Who is responsible for postdoc education? This was the focus of intense discussion at the first Postdoc Network national meeting . During a session for faculty and administrators, half of those in the room raised their hand when asked if their institution had a formal office assigned to postdoc affairs. The other half--some 30 people--"are trying to sort out what their options are," says Joan Lakoski, professor of pharmacology and anesthesiology at Pennsylvania State University College of Medicine and co-chair of the Committee on Postdoctoral Fellows  at Penn State.
During the discussion, it became "really clear that there was no general consensus as to how postdoctoral education should be handled," according to Sharon Milgram, associate professor of cell and molecular physiology, faculty advisor to the postdoctoral association at the University of North Carolina, Chapel Hill, and director of the UNC Postdoctoral Initiative. Milgram was surprised "that it's taken so long for academia to pay attention to this. Many schools are just beginning to deal with postdoc education, and nobody knows what anyone else is doing."
Milgram and Lakoski worked together with the Next Wave staff to put together an informal poll to learn more about the organizational structures that already exist for postdocs at various institutions. It was clear from the meeting that information was lacking to assist those wishing to develop a response to postdoc needs and that those who had done so were unable to compare their progress to that of others. It was hoped that this survey would prove useful to those institutions just beginning to figure out how to address postdoc needs. "There's no reason to recreate the wheel," Milgram says.
Administrators who are running offices for postdocs often experience roadblocks while trying to negotiate additional funding or other resources to support postdoctoral education. According to Milgram it can be very helpful during negotiations to know what other schools--i.e., one's competitors--are doing.
The poll was sent via e-mail to the more than 110 attendees of the Postdoc Network national meeting, approximately 350 attendees of the previous day's COSEPUP convocation , and attendees of a recent GREAT Group  meeting. The total number of unique recipients is unknown because many people attended more than one meeting. Useable responses were received from 70 institutions in the U.S. and one in Canada. Thirty-six responses came from public universities (56% with a medical school), 18 from private universities or colleges (89% with a medical college), 9 from private, independent institutes, and 8 from government research facilities or national laboratories (i.e., government-funded, university-run laboratories). University and college responses include 14 of the top 25 academic institutions with the largest number of postdocs in 1998, according to the COSEPUP  report.
It is unclear how many universities with postdocs on campus never received the poll or how many failed to respond because there was nobody available to even answer the questions (persons who received the poll--many of whom were postdocs themselves--were asked to forward it to the appropriate administrator or faculty). Many of the percentages presented in Tables 1  and 2  regarding administrative oversight and the existence of postdoc offices may be underestimates since the persons who responded to the poll are more likely to represent institutions where somebody has been specifically assigned the task of directing or overseeing postdoc education, according to Lakoski. "Perhaps individuals who didn't respond to the survey don't yet have a mechanism for responding or addressing these issues," Lakoski says.
The results of the poll reveal substantial variation in how institutions handle postdoc affairs--not just because of the varying natures of the institutions, but because "we're all at different stages, doing different things," Lakoski says.
For example, among both public and private universities, the graduate school dean or an associate dean is the person most likely (37%) to be in charge of administrative oversight of postdocs ( Table 1 ). But 28% of all responding universities reported that they had no one designated at the institutional level to assist postdocs ( Table 3 ). Only nine universities out of the whole sample reported the person in charge as someone with the word "postdoctoral" in their official title and, in many cases, the responsibilities of this position are unclear. For example, one respondent identified as "Director of Postdoctoral Studies" remarked "I am temporarily responsible, but it is more in my capacity to survey the postdocs rather than serve a long-term commitment." At independent institutes and government laboratories a variety of administrators with titles including "Postdoctoral Fellows Advisor," "Divisional Administrator," and "Deputy Director for Science and Technology" assume the role of administrative oversight.
No matter who oversees postdoc concerns, says Milgram, it is important that she or he be a fairly powerful senior administrator who can implement change and somebody who the postdocs feel comfortable approaching. Deans and associate deans have that power and, as Lakoski points out, are in charge by default if there is nobody else specifically assigned the task. But most Deans and associate deans are very busy people with full plates. Lakoski and Milgram agree that it's very important that somebody be specifically assigned oversight of postdoc affairs.
One feature of organizational structure that Milgram, Lakoski, and Next Wave staff were most curious about is where on campus the postdoc office is housed. In public universities, 61% of postdoc offices are in the graduate college ( Table 1 ). "Many institutions automatically have placed oversight for postdocs in the graduate school," Lakoski says, "because graduate programs are closely linked to postdoc training, and the grad offices are already set up. But when you're a postdoc, you don't want to be treated like a graduate student." Postdocs are professionals with different sets of issues and concerns. Their career development needs, for example, are different than those of graduate students. Although the grad school could provide career development support for postdocs, a faculty career development program might be more appropriate, and postdocs might be better off in the research arm of the university, says Milgram.
In contrast to public universities, at private universities and colleges the majority of postdoc offices--75%--are housed in the medical school, not the grad school. Still, this can be problematic, says Milgram. Unless the medical school is working together with the college, the concerns of postdocs who working on other parts of campus, i.e., outside the medical school, can be easily overlooked.
No matter where on campus it's housed, once a postdoc office is established, it needs a budget. The sources of funding and approximate annual budgets are reported in Table 1  and Table 2 , but it's not at all clear from the data how the budgets are apportioned. Several respondents remarked that their entire budget went toward salary. One office with an annual budget of $80,000 reported this barely covered staffing costs and that they divert money from endowments to defray the costs of programming. Several respondents indicated that money was allocated specifically for postdoc education programs. At one institution, for example, the budget went toward funding two postdoc research fellowships per year for a 2-year period each. At another, the budget includes institutional support for postdocs to travel to scientific meetings.
The poll did not specifically ask about the amount of postdoc association funding. Nonetheless, several respondents provided details of their postdoc association budgets. Several associations reported having only enough money for monthly "pizza and beer" gatherings, whereas others are able to bring in several seminar speakers per year. The information shared showed an annual budget range from $1000 to $10,000.
The institutional and postdoc culture is different in private and government laboratories compared to most academic settings and likely warrants a different approach to handling postdoc affairs. According to the poll, most private institutions and government labs do not have postdoc offices and have no plans to create one ( Table 1 ). One respondent from an independent institute said that they are small enough that "everyone naturally comes together without the need of an official program, association, or office." In this case, there is an administrator responsible for postdoc affairs even though that person isn't housed in a postdoc office per se.
Like good science, this poll raises more questions than it answers. For example, although postdocs may "naturally come together" at a smaller institute where there are only 30 or 40 postdocs, what about the larger institutions that accommodate as many as a 1000 or more postdocs, who presumably don't naturally come together? And what about institutions, like liberal arts colleges , where there may only be a handful of postdocs?
As another example, at nearly half the institutions with postdoc offices (13 of 29), the person heading the office is not the same person who is responsible for administrative oversight of postdocs. In these situations, how do the roles of the person with administrative oversight and the head of the postdoc office differ? The advantages and disadvantages of these various administrative arrangements remain unclear, says Lakoski. Milgram expects that this split administrative structure may work well. As faculty advisor to the postdoc association at UNC, Milgram interacts daily with postdocs to facilitate communication between postdocs and the administration, assist postdocs with program development, etc. But she reports to the dean of the College of Medicine and the dean of the Graduate School, both of whom have power to implement changes in policy that affect postdocs.
Finally, universities with medical schools seem to have a better handle on administrative oversight of postdocs than do universities without medical schools. For example, 51% percent of all universities with medical schools have postdoc offices, compared to only 27% of universities without medical schools. But how well do these offices cater to the needs of postdocs who are working on other parts of campus?
These results are a preliminary "snapshot of where institutions are in the process of making things better for postdocs," says Lakoski. "The key phrase is in transition. Some institutions are further along in formalizing services than others." By exchanging this kind of information, institutions can begin to learn from the lessons of others.