No man or woman is an island. This is especially true in research.

The scientific community is like one big, extended family, full of close and distant connections, crazy old aunts, pontificating father figures, and distant cousins you've never met whose names you can't remember. But a small group of you is on an island together, forming the most basic social unit of scientific research, the research group. Until you get your degree (unless you decide to bail out), you're stranded with this group of people, like them or not.

If you haven't chosen a research group yet, it can be helpful to consider which type of group you're likely to be most comfortable in. Sure, you need to choose a group that does the science you want to do--hopefully you've given that a lot of thought already--but factors such as the size and chemistry of the group are also important. Even if you've already chosen a group, thinking critically about its dynamics can be beneficial.

Your group is probably a variation of one of the archetypes discussed below.

The start-up group

Full of exciting new ideas, a starting professor often operates more like a more senior partner than a boss. On the downside, there is the pressure to get tenured--and management and interpersonal skills don’t come naturally to everyone.

If your Ph.D. adviser is just as new as you are, or nearly so, you can expect him to be full of energy and ideas and eager for data, which you and a colleague or two will have to acquire. Young thesis advisers have the tendency to design overly ambitious research programs, and new professors are rarely quite sure which aspect of their research is going to bear fruit. So expect some course corrections, which may cost you time.

Most--indeed, almost all--assistant professors lack experience supervising students, but what they lack in experience they make up for in time and intensity. Your young adviser probably has fewer teaching and administrative duties, and rather few students, so you can expect frequent, intensive interactions. The bad news is the reason for those frequent meetings: Your adviser needs that data desperately, and that means pressure.

In such a setting, it is crucial that you get along well with your adviser, so choose well and invest in the relationship. If you don't get along with your adviser, your life in the lab is bound to be rocky.

In a start-up group, there are probably just one or two other Ph.D. students or postdocs, so the success of your projects will be intertwined. Lacking the infrastructure of an existing group, all of you probably will spend a lot of energy building equipment, designing new models, or writing new computer codes; that's time lost from collecting data. This, together with the fact that young groups often lack established rules for whose name gets on the paper (for example), can cause intense disagreements. So make fair arrangements with your colleagues about how to share the bounty when it is time to harvest data. And when conflicts arise, sort them out.

How can you tell whether a new group leader, who probably has no track record, has his act together? Think about your job interview as a two-way selection process; interview your boss. If you can, find out how the professor functioned as a postdoc. Was he already acting as a professor-to-be, or was he still working like a Ph.D. student? Finally, go with your gut: Does he inspire confidence? Does he make you want to work for him?

The up-and-running group

Around the time an academic scientist has delivered her first Ph.D. graduates out into the world, she is usually tenured and promoted to associate professor--or fired. Assuming it's the former and not the latter, the research of the new associate professor has made some impact on the scientific community, grant money is easier to come by, and the group is able to expand. The PI's job security, the availability of more resources, and--usually--the larger size of a group like this make it a very different place to work than the lab of an up-and-comer.

It's easier to do research in a lab with more resources and fewer insecurities, but most established scientists spend less time in the lab and more time traveling to conferences and invited lectures, in grant-review meetings, and on other activities. Then there are the committee meetings and other institutional commitments. How well the associate professor copes with these new burdens will depend on her organizational skills--and will determine, to an extent, how good a place her lab is for a graduate student to work.

In a group like this, guidance from the adviser is less frequent than in a start-up group. Accept this reduced interaction with your adviser as a fact of life; the ability to work well independently is a key career skill, so develop it. Discuss how the two of you can have effective interactions when you do get together. Prepare for those meetings by talking things over with your lab mates. Turn to the lab's senior graduate students and postdocs for help, but be sure to offer something in return. Labs like this thrive on lots of small collaborations, so learn to work well with your mates.

If you're sure you want to stay in academia, an up-and-running group is a good choice, because your adviser will be better able to help you find an excellent postdoc position--a necessity for a good academic job--than will a brand-new assistant professor.

The small-but-established group

After several years as a successful associate professor, academic scientists usually are promoted to full professor. At this stage, some professors feel they are finally able to relax after many years of hard work. They may become more interested in the administrative aspects of running a research group or a department, and their interest in research may start to fade. They have enough experience to keep a small group going and still periodically have decent results to publish, even working part-time on research. But in a group like this, you may have to work extra-hard to generate enthusiasm for your project. Although such a group can get you a degree--a valuable asset in your future career outside academia--it is probably not the best place to start an academic career.

Please note, however, that not all small, established groups fit the description above. Plenty of full professors enjoy doing science so much that they want to remain deeply involved in hands-on research, which is impossible with a large group. So they choose to focus their attention on a small laboratory with only a couple of Ph.D. students and maybe a postdoc. These small enclaves of pure, intensive research can be wonderful and stimulating places for graduate work. Assuming you get along well with the professor and your fellow Ph.D. students, you will thrive in an environment like this.

The empire

Some successful professors choose the opposite course. Instead of keeping a hand in the research, they allow their group to expand and take on an administrative role. Scientific emperors still care deeply about research, but the size of the lab keeps them at arm's length.Such groups can easily have 10 to 20 Ph.D. students--or more--along with several postdocs. Not every research topic will be a winner, and some will fail altogether. But the availability of sophisticated instrumentation and a vast skill base should enable you to acquire data quickly. And if your project fails, there are other projects to fall back on.

If you choose a lab like this, don't expect to see your adviser very often. More than in any other group, your interactions with your peers will determine your success--and how much pleasure you get from your Ph.D. research. Because you have access to a whole army of young researchers, you have the luxury of finding a few with whom you click. That's probably the key to succeeding in an empire: You need a few good friends.

Unfortunately, some empires have a cutthroat atmosphere. With several Ph.D. students finishing their degrees in a given year--some probably even applying for the same faculty jobs, all seeking letters from the same adviser--the jockeying can be intense.

Before you commit to working in a lab like this, talk to some of the students in the lab. Try to find out what it's like to work there. A little competitiveness can be an excellent thing, but too much can produce an environment that's unhealthy for science--and scientists.

In some European countries--Germany, for instance--full professors all seem to run such empires. But what looks like an empire is often instead a loose confederation of smaller labs, each run by a lieutenant. The dynamics of labs like these usually resemble those of a start-up or an up-and-running group.

Surviving in a nonsupportive group

The relationships you establish within your research group, with your colleagues and adviser, will have a big impact on how well you perform during--and after--your Ph.D. project. So find a group that does the kind of science you're interested in--but also one that fits your personality.

Unfortunately, that's not an easy determination to make. Much depends not just on the size of the lab but on the particular people in it--and, especially, the person who runs it. And even if you make a good choice, there are likely to be times when you struggle with group chemistry, or when you don't feel you're getting the support you need. Here's how to get the most out of such a situation:

  • Think positively. Focus on the support that is available rather than sitting in isolation, frustrated about the support you're not getting or blaming those who should be helping you. Use this time as an opportunity to develop some independence.

  • Find the help you need. No member of the group, including your adviser, can solve all your problems and fulfil all your needs. Some colleagues and advisers are better at designing new projects, others at debugging computer codes, still others at editing manuscripts. So search around, and don't limit yourself to your research group; you might be, metaphorically, stranded on an island with these folks, but the island has a phone and even a fast Internet connection. Don't share prepublication data without your PI's permission, but find the help you need where it exists, whether it's within or outside.

  • Identify your showstoppers. It’s not necessary to solve all of your problems in order to make progress, at least not all at once. Set priorities and address the most important issues first. If you notice that you're spending a lot of time on problems that seem pressing but, in a careful analysis, don't seem all that important, read our previous column and revise your approach. Focus only on those problems that really stop you from making progress with your project.

If none of the above suggestions works and you envy students working in a seemingly much more productive and pleasant group, it might cross your mind to change labs. This is a delicate issue, so double-check how much greener the pastures really are over there--is it really worth it to change? Identify the procedures at your university--is changing labs commonplace, or are you considering something that's unheard-of? Good diplomacy will improve your odds of changing labs successfully.

The research group you choose--and how skillful you are at working within it--will determine, to a large extent, how successful you will be in graduate school and beyond. So evaluate dynamics, identify allies, and put some effort into key relationships.

Comments, suggestions? Please send your feedback to our editor.

Bart Noordam is a professor of physics at the University of Amsterdam, the Netherlands, and director of a regional audit organization. He has also worked for McKinsey & Co.

Patricia Gosling and Bart Noordam are the authors of Mastering Your Ph.D.: Survival and Success in the Doctoral Years and Beyond (Springer, 2006). Gosling is a senior medical writer at CMPMedica in Malaysia and a freelance science writer.

10.1126/science.caredit.a0700010