The Next Generation of Scientists: Examining the Experiences of Graduate Students in Network-Level Social-Ecological Science

By integrating the research and resources of hundreds of scientists from dozens of institutions, network-level science is fast becoming one scientific model of choice to address complex problems. In the pursuit to confront pressing environmental issues such as climate change, many scientists, practitioners, policy makers, and institutions are promoting network-level research that integrates the social and ecological sciences. To understand how this scientific trend is unfolding among rising scientists, we examined how graduate students experienced one such emergent social-ecological research initiative, Integrated Science for Society and Environment, within the large-scale, geographically distributed Long Term Ecological Research (LTER) Network. Through workshops, surveys, and interviews, we found that graduate students faced challenges in how they conceptualized and practiced social-ecological research within the LTER Network. We have presented these conceptual challenges at three scales: the individual/project, the LTER site, and the LTER Network. The level of student engagement with and knowledge of the LTER Network was varied, and students faced different institutional, cultural, and logistic barriers to practicing social-ecological research. These types of challenges are unlikely to be unique to LTER graduate students; thus, our findings are relevant to other scientific networks implementing new social-ecological research initiatives.


INTRODUCTION
Many of today's major scientific challenges, e.g., climate change, atmospheric modeling, epidemiology, genome sequencing, and high-energy physics, are inadequately addressed by a single research team at a single institution with a single disciplinary focus (Wilson 1999, Ewel 2001, Díaz et al. 2011. Complex research questions often require a collaborative approach that reaches across disciplines, which is becoming an increasingly prevalent scientific model (Nature Editorial Board 2003, Norgaard and Baer 2005, Carpenter et al. 2007. As a result, scientific endeavors are often carried out by organizations of researchers spanning dozens to hundreds of geographic sites. Network-level science is one term used to describe this mode of large-scale, geographically dispersed, integrative research made possible by connecting researchers, data sets, and resources to solve a common set of scientific problems (Karasti andBaker 2008, Porter et al. 2009).
Network-level science has proved effective in scientific investigations that involve collection, synthesis, and analysis of massive amounts of data. This approach is also needed when the range of phenomena under examination requires a comparable range of academic expertise, as in contemporary ecology, where research on the complex social-ecological drivers of regional and global environmental change necessitates collaborations across the biophysical and social sciences Baer 2005, Carpenter et al. 2007). As such, several large-scale distributed scientific networks, e.g., the Resilience Alliance, the National Ecological Observatory Network, and the Long Term Ecological Research (LTER) Network, have been established to conduct social-ecological research across broad spatial and temporal scales. However, not all scientific questions necessitate a network of researchers undertaking massive data collection and analysis, and scholars have long raised concerns about how the inclusion of multiple perspectives in research may impact the scientific process (Shapin and Schaffer 1985). Despite this, few studies critically examine the ways in which scientific networks operate. Existing research largely focuses on logistic and communicative issues that arise when connecting researchers with different institutional, cultural, disciplinary, and methodological backgrounds (Heemskerk et al. 2003, Rhoten andParker 2004).
Network-level science is often advanced by early career researchers, who are typically products of innovative and successful graduate programs (Graybill and Shandas 2010). Historians and sociologists of science have shown that graduate students and research assistants are profoundly influential in how science is produced and distributed across geographic and disciplinary boundaries (Shapin andSchaffer 1985, Galison 1997). Scientists have also acknowledged myriad challenges, such as differences in theories, methodologies, and overall values, associated with collaborative team research across disciplines and suggest the need for continuous dialogue and programmatic restructuring to address issues as they arise (Nature Editorial Board 2003, Lélé and. Graduate students make important contributions to ongoing debates about interdisciplinary Our research traced how LTER graduate students interpreted, put into practice, and reformulated the novel ideas put forward through the ISSE initiative. We examined how graduate students conceptualized and practiced social-ecological research at three scales: the individual/project, the LTER site, and the LTER Network levels. Although the results are highly relevant for members of other research networks that seek to integrate social and ecological science, we have explored potential issues for all interdisciplinary, network-level scientific projects. We found that the disparate experiences of LTER graduate students revealed the complex nature of how scientific networks operate and the unique concerns that their members face. Although not representative of network-level science in general or even the LTER Network as a whole, the different ways in which graduate students understood and negotiated what it meant to be networked, social-ecological scientists provided insights into how network-level science operates, how networks can best adopt novel social-ecological research initiatives, and how to build a strong foundation for the future of network-level science.

METHODS
To document the experiences of graduate students with socialecological research in the LTER Network, we used a study design that combined focus group interviewing with survey research and semistructured interviews. A mixed methods approach enabled us to collect study data at multiple scales and to triangulate our data (Fig. 1).

Fig. 1.
In this study we employed three iterative rounds of data collection from surveys to interviews. Analysis of the coded interviews alongside the survey data revealed general trends in responses across different scales in the Long Term Ecological Research (LTER) Network.
We began with a focus group among graduate students, called "Benefits and Barriers to Social-Ecological Research," at the 2009 LTER All Scientists Meeting (ASM) in Estes Park, Colorado. A focus group is a structured or unstructured group discussion about a topic of interest led or observed by researchers (Edmunds 2000, Berg 2007). In the session, 19 graduate students from 6 LTER sites discussed the challenges they faced in carrying out social-ecological research in the LTER Network. It is important to acknowledge that we, the authors, were LTER graduate students who initially participated in the ASM focus group and then subsequently assumed the role of project investigators observing other LTER scientists. We recognize our positions within the LTER Network and make no claims to pure objectivity. Instead, our Ecology and Society 18(3): 42 http://www.ecologyandsociety.org/vol18/iss3/art42/ research represents a kind of "situated knowledge" (Haraway 1988) and is the product of our own embodied and contingent experiences working in and studying the LTER Network.
The focus group proved a useful strategy for gathering preliminary data on LTER graduate students' experiences with social-ecological research, and we used this information to frame specific project research questions and to develop two separate, structured surveys. Each survey contained a combination of closed and open-ended questions and was pretested with volunteers. First, we distributed a broad, 31question online survey to all the graduate student members of the LTER Network's student Listserv (Appendix 1). The LTER Network's student Listserv includes approximately 650 LTER graduate and undergraduate student members. Then, to collect more specific data from students who had conducted social-ecological research, we e-mailed a targeted, 17question survey to 42 graduate students identified by senior LTER scientists as being involved with social-ecological research (Appendix 2). We gathered data from 103 students, 57 fully completed responses, for the broad survey and from 41 respondents, 33 fully completed responses, for the targeted survey. Data for categorical questions were tallied and reported as percentages of the total number of responses.
The broad survey asked questions about the types of research students were conducting, institutional and disciplinary affiliations, familiarity with and attitudes toward socialecological research, and the challenges and opportunities related to the pursuit of social-ecological research. The targeted survey expanded on the broad survey by asking more specific questions about students' definitions of and direct experiences with social-ecological research in the LTER Network, and targeted survey respondents were asked to indicate their willingness to participate in follow-up interviews. Results from both surveys were analyzed to reveal patterns across survey responses. To add greater depth and context to the survey data, we used the findings as a guide to design a semistructured interview protocol (Appendix 3). Interview questions asked graduate students to more specifically describe their engagements with social-ecological research in the LTER Network at the individual/project level, the site level, and the network level. Hour-long semistructured phone interviews were carried out with 10 current or recently graduated students who volunteered to participate and who self-identified as conducting a social-ecological project in the LTER Network. This phased survey and interview approach employed a sampling design with intentional biases. Although we used random sampling for the broad survey, we chose purposive sampling for both the targeted survey and the interviews to locate students who had experiences with socialecological research in the LTER Network.
We used a combination of inductive open coding and deductive coding to analyze our interview data. Using ExpressScribe and Microsoft Word, we transcribed and then coded the interviews into specific categories related to challenges in social-ecological research. In an initial round of coding, we searched for quotes and expressions relevant to preidentified categories such as "funding challenges," "site challenges," "network challenges," "adviser challenges," and "methodological challenges" (see Ryan and Bernard 2003). In a second stage of coding, we used open coding to inductively identify key concepts and categories beyond those included in the original list of categories (Corbin and Strauss 2008) but related to the definitions and challenges of social-ecological research. We constantly compared notes to develop a consistent framework for coding the interviews. Analysis of the coded interviews alongside the survey data revealed general trends in responses across data sources and in experiences across different scales in the LTER Network ( Fig.  1).

RESULTS AND DISCUSSION
Our results demonstrated that graduate student experiences with social-ecological research in the LTER Network were defined by several key conceptual challenges. We discuss how these challenges manifested at three scales in the LTER Network, i.e., the individual/project, the site, and the network levels, and along two dimensions at each scale, i.e., conceptualization and practice. Table 1 provides examples of the types of unstated conceptual challenges that LTER graduate students may experience at these scales.

Conceptualization
Graduate students had different definitions of socialecological research and different understandings of what it means to be a social-ecological researcher in the LTER Network. The LTER Network's ISSE plan asserted that "new research must focus on understanding the long-term dynamic processes that are unique to social-ecological systems versus purely social or purely biophysical systems" (U.S. LTER Network 2007:6). However, the results from our broad survey demonstrated that graduate students struggled to concretely define the term social-ecological. Only 4% of respondents felt that there was an agreed upon definition of social-ecological research within the LTER Network, and only 33% felt that they could clearly define social-ecological research. Indeed, when asked to define social-ecological research in their own words, only 57% of graduate students chose to respond to the question.
Those who did respond offered a general definition of socialecological research as the study of the interactions between human systems and ecosystems. A simple word-frequency analysis (Fig. 2), created using http://www.wordle.net, revealed that the top five terms students used to define socialecological research were research, interaction, human, ecological, and systems. However, graduate students described the interactive relationship between humans and ecosystems in many different ways. Several students characterized social-ecological research as a one-directional research field, emphasizing the study of human impacts on the ecosystem rather than examining humans as part of the ecosystem. Other graduate students characterized socialecological research as a type of outreach or education to the general public, a core aspect of the LTER Network's mission.
In addition to divergent ideas about its definition, we found that graduate students involved in social-ecological research articulated different understandings of what it means to be a social-ecological researcher. In the interviews, students frequently referenced the depth-versus-breadth debate surrounding disciplinary specialization (Gilbert 1998). The depth understanding is multidisciplinarity, which holds that students should primarily identify with a single social or ecological discipline, becoming an expert in the theories and methods of that one particular approach. Students then become social-ecological researchers by applying their disciplinary specialization alongside other researchers in a multidisciplinary setting and learning to communicate with other disciplinary experts. As one interviewee stated: "We should all have expertise ... the disciplinary knowledge is really important, but I think we should have enough background in each other's disciplines to be able to read studies, read papers, evaluate methods ... we each bring different skills to the table and we need to rely on those sometimes to get the actual work done in the best way." In contrast, students with a breadth view were more aligned with interdisciplinarity or transdisciplinarity and did not necessarily identify with or consider themselves experts in a single discipline. They were trained to have a range of knowledge, skills, and expertise that allows them to conduct research that integrates multiple disciplines, either on their own or in collaboration with other researchers. One student described how she was "... trying to frame myself as more of a synthetic person, someone who knows enough about all of these other disciplines that I can piece together the puzzle. Not trying to actually be an ecologist, be a planner, be a geographer, be all of these things ... people tell you that there are going to be challenges in the future, finding jobs, because you need to have a discipline. But I really don't think that's going to be the case."

Practice
The concept of specializing in one field versus gaining broader training in multiple fields was also evident in how socialecological projects were carried out. From our interviews, we found that students who had participated in one or more collaborative social-ecological projects described these projects as falling along a continuum from a siloed, or multidisciplinary, project approach to a holistic, or transdisciplinary, project approach (Eigenbrode et al. 2007). Their narratives pointed to the variety of modes of practice that currently exist for social-ecological research in the LTER Network ( Fig. 1).
Graduate students involved in siloed projects described how disciplinary researchers were seen as independent from each other and were asked to approach a clearly delineated part of a research project from within their single area of expertise. Some interviewees viewed the siloed model as a normal, accepted, and even exemplary approach to social-ecological research, whereas others were critical of it. For example, one student working on a study employing social and ecological methods was critical of her experience with a siloed collaborative process, stating: "Each researcher is bringing their one component to the project that will then somehow be synthesized in the end. There doesn't seem to be much emphasis on having everyone on the same page or having everyone understand everything else." Graduate students involved in holistic projects described how researchers were more involved in each other's activities throughout the project, jointly forming and reformulating http://www.ecologyandsociety.org/vol18/iss3/art42/

Fig. 2. A visual representation of the most frequent terms Long Term Ecological Research (LTER)
graduate students used when asked to define social-ecological research, with the size of the word indicating the frequency of its use.
research questions, as well as collaboratively collecting and analyzing data. For instance, one student praised this holistic approach: "Rather than just doing ecology and then bringing in the social scientists on the side, it's about creating interdisciplinary ideas and methods, and thinking about that from the beginning, so that at the end you have a project that is integrated throughout."

Conceptualization
Graduate students discussed how LTER sites varied in the way they incorporated social-ecological research into the scale and scope of their research, raising questions about traditional LTER site organization. As other scholars have noted, integrated social-ecological projects often face the challenge of scale mismatch (Pickett 1999; that is, the biophysical and social variables are not studied at the same geographic scales. Scale and integration were definite challenges for graduate students working on LTER socialecological research. One interviewee working on an LTERaffiliated social-ecological project with broader boundaries than his official LTER site noted that his interactions with other site personnel and his overall sense of connection to his LTER site were limited because he was not based at the traditional field site during the regular field season. He noted: "We are not terribly well integrated into the LTER. [Our site] tends to think of its research as what goes on 'within the fence' ... Since we are not physically on site during the season the way that most researchers are, it's a little peripheral to that." Another interviewee, who also worked on a social-ecological project with boundaries broader than those of his official LTER site and was not based at the site's research station, observed: "I know that there is a kind of field study house located within [and] maintained by the site [that] is used for doing the very traditional ecological research ... I get the impression that there is a much more closely connected network of researchers out there who work on ecological issues. I don't really have any connection to those folks." The remarks of the interviewees belie a sense of frustration at being in a more marginal position to the main hubs of research activity and intellectual exchange at their LTER sites.

Practice
When considering the practice of social-ecological research at the site level, some graduate students discussed challenges associated with spatial and cultural distance, whereas several others described strong site-level support. Graduate students described varied levels and modes of involvement at their home LTER sites. Although some LTER sites have principal investigators (PIs) strongly linked to one particular university and a field location with specific boundaries, others have PIs dispersed throughout the country and study areas that vary widely in geographic location and scale. In fact, 64% of the graduate student respondents in the broad survey were located at universities greater than 50 miles from their LTER sites.
Students at sites characterized by dispersed researchers, universities, and field locations tended to have greater difficulty in engaging with their site and, therefore, understanding if or how social-ecological research was operating at their site. One student attending a university separated from the LTER site by hundreds of miles had scheduling conflicts. He told us: "The tough part about http://www.ecologyandsociety.org/vol18/iss3/art42/ working with [the LTER site personnel] was that their schedule was based on the semester system and [my university] is based on the quarter system, so we could never go to their meetings." Other students struggled to find support for the social sciences or other social scientists working at their LTER site. One interviewee observed that incorporating the social sciences was an NSF mandate that was not yet generally accepted within his site. He explained that "the renewal of their LTER was really contingent on becoming more transdisciplinary, involving social science. And I think it's been a really challenging transition for them." However, we also interviewed students who felt a strong level of integration into their local sites. These students reported that other students and faculty at their sites were conducting a substantial amount of social-ecological research, as well as collaborating on various short-and long-term projects of their own. In particular, several interviewees were participants in an Integrative Graduate Education and Research Traineeship (IGERT) program based at the university anchoring their LTER site. These students reported finding support for socialecological research at their sites. When asked if socialecological research existed at his site, one interviewee felt strongly influenced by the work of his site's PI. He told us that the PI had been working over the past decade "towards the integration of the social sciences. I mean, that's what the whole IGERT program has been developed to do." Another student described how, as an IGERT student, she was "automatically involved in the [LTER site] grad community and required to go to the meetings." Students also described different ways that the IGERT program allowed, even required, them to receive training in social-ecological theories, methods, and practice. These included the requirement to pursue a minor in either the social or ecological sciences and to include a collaborative chapter in their dissertations.

Conceptualization
Graduate students had high general recognition of nascent social-ecological research initiatives in the LTER Network, but a more fragmentary awareness of network-level documents and resources available to facilitate this emerging area of investigation. According to the network-wide broad survey, the majority of LTER graduate students were familiar with social-ecological research, recognizing it as an important, interesting, and novel research initiative within the LTER Network. Ninety-two percent of respondents had previously heard about social-ecological research. Graduate students expressed a high level of enthusiasm for the field, with 59% stating that they were "interested" or "very interested" in social-ecological research and 80% agreeing or strongly agreeing that social-ecological research is important to the future of the LTER Network.
Even with this enthusiasm, 71% of the graduate student respondents to the broad survey stated that they had not read the ISSE initiative. One interviewee with a social science background had been working as a research assistant on an LTER social-ecological project for more than six months, yet was not aware of the ISSE or that there were other researchers at the network level engaged in social-ecological research. Conversely, some students were not only knowledgeable about the network but also participated in its higher level organizational and research-oriented activities. Other students had a moderate level of awareness, having been informed of certain network-wide initiatives regarding social-ecological research. However, when asked to describe network-level discussions and strategies, these same students expressed an incomplete understanding of the network itself. For example, one interviewee was quite familiar with the widely circulated press-pulse conceptual diagram, which illustrates how social and ecological research should be integrated and has become the centerpiece of the LTER Network's ISSE initiative. However, the student did not refer to the diagram as an LTER artifact but instead described it as a product of her LTER site.

Practice
Despite overall awareness of and interest in social-ecological research, relatively few LTER graduate students actually practiced social-ecological research. Findings from the broad survey revealed that the majority, 68%, of LTER graduate students were engaged in strictly ecological research in comparison with social scientific and social-ecological research (Fig. 3). The broad survey results showed that out of the 59 respondents who indicated that they were currently involved in at least one LTER research project, only 12 students, 20%, were involved in social-ecological or social scientific research projects. Moreover, those students actively involved in social-ecological research were still spending a significant amount of time on other ecological research projects. Nine out of the 12 students referenced previously were involved in at least one other ecological science research project. When asked to estimate the amount of total research time spent on their ecological, social, and social-ecological projects, these 12 students reported spending an average of 15% of their time on social-ecological and social scientific research.
We found a spectrum of network participation and awareness across our interviewees. A student with limited exposure to the network expressed regret for not being aware of the LTER Network earlier in his studies and provided a suggestion for those bringing new students into the network: "If initially I was told I was going to be working [in a network], that would be a great opportunity and resource, but I think it has to be straightaway. Because once you get involved in your research and you have this timeline right in front of you, you're budgeting your time, and it's hard to get involved in other things." Another student, although in the LTER directory, did not consider himself to be part of the network. He felt this way despite his membership in a cohort of students funded by a grant led by collaborative LTER researchers. As he described his experience: "We didn't really have a great relationship with [the LTER Network] outside of our professor. It was always sort of from outside ... I didn't have the opportunity to get involved." In contrast, other students found the LTER Network to be instrumental in their work and characterized their advisers and/or lead researchers as very vocal about the network and its social-ecological research initiative. As one student described, her project would not exist without the network: "We're doing a cross-site research project with a couple other LTER sites, so using the network to facilitate the cross-site study and collaboration has been really helpful."

CONCLUSION AND RECOMMENDATIONS
Our research has captured the conceptual challenges confronting graduate students in the LTER Network as the network strives to adopt a novel research program focused on social-ecological scholarship. These conceptual challenges highlight key findings that emerged from our research: (1) At the individual scale, graduate students hold different underlying ideas when they define social-ecological research and experience different modes of social-ecological research in practice in the LTER Network.
(2) At the site scale, graduate students involved in social-ecological research often faced institutional, communicative, spatial, and cultural obstacles to their integration into the LTER community.
(3) At the network scale, although graduate students were very interested in and aware of social-ecological research in general, they were less aware of specific network-level social-ecological initiatives, and only a small number of graduate students were actually engaged in social-ecological projects. We believe that these challenges are neither unique to the LTER Network nor signs of the ISSE initiative's failure but instead provide fertile ground for learning about the dissemination and adoption of new ideas in network-level science consortiums. At each scale described in our results, we found commonalities in the ways that graduate students successfully overcame such challenges. We discuss these successful strategies and offer suggestions for LTER and other network-level science leaders and personnel to consider as they take on and implement new research initiatives.

Recommendation 1: facilitate the creation, dissemination, and discussion of boundary objects
Our research demonstrates that LTER graduate students struggle with the term social-ecological and therefore hold a plurality of definitions of this type of research instead of a shared concept of what this term means. In general, shared concepts are critical because they allow interdisciplinary researchers to communicate across disciplines (Pickett 1999, Wear 1999, Heemskerk et al. 2003) and can form a foundation for implementing successful new research initiatives. However, novel concepts like social-ecological research require a certain amount of discussion, practice, and debate to refine (Miller et al. 2008).
One way to focus discussions and engagements with novel research initiatives is for networks to develop boundary objects (Star and Griesemer 1989) to facilitate discussions about novel research concepts and to create a wider collective awareness about them. A boundary object is a concept portrayed through various media, such as a diagram, map, or scientific specimen, that can travel across different scientific communities without losing its meaning, although the interpretations of such meanings may vary across different institutional, geographic, or disciplinary communities.
In the LTER Network, the ISSE diagram (Collins et al. 2011) has acted as a key boundary object. This conceptual map of how social scientific research can be integrated with traditional ecological research has been disseminated, reworked, and used to encourage discussion about the theory and practice of social-ecological research among diverse researchers at LTER sites. Our results showed that some graduate students were actively engaging with the ISSE research framework, but the large majority remained unaware of the ISSE diagram or the ISSE report. Clearly, graduate students had insights to offer about the design of social-ecological research in the LTER http://www.ecologyandsociety.org/vol18/iss3/art42/ Network even if they were not aware of the ISSE. Thus, it is important for networks to actively encourage graduate students' awareness of boundary objects and to encourage evaluation and reconceptualization from the ground up as well as from the top down.

Recommendation 2: provide network-level support for communities of practice
It is important to support communities of practice within largescale research networks. In our research, we found that some students felt as if they were peripheral, rather than integrated, members of their LTER site. In some cases, this was attributable to physical distance from the site. In addition to Grimm and Redman's (2004) discussion as it relates to LTER interdisciplinary research, the "distance matters" issue in collaborative research has been covered extensively elsewhere (Olson and Olson 2000). However, we interviewed several graduate students conducting social-ecological research without a geographic distance barrier who reported a feeling of cultural distance. Miller et al. (2008) addressed the causes behind this cultural distance, suggesting that LTER sites tend to have one dominant research paradigm that is not open to other ideas. The LTER Network has worked to encourage synthetic science across disciplines with the creation of the National Center for Ecological Analysis and Synthesis (Hackett et al. 2008) and to fully integrate social science into the network . We suggest that these shifts do not always trickle down to the most junior levels of the network. In our results, established communities of practice, such as working groups and integrated training programs, often seemed to help counteract students' sense of distance.
Several students participating in an IGERT program reported finding a supportive community for social-ecological research. Because participants meet frequently, the IGERT structure creates cohesion and opportunities for transmission of interdisciplinary knowledge and insights from higher level personnel to junior personnel. Although we recognize that all IGERT programs are not the same, our interview results bolster the findings of studies that have provided more formal evaluations , August et al. 2010, offering support for continuation of these programs and perhaps for implementation of similar interdisciplinary training programs at LTER sites that are currently without them. Additionally, the emerging cohort of LTER scientists that work on urban ecology (see Grimm et al. 2000, Pickett et al. 2008) provides a research community in which several students reported finding support. We expect that this and similar self-organized communities of practice will continue to develop as socialecological research becomes the norm in LTER, and we would recommend that research networks support these groups when possible. These types of interdisciplinary communities act as a space for graduate students to collectively respond to and reformulate higher level blueprints like ISSE.

Recommendation 3: ensure that lead researchers and advisers educate current and potential members about the network, across all scales and locales
As our research demonstrates, graduate students involved in social-ecological research often faced institutional, communicative, spatial, and cultural obstacles to their integration into the LTER Network and its research sites. There were students working on social-ecological research without knowledge of the network-level support that exists for this type of integrated research. Few students had read the ISSE initiative, and most were not aware of the explicit support for social-ecological science that is written into the plan. In addition, several students did not become aware of the LTER Network or its resources until well into their degree programs or research projects. At that point, becoming part of the network becomes more of a burden than a benefit.
To overcome this challenge in all research networks, it is important to inform graduate students about the research network and its initiatives as soon as they enter an academic program or become employed by a research study. This must include educating and training potential research advisers because they are largely responsible for overseeing the development and execution of social-ecological research carried out by new network members. Indeed, nearly all of our interviewees indicated that the influence of their graduate advisers or PIs strongly impacted whether or not they felt supported in the pursuit of social-ecological research projects and whether or not they became involved in LTER Network activities.
We offer these suggestions as potential ways to bridge the gap between the senior network personnel who formulate novel research initiatives and the junior researchers who often implement these initiatives. Large-scale, distributed scientific research networks with multidisciplinary, interdisciplinary, and transdisciplinary approaches have changed the way that science is practiced. It is essential to ensure that the next generation of scientists is equipped to carry forward new research initiatives and to continue to advance network-level science.
Responses to this article can be read online at: http://www.ecologyandsociety.org/issues/responses. php/5606

APPENDIX 1
Broad survey about graduate student socio-ecological experiences and attitudes in the LTER Network

SURVEY INTRODUCTION
This survey is about graduate student attitudes towards and experiences with socio-ecological research in the LTER Network. This survey should take approximately 10-15 minutes of your time. It will contribute to a research study that a group of LTER graduate students has organized as follow-up to a 2009 ASM working group on this topic. Survey results will be used as data for a research publication and will be presented to LTER leadership. Your responses will be kept confidential. Your name or other personal information will not be publicly linked to your survey responses. Thank you for your time! Should you have any questions about this survey or our work, please contact us.

SECTION 1: SOCIO-ECOLOGICAL RESEARCH in the LTER NETWORK
This first section of this survey asks you about your familiarity with socio-ecological research and your views of it. (Section 1 of 4) Q1 Since joining the LTER network, how often do you typically hear about socio-ecological research?  Never (1)  Rarely (2)  A few times (3)  Frequently (4)  All the time (5) Q2 How interested are you in socio-ecological research?  Very disinterested (1)  Disinterested (2)  Neither interested or disinterested (3)  Interested (4)  Very interested (5) Q3 What socio-ecological research topics interest you?
Q4 What other kinds of research topics interest you?
Q5 Please indicate how much you agree or disagree with the following statements: Strongly Disagree (1) Disagree (2) If the respondent indicates that he or she "disagrees" or "strongly disagrees" with the statement "I am able to spend as much time studying socio-ecological research topics as I would like" in question 5, then ask: Q5A Why are you not able to spend as much time studying socio-ecological topics as you would like?
Q6 In your own words, how would you define socio-ecological research?
Q7 Please indicate how much you agree or disagree with the following statements: Strongly Disagree (1) Disagree (2)   LTER site(s) and field sites or stations affiliated with another organization (e.g. US Forest Service, a University. Please specify) (6) ____________________  LTER site(s) and other sites (please specify) (7) ____________________ Q9 Is your research collaborative? Collaborative research typically involves actively working with other colleagues as part of a larger research team to answer a research question. Researchers work within their own disciplines to make a contribution to group data collection and analysis.  Yes (1)  No (2)  Not sure (3) Q10 Is your research interdisciplinary? Interdisciplinary research synthesizes the theories and methods of multiple academic disciplines (e.g. anthropology and biology) to answer a research question.  Yes (1)  No (2)  Not sure (3) Q11 How many of the following types of research projects are you involved with? ______ Socio-Ecological (1) ______ Social Science (2) ______ Ecological science (3) ______ Other (4) If the respondent indicates that they are involved in one or more socio-ecological research projects in their response to Q11, then ask him/her these questions: Q11A What statement best describes the socio-ecological project ( If the respondent indicates that he or she "agrees" or "strongly agrees" with the statement "Graduate students face challenges when conducing socio-ecological research" in question 14, then ask: Q14A Select the challenges related to socio-ecological research that you think are the most difficult.

SURVEY INTRODUCTION
Welcome to our survey about graduate student experiences with socio-ecological research in the LTER network. It will only take a few minutes of your time to complete. Your answers are very important and will contribute to a research study that a group of LTER graduate students has organized as follow-up to a 2009 ASM working group on this topic. Please note that your responses will be kept confidential. Your name or other personal information will not be publicly linked to your survey responses. Should you have any questions about this survey or our work, please contact us.  (2) Q7 What academic discipline(s) do you most closely identify with?

SECTION 3: DEFINING SOCIO-ECOLOGICAL RESEARCH AND ITS CHALLENGES
In this next section, we will ask you some general questions about socio-ecological research.
(Section 3 of 4) Q8 In your own words, how would you define socio-ecological research?
Q9 Do you think there is an agreed-upon definition of socio-ecological research among LTER researchers? o Yes (1) o No (2) o Not sure (3) Q10 How would you rate the following challenges associated with socio-ecological research? Not problematic (1) Somewhat problematic (2) Highly problematic (3) Funding (1) Institutional support (2) Advisor support (3) Peer support (4) Theoretical issues (5) Methodological issues (6) Finding collaborators (7) Bridging disciplines  (5) o Sixth year (6) o Seventh year (7) o Other (8) (2) If respondent answers "No" to Q17, then skip to End of Survey If respondent answers "Yes" to Q17, then ask: Q17A Thanks for saying "yes"! Please provide your contact information below (name, e-mail address and phone number) so that we may get in touch with you. Please note that your contact information will be kept confidential at all times: SURVEY CLOSING: You are now finished with the survey! Thank you for your time. Your answers are very valuable.