International Collaborations in Behavioral and Social Sciences: Report of a Workshop (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Enhancing International Research Collaborations International research collaborations have many benefits to institutions and researchers both in the U.S. and abroad; thus, workshop participants were enthusiastic about the importance of such collaborations and hoped to encourage more U.S. behavioral and social scientists at all phases of their careers to engage and learn from their research colleagues in other countries. Workshop participants were committed to finding ways around the obstacles that often impede the success of international collaborations. Their recommendations centered on developing research capacity around the world among early-career as well as more established scholars, and, further, by facilitating these types of research interactions by addressing specific difficulties that international collaborators encounter. They also had suggestions for early-career scholars and funding agencies. DEVELOPING RESEARCH CAPACITY AROUND THE WORLD: TRAINING AND INFRASTRUCTURE By developing capacity around the world, behavioral and social scien- tists across countries ensure having highly competent colleagues with whom to collaborate on international projects. Research can and should have long- term benefits for individuals in the countries in which it is being conducted. The development of research capacity ensures that researchers within those countries are prepared to continue and adapt the process either individu- ally or as part of national or international collaborations. Recommenda- 24 

ENHANCING INTERNATIONAL RESEARCH COLLABORATIONS 25 tions for improving capacity focused on two areas: human resources and institutional infrastructure. Developing either in isolation will not result in sustainable local capacity. Individuals who receive training will be unlikely to remain in their home institutions if those institutions lack the resources for research. At the same time, institutions need skilled individuals to put to use the resources that institutions can provide. Workshop participants examined a number of possibilities for improving research capacity along both of these avenues. Opportunities for advanced research training should include a menu of flexible options. Researchers of different skill levels, or at different points in their careers, with access to varying resources or differing degrees of flexibil- ity in their schedules and commitments, could then avail themselves of the appropriate option. Among the approaches to training discussed were brief workshops focused on a single skill (such as a particular coding technique or writing grant proposals), visiting-scholar programs, extended summer training programs, supplemental or partial graduate training programs, and formal graduate degree programs. Oscar Barbarin (now at the University of North Carolina) explained and reflected on his experience leading the University of Michigan’s South Africa Initiative Office. One notable project was the Moody Scholars Program for South African faculty who were simultaneously working as lecturers and completing the Ph.D. at their home institutions. The program provided stipend and travel expenses to permit young faculty members to spend the summer at the University of Michigan, devoting their time to writing their dissertation. The summer was chosen because affordable hous- ing, computer facilities, and office space were more readily available then. Each year two or three scholars spent their time doing library research, consulting with senior scholars, and participating in a structured research seminar along with University of Michigan graduate students. This op- portunity proved critical to the South African scholars’ completion of their dissertations and to their careers. The Quantitative Program for South African Scholars, also carried out at the University of Michigan, provides another training model. This pro- gram brought 20 South African scholars, selected from groups historically underrepresented among researchers, to participate in short-term courses offered through the Institute for Social Research. The scholars participating in this Mellon Foundation-supported program were at the University of Michigan for three consecutive summers. They enrolled in statistics courses with increasing levels of difficulty and also participated in a weekly inte- 

26 INTERNATIONAL COLLABORATIONS grative seminar addressing the research process. To complete the program, each fellow conducted an independent research project and wrote a research report involving application of the quantitative skills they had developed to a large dataset (e.g., the South African Living Standard Measurement Survey or the South African October Household Survey). Throughout the program the South African scholars had access to the University of Michigan’s library and computer facilities and had many opportunities to attend and give talks. They built connections with each other, with other Michigan gradu- ate students, and with faculty. The program provided formal coursework to fill in gaps in their training and also initiated them into a community of scholars and enhanced what Barbarin referred to as their “meta-research” skills—such as setting a research agenda, framing a research design, giv- ing and receiving feedback. The program culminated with a conference at which the scholars presented their work in South Africa. The summer training program helped develop enhanced methodological and statistical skills among South Africans who will train and mentor the next generation of South African students. Kenneth Rubin (University of Maryland, College Park) also spoke of his experience with the International Society for the Study of Behavioral Development, a multidisciplinary organization committed to developing the capacity of young scholars from countries with limited resources for research. This fairly small group (with a current membership of 1,200) charges lower dues (and sometimes gives free membership) to scholars from Sub-Saharan Africa, Latin America, and Eastern Europe. It holds biennial conferences focused on research topics developed by the conference hosts. A conference in Gaza, for example, addressed the effects of political violence on children’s development, while a conference in Recife, Brazil, explored the impact of children’s homelessness on child-parent relations. The confer- ences are preceded by sponsored workshops that provide travel and housing for junior scholars. Training workshops address particular skills, such as secondary analysis of datasets, methods for working with longitudinal data, or preparing manuscripts. These types of mechanisms for supporting early- career researchers as well as more established scholars to attend international meetings are important. Workshop participants mentioned other models for training, such as holding preconference meetings using formats that promote problem-solv- ing discussion and address the developmental needs of early-career scholars.  See www.issbd.org, accessed October 26, 2006. 

ENHANCING INTERNATIONAL RESEARCH COLLABORATIONS 27 These meetings could enable more junior scholars to consult with senior researchers to hone a research question or design. An alternative model is to send an experienced researcher or a group of advanced scholars to provide a minicourse in statistics at an institution lacking this resource. A third alternative is to develop Internet Listservs in which early-career scientists could ask questions and solicit advice on an array of topics. Workshop participants emphasized the importance of providing a menu of options, since the effectiveness and cost-benefit ratio of any train- ing model will depend on many factors. Lengthy degree programs are valu- able but are expensive. Furthermore, some countries or institutions may avoid sending their scholars to such degree programs because they fear that these individuals, with new credentials, might take a job outside their home country. Single workshops can impart a specific skill but are not adequate for building broader capacity. In Barbarin’s view, programs of moderate intensity, such as repeated summer programs, are the most productive and cost effective. Barbarin provided data on the costs for graduate students in the doctoral program he directs at the University of North Carolina, Chapel Hill, and what it costs to operate the programs there. (See Table 1.) Barbarin also advocates that senior scholars do their best to identify any possible “hidden little pots of gold,” such as research assistantships, library privileges, internal discretionary funds, or tuition swaps between institu- tions that could enhance the range of training opportunities. Senior researchers need to attend to the mentoring responsibilities present in any training opportunity. Preparing and welcoming someone into a community of scholars involves not only imparting skills but also offering guidance, investing in an individual’s development, building trust, and fostering reciprocity. Through such relationships researchers can work toward a common language and common goals, moving beyond research dependence to interdependence. Trained individuals need good home institutions in which to work in order to make their training time and effort worthwhile and to put their skills to use. Mark Nichter had several recommendations for improving institutional capacity in countries with limited resources. Libraries should be strengthened and regional repositories established for journals and other scholarly materials. Costly journal subscriptions should be made more af- fordable, and electronically stored material should be made available via the Internet. Senior researchers can serve as filters for some of this material, se- lecting the most useful recent journal articles and ensuring their availability to dispersed libraries and other repositories and on the Internet. 

28 INTERNATIONAL COLLABORATIONS TABLE 1 Estimated Costs Associated with Training Models for South African Students at the University of North Carolina, Chapel Hill Stipend Training Total Home Model in U.S. Costs in U.S. (U.S.) Country Workshop — $1,000 $1,000 $400 Visiting Scholar (3 Months) $5,000 $1,000 $6,000 — Non-degree Training Program (3 years) $15,000 $21,000 $36,000 $9,000 Advanced Degree Masters (2 years) $36,000 $24,000 $60,000 $6,000 Advanced Degree Ph.D. (5 years) $90,000 $60,000 $150,000 $5,000 SOURCE; Oscar Barbarin III, workshop presentation, October 6, 2006, Northwestern University, Evanston, IL. Nichter suggested that countries with few resources establish national research centers to serve as a hub for research networks. For example, India’s recently established first school of public health is now the hub of a re- search network on tobacco cessation and locally appropriate interventions. Research networks should build on existing resources—for example, by getting existing medical schools to collaborate with one another, attracting students at different levels, linking dispersed researchers, and increasing the visibility of research. Research networks, Nichter argued, can not only ac- complish greater results, but these partnerships also create fora for learning and foster a common sense of identity, improving the morale and motiva- tion of researchers. Developing research capacity elsewhere, both through a menu of training options for individual scholars and by strengthening their home institutions and networks, can begin to address many of the obstacles that result from differences within international research teams. By sharing common skills, methods, and approaches—whether through workshops, summer programs, or degree programs—scholars can gradually build their repertoire of shared “practices,” as Goodnow used the term. By increasing local research capacities, some of the power asymmetries that impede col- 

ENHANCING INTERNATIONAL RESEARCH COLLABORATIONS 29 laboration can begin to be addressed. Both can build the relationships of reciprocity and trust that are essential to successful collaboration. COMMUNICATION Good communication between collaborators will not only avoid mis- understandings on substantive issues but also build trust. Researchers need to be aware and respectful of cultural differences in styles of communica- tion. This may include anticipating that a collaborator will find it difficult to voice a challenge or criticism of an investigator who has brought funds to support the project. Nichter contrasted a confrontational style more ac- ceptable in India with the much greater reticence encountered among col- laborators in Indonesia. Helwig noted hesitancy among Chinese collabora- tors to express criticism. Workshop participants urged researchers to create a variety of opportunities in which collaborators can share any discomfort they have with the research questions or methods and make it possible to return to these issues at different times in the project. One possibility is to make an explicit call to participating researchers to reassess the strategy at the midpoint of a project or when a progress report is being prepared for funders. Without such opportunities for communication, criticisms may not be voiced and, consequently, corrections and improvements may not be made. Workshop participants emphasized that at least some communica- tion in any collaboration should be face to face, whether through visits to research sites or gatherings at international conferences. All found periodic face-to-face interaction to be invaluable for addressing or preventing misun- derstandings and for building long-term relationships of trust. E-mail and electronic conferencing are excellent tools, particularly for frequent updates, minor adjustments or corrections, and joint editing of texts. But workshop participants agreed that it is insufficient for building the relationships and cultivating the trust and reciprocity that are essential to collaboration; for these, in-person interaction is necessary. PROJECT DEVELOPMENT Addressing the specific obstacles they had identified earlier, workshop participants offered a number of comments and suggestions. Anyone considering an international collaboration should appreciate the longer than usual lead time that the first phase of a research project will require. 

30 INTERNATIONAL COLLABORATIONS Collaborators need to remain attentive not only to discussing the major constructs and their meaning but also to the layers of translation (forward translation, consensus choices, back translation, field testing). Political sensitivities, cultural relevance, and appropriate administration or study questions and procedures also require attention. Such tasks will lengthen the period of project preparation, and international researchers should plan accordingly. In securing collaborators and then recruiting staff, researchers also need to be aware of differences in practice regarding such matters as workload, supervision, pacing, and vacations. Communicating early and often, so as to prevent misunderstandings, is the best means for handling these issues. Workshop participants recommended that researchers begin anticipating and addressing these issues at the earliest stages of their projects. ETHICS REVIEW PROCEDURES Regarding institutional review boards (IRBs), workshop participants had a number of suggestions. From his experience with BEIP in Romania, Nelson recommended that international research projects contact local nongovernmental organizations for advice on such matters as culturally appropriate consent forms and local pay rates. Researchers should also try to gather precedents that have previously been used to satisfy IRB require- ments. Because the home institutions of many potential collaborators will not have formal IRBs, workshop participants urged that some sort of inter- national guidelines be developed. This would not be a template, mimicking the legalistic and bureaucratic approach found at too many U.S. institu- tions. Rather, it would be a general framework for establishing an ethics review process aimed at protecting human subjects, with attention to elic- iting truly informed consent and methods appropriate for minimum-risk research. Such guidelines would assist institutions in meeting the demands of U.S. IRBs without having to adopt the U.S. pattern or start entirely from scratch. Workshop participants noted that no one holds accountable or regulates IRBs and that these boards sometimes abuse their power when they shift from ethical to scientific oversight, challenging research designs that have already passed peer reviews. Nichter recommended that scholars look into whether their universities are indeed required to submit to IRBs for all research activities. Caroline H. Bledsoe, “Hope in the IRB Mire? The Federal-Wide Assurance Box 4(b)  Option,” workshop presentation, March 27, 2006, Northwestern University, Evanston, IL. 

ENHANCING INTERNATIONAL RESEARCH COLLABORATIONS 31 DATASETS Researchers need to ensure that all collaborators are working with comparable constructs or at least are prepared to document and work with the variations. They also need to confirm that their questions meet at least a minimum level of political sensitivity and cultural appropriateness. As Helwig learned from conferring with Chinese collaborators, by eliminating hypothetical scenarios involving families with multiple children, questions directed at issues of rights and autonomy could still be asked in a society that promotes the one-child family. Translations of questions also need to be handled carefully. Alexandra Quittner provided an example of how a series of options for the term “cough” were identified; consensus was reached among the options; and the consensus items were then translated, back translated, and tested in the field. Equally careful decisions are necessary to avoid biases in instruments, measures, and administration. These are difficult tasks. Quittner advocated that guidance be made available on how to develop an instrument, particu- larly one that will be internationally relevant. Suzanne Bennett Johnson urged that, once such instruments are devised, they be made available to other researchers. Since measurement and instrument development for use in international research is so time consuming, a repository of avail- able instruments with their translations could permit more rapid scientific advances. Data management also needs to be improved to facilitate international collaborations. This includes initial explicit agreement among collaborators and then eventual standardization across the field, regarding such matters as cleaning data, handling missing data, and incorporating late submissions. Some large and well-established international research organizations have procedures in these areas that can provide guidance to investigators conduct- ing smaller projects. The infrastructure of data management should be de- veloped and shared with the aim of facilitating international collaboration. Greater resources should be devoted to documenting completed datasets so that they can be used for secondary analysis. Van de Vijver noted the Data Documentation Initiative, an international effort to establish a standard for technical documentation describing social science data. Collaborators also need to communicate clearly about ownership of and access to datasets. Several models are possible. For example, each principal investigator could retain control over how his/her data are shared, including what parts of the  See www.icpsr.umich.edu/DDI, accessed October 26, 2006. 

32 INTERNATIONAL COLLABORATIONS data are shared, whether or how the data are deposited (attending to IRB regulations concerning data storage), whether the data may be analyzed at a remote site, and whether other researchers need to apply in order to gain ac- cess or for what period of time access to the data may be restricted. These are all decisions to be reached among research collaborators. Goodnow noted that the population under study may well presume or insist on ownership of the collected data. Resolving issues of ownership and access will require communication as well. PUBLICATIONS As with datasets, there needs to be ample early communication about publications that will result from research collaborations. The issues in- volved touch on both dissimilarities in practice and asymmetries of power. What will be the order in which authors are listed? What level of involve- ment will meet criteria for authorship? In some research settings it may “go without saying” that principal investigators will be listed as first authors. In other settings the standard may be for all collaborators to form a group, with all publications under the group’s name. Some researchers may expect the freedom to publish their own segment of the research separately or may feel justified in presenting the work of the group without acknowledging individual group members’ contributions. These issues need to be discussed and, to the extent possible, decided in advance. The handling of publications can also contribute to building capac- ity. Several workshop participants urged that research be published in the language(s) of the country where the work was done. Efforts should be made to help scholars whose mother-tongue is not English publish in international English-language journals. This may include workshops on crafting a manuscript suitable to such journals or creating more opportu- nities for joint authorship with established scholars. There was also agree- ment among workshop participants on the need to educate journal editors about the unique contributions of international collaborative research and to suggest greater flexibility when reviewing submissions that come from researchers for whom English is a second language. Unnecessarily rigid style requirements can prevent valuable research results from being shared. International publications will also help foreign scholars gain access to funding streams, increasing the chance that they will be able to initiate new collaborations. 

ENHANCING INTERNATIONAL RESEARCH COLLABORATIONS 33 DISSEMINATION In addition to professional publications, international research col- laborations should give enhanced attention to disseminating their results to relevant audiences outside the academic and scientific communities. Not all research is intended to be of direct public interest or utility, but dissemi- nating research results, at interim points as well as at the conclusion of a project, will further two objectives. If handled astutely, it has the potential to deflect the suspicions, conspiracy theories, and unwarranted attacks that a number of international research projects have experienced. It may also raise the profile, status, and support for local researchers, thereby contribut- ing to local capacity. Workshop participants urged researchers to think about the different audiences that need to be informed and how to make their research acces- sible to those audiences. Governments, policymakers, health care providers, educators, communities, or parents will need to be addressed differently and through a variety of venues. Web sites might be appropriate in certain set- tings, whereas formal announcements incorporating local dignitaries, press conferences, or radio interviews may be useful in others. Local support and engagement should be encouraged whenever appropriate. EARLY-CAREER SCHOLARS Workshop participants were well aware that the obstacles to interna- tional collaboration, while daunting to senior researchers, may appear insur- mountable to younger scholars, particularly junior faculty concerned about tenure. They offered several recommendations. Professional societies should add workshops or small group meetings to their existing international meet- ings where junior researchers could identify potential collaborators and discuss possible projects. At similar workshops, experienced international researchers should help review proposals by junior scholars for international collaborations. Guidance should also be provided for navigating the IRB process. Junior scholars should be encouraged to pursue long-term collaborative strategies. Research projects should be designed to build on one another, so that as a researcher’s linguistic competence, cultural familiarity, and relationships with collaborators deepen, more challenging projects can be conceived and initiated. 

34 INTERNATIONAL COLLABORATIONS FUNDING AGENCIES Clearly, many of the recommendations for junior scholars or senior researchers are not cost-free. Spending further time in the formation of a project or the careful translation of instruments will consume resources, as will traveling to engage in face-to-face communication with collaborators. Documenting and managing a dataset both require funds, as does dissemi- nating results. Training programs, no matter how minimally budgeted, are still costly. Holding methodological or content-focused sessions, even if tagged on to existing conferences, requires sponsors. Workshop participants therefore generated a list of comments directed specifically to funding agencies. Workshop participants urged greater flexibility on the part of funders. For example, funding agencies might permit research grants to cover more of the phases involved in international collaboration instead of limiting support to data collection, analysis, and publication. Scholars need to travel to potential research sites, meet with possible collaborators, and do some preliminary exploration into the feasibility of research, with particular at- tention to issues that might arise over translatability and cultural appropri- ateness. Workshop participants emphasized that this sort of work can only be done on the ground and in person. Institutions and funding agencies might provide small grants for this essential exploratory travel and project planning. This may be particularly important for early-career investigators, who often do not have access to as many resources as their senior col- leagues. Funders should also recognize the additional lead time needed to plan international research collaborations, and should ensure their funding mechanisms have enough flexibility to take such needs into account. Rather than requiring proposals to have a predetermined design and translated instruments, funding agencies could permit research collabora- tors to develop locally appropriate aspects of a design once a research agenda has been clearly defined. Research projects might also be permitted to put more of their grants toward training local staff to undertake the research, rather than solely toward the production of data by experienced researchers from outside the country. Documentation of data for secondary analysis and dissemination of results need to be recognized as valuable but costly aspects of international research projects, meriting designated funding. Workshop participants expressed concern about funding agencies’ requirements that all principal investigators attend central meetings in the donor country—a 

ENHANCING INTERNATIONAL RESEARCH COLLABORATIONS 35 practice that usually stresses the finances and time of the international col- laborators on the project. In addition to greater flexibility, workshop participants encouraged funding agencies to consider creating new funding mechanisms specifically for enabling early-career scholars to become engaged in international col- laborations and for offering training, via a variety of innovative programs, for scholars from countries with limited research resources. There are many steps that could be taken to facilitate international collaborations. Workshop participants made an array of recommendations relevant to journal editors, ethics review boards, and funding agencies and offered suggestions to collaborating researchers who find themselves crossing many national, cultural, or disciplinary boundaries. Obstacles to international research clearly exist. Nonetheless, workshop participants enthusiastically endorsed research collaboration for its great potential to advance the psychological, behavioral, and social sciences.