborations”

Dr. S: “I think this is field dependent. You gain some skills as a grad student. As a postdoc you use those skills, you might pick up a few more but you’re primarily developing skills and direction. In a postdoc, you breakout into your own direction. My opinion is when you realize you need a collaboration, think of it as a business arrangement, have something to offer or start out equally unskilled & unknowledgeable. Both approaches are what Dr. D is describing – collaborating. Part of this preference is to do with personality!”


“Wouldn’t you say that a grant with a single PI sticks out? I looked at what NSF are funding – there’s grants with 3-5 PIs, I don’t think I saw one funding in a single PI. Is it odd to submit a grant with one PI and no co-I’s?”

Dr. S: It’s odd not to have co-I’s but that’s not necessarily a full-on collaboration. That might be just consulting work. It’s odd to have 1 PI and nothing else but it does happen. There’s usually someone for stats, some clinical person. The Co-PI thing is rare.”

Dr. D: “This again I think is discipline specific. It would be odd to have a single PI. Submitting to NIH – one of the scoring criteria is your team.”


“You’ve mentioned pulling in someone from outside. How do you identify someone to collaborate with?”

Dr. S: “I had an idea of what I wanted to study, I read background research, found papers that I liked and called the author who ended up recommending someone else. Usually you have an idea and you know the other people in the field who think your idea might be good.“

Dr. D: “My approach is different. Depending on the environment where your PhD is – I came out of a lab where you represent the lab & it was positive in pushing students forward. They gave plenty of opportunities for authorship. This has built me a network of colleagues I trust, so if I was looking for collaborations I would rely on my own senior mentors as well as other peers. At this point in my career, I can allow some of those collaborations to be strategic.

At some point, you’ll be bombarded by people who want your time, attention and collaboration – be aware to not over-load yourself. Your field becomes narrow – in my area, there’s maybe 10 people globally who do what we do. It’s very narrow! You get to know these people through your network, knowing who you can rely on and trust in the field and who would be suitable to collaborate with.”

Dr. S: “I think you need to be aware that there’s predatory behavior in senior faculty. That’s why I’m saying its good to have something to offer. Be careful to not get too one-sided in your relationships. There are senior faculty will take advantage of people coming in and not think what’s best for those people’s careers:”

Dr. R: “You’ve got to go in with a savvy mind. I’ve worked with several collaborations & the most impactful one was a PI who was aggressive & you had to push back. But it ended up being very productive. I was an assistant professor at the time, maybe a bit vulnerable. It became an understanding that he opened up avenues that weren’t attainable at the time, and I was providing publications”

Dr. S: “Essentially two people using each other – and understanding how they’re using each other.”

Dr. R: “We weren’t friends but it was a very productive collaboration. Of course it’s more fun if you get on with them but it’s not always necessary and this collaboration led to some really great work.”

Dr. S: “It’s a business arrangement. If you need what they have, and you have something to offer what they need, then I suppose you should work with them regardless of personality as long as you can be productive.”

Dr. R: “We had extremely good clinical links, a pool of subjects you could recruit from. It would have taken a career to build up the environment, but this collaboration enabled me to use this. It was a hierarchical system as well. It is a lot more fun if you get on with your collaborators but it can work otherwise”

Dr. D: “I think the set of clear expectations at the start is key. Before you enter any collaboration, what are you both bringing to the table? What are you producing? What are your expectations? Even with the greatest laid plans, when things change, how do you continue to work through when expectations change? How do you navigate the muddy water?”

Dr. S: “What Dr. D is saying, there’s a lot of work time and effort to go into maintaining collaborations.”


“What point do you get into a collaboration? When do you decide to back out?”

 Dr. S: “Once you enter into something, you’re going to see that person at conferences. You don’t want to enter into anything lightly. There’s grants, authorship & papers – that’s what we spend all this time on. That’s what you have to negotiate, you really have to work these things out. What responsibilities are, how you’re splitting grants. That grant has to go through one center, and that determines who takes the leadership role.”

Dr. D: “To add to that, I haven’t burnt a bridge and I’m proud of that. I’ve had collaborations where I’ve been asked to be Co-PI and nothing came to fruition. Sometimes you have to respectfully decline & not go there again. Give everyone a good first effort, if that first effort doesn’t work out well then I don’t do that again. I’m careful to not roll that coin again.”


“Do most collaborations start at the grant level or how are they born?”
Dr. S: “I needed someone to recruit patients  & do the neuro-psych aspect. We didn’t write a grant off the bat. We put something concrete into it. Then we started to collect data & then pretty quickly work on a grant. I think you need to have resources to be able to sustain collaboration. It’s different when you’re working with something on campus, where you can play with ideas and experiment in your lab. With 2 big groups it has to turn into resources.”

Dr. R: “My most intellectually fun collaborations don’t start off as grants. You have colleagues/friends who share an interest, passionate about what you do, and its fun. Those can sometimes lead to grants, and once they do its important to establish some expectations. You don’t want to fall out then! You’ve had a fun experience and now it’s gotten to a point where you have money. What we did is we had a document, like a contract & guidelines for publication, from a grant with 8 PIs on it. This agreement would start off with everyone on the paper, and then it was left to each PI to decide what study they worked on. We could call out our co-PIs and say they didn’t do anything, you shouldn’t be on the paper. I don’t think that happened ever – it was understood that if you didn’t have any role in that publication other than being on the grant, people stepped back off authorship for that publication.”

Dr. S: “We’re working on a coordinated collaboration. We won’t get a grant unless we’ve shown we can work collaboratively together.”

Dr. D: “I’ve started collaborations that have been about a paper, contribute something and that leads to something collaborative in the future. We can generate a lot of papers and I contribute to the bigger grant. I’ve had other situations where I contribute to grant proposal as a consultant, thru the grant writing process in the meantime we should get a paper out. If it’s successful it’s generated papers in the interim between projects.”

“When someone approaches you with a project, how do you evaluate whether it’s a collaboration you want to work with?”
Dr. S: “You have to respect the person scientifically & believe that they share your ethical perspective. As you go on in science, you build a reputation, you get more requests for collaborations. You have to know your limitations. Especially early on there’s a tendency to run in every direction & that can hurt you. Make sure its something your passionate about & it fits your goal scientifically. If it does, great!”
Dr. D: “What’s their motivation? Is it predatory? Can you read between the lines, what’s their need? Is it mentoring for you? What are you going to get out of it your end – if it’s only one paper but a new positive collaboration is it worth it? The more you go in understanding expectations & cost-benefit, you make a strategic decision.”

Another professor from the audience: “Sometimes people approach because you have something they want, but their question isn’t interesting to you. That doesn’t mean you can’t work together, you could create a quid-pro-quo but that’s not a collaboration. There has to be some interest in the same question.”

Dr. D: “Additionally ask some questions. Where’s the proof? I’ve been working with so and so, where is that on the CV? If you’re investing lots of time that’s not resulting in an outcome, maybe intellectually that’s great. In the early part of your career, you should be focused on the productive pathways for you. Later on in your career, then you can do a few more of those intellectually stimulating things that have less productive outcomes. We can maybe feed those things then. Think about how many of those collaborations you have and what the outcomes are for that stage in your career.”


“Have you been part of something you didn’t like how it turned out? The end-product isn’t something you want your name associated with? How do you explain it on your CV?”

 Dr. R: “I’ve taken my name off because I felt I couldn’t contribute. I’ve been in an awkward collaboration before, I was close to saying I’ve had enough & things didn’t turn out so well on that particular project. In the acknowledgements, they said I said I cleaned the computer screen! They diminished my role in the paper that came out.”

Dr. S: “I had a postdoc who was with me a while. We had an argument about the interpretation of one of our findings. You are your faculty’s children in a way, and we really feel very passionate about our research. This thing happens when you’re a grad student where you move out the house, there’s a bit of friction about who gets to do what. I said to our post-doc, if you do this, take my name off the paper. I was being a baby about it, I was an idiot! A few months later, I see the paper without my name on it. I thought that’s not what I meant! I have never said take my name off it since! That was stupid of me. It’s difficult to transition from the personal mentor-mentee relationship, it is like your kids growing up.”

Dr. D: “ I have not pulled my name off but I will. I think more important than the cut is navigating those conversations in how you express with what you’re unhappy with & leading you to speak up. I’ve had incidences where I’ve had to have those difficult conversations. I was clear saying I will support this endeavor, if the grant is funded we will have a very clear discussion on how this will work. I was clear – communication is key. Going forward here are the expectations.”

Dr. R: “We’re talking about us as PIs on collaborations and pulling out of papers. In your early career, as a grad student, you’re much more vulnerable. Maybe you collaborate with a friend, it’s important to establish some guidelines & expectations. All of a sudden you can be in a position where their PI asks what did they do.”
Dr. S: “Really good point. When I was a grad student, we made our data collection apparatus. I had constructed some things & written some code. I did it for my work, some others in our lab used it. I didn’t get even an acknowledgement. That’s something I’m sensitive to in my lab, and if someone’s contributing they have to be an author. You have to know your mentor. Bring this up!” 

Dr. R: “If you’re spending a lot of time on it, you should get authorship.”

“People who do essentially the same work, bringing same skills – they’re just really good friends and want to work together. Is there anything wrong with working together just because it’s fun even if no one is bringing anything new to to the table?” 

Dr. S: “I’ve been toying for 20 years with a project to do with a friend, and we haven’t figured it out, so we just drink together. Here, we were out to lunch with a collaborator we have different theoretical approaches and that turns into a collaboration. If you don’t have something substantial from 2 different places to turn into a collaboration it can be difficult.”

Dr. D: “In our field, a meta-review, stats analysis we write with colleagues who have overlap and respect. Right now, I’m working with a colleague because we like each other and get along. This is a great way to produce something relevant – it’s a book chapter. Those are a little bit more selective”
Dr. S: “Same with conference organization.”

Dr. D: “When you’re in charge of a conference and you need people to help you, your collaborators and their students are the ones you want to be pooling from.”


“When you have collaborators how do you figure out authorship?”
Dr. D: “I think we do all. If there’s a clear distinction between areas, it’s clear. Otherwise right from the start, we make a manuscript plan. Lay out expectations clearly, how do students fit in, who’s getting first author etc.”

Professor in audience: “First author is primary responsibility for work. Last author is the advisor. I know this varies by field.”
Dr. R: “I think for the most part, it’s pretty clear. There’s not disagreements who go first. It’s the students typically who go first. Mainly the discussion comes with who goes last, I think it’s clear but if it’s not, you can either rotate, or we use two corresponding authors. Stating they’ve done equal work or some other concession.”
Dr. D: “It’s definitely field dependent. For many of you who will go into assistant professor positions, if you’re going thru promotion and tenure process. There are different ways of looking to being first author on your mentor’s papers, and then switching to last author. It’s not that fast, but then you need to be first author on your own papers. Maybe you don’t have a grad student who’s fully trained, or you work fully alone. As you build a lab and develop a deeper mentoring relationship, as you’re growing, you primarily build your senior author papers.  One of the feedback we give junior faculty reviewing their CV’s, where’s the first authorship on your work? You go from first author on your mentor’s work to being senior author on your own work.”

Dr. R: “I think that’s definitely field specific. I didn’t feel that. I started with an honors student and they need to go first. There’s times where I’ve had instances where a student left, and I finished doing the papers and I felt awkward going first because I didn’t want people thinking I’d stolen the paper from my student!”

Dr. S: “All things being equal, going back to collaborations, you should rotate being last author. Usually the trainee or student is working very closely with one of the senior authors. Normally they will be last author.”


“How much of that transition you talked about happens as a post-doc?”

Dr. S: “You’re probably first author. But if you’re a post-doc and you mentor a student, then you should be senior author. But usually it’s when you start striking out on your own.”

Dr. D: “Often as a post-doc you’re still first author.”

Professor in audience: “It’s fair to say you don’t want a bunch of middle-authorships when you go up for promotion and tenure.”

Dr. D: “You need that transition separating from your mentor, showing your independence and then showing your mentorship.”

Dr. S: “There’s nothing wrong with having middle authorships as long as you still show your transition.”


Thanks to our professors for this talk on collaborations, it raised some very interesting points to consider.


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 )

Google+ photo

You are commenting using your Google+ 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