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.

Appendix D Results of a Survey of International Collaborative Research in Psychology: Views and Recommendations from Twenty-six Leaders of Projects Judith Torney-Purta, Ph.D. Professor of Human Development University of Maryland, College Park jtpurta@umd.edu Member of the U.S. National Committee for the International Union of Psychological Science Survey Consultant (2005-2006) Workshop on International Collaborations in Social and Behavioral Sciences Research October 5-6, 2006 Northwestern University Norris University Center Evanston, Illinois 64 

APPENDIX D 65 INTRODUCTION This survey covering international collaborations was designed in June-July 2005 by the survey consultant after extensive input from the U.S. National Committee for the International Union of Psychological Science. It was sent to 53 persons identified by the committee members. There were 21 respondents (as of January 5); after follow-up, 26 had re- sponded (as of May 1, 2006). The response rate is just short of 50 percent. One respondent noted that filling out the survey was “burdensome”; others did not comment (or indicated that they were pleased to reflect on these topics). There was variation in the length of responses, but on the whole they were thoughtful. The questions seem to cover the major themes. The projects reported on had been funded by a variety of governmental and nongovernmental sources (inside and outside the United States) and ranged in duration from several decades to quite brief periods. In the opinion of those answering this survey, international collabora- tive research is making a positive contribution to many subfields of psy- chology. Psychologists who are involved seem ready to reflect on ways to make it more feasible and more attractive to other scholars. International collaboration has a role to play in advancing psychology as a science, in building scientific capacity, and in informing policy and practice in the United States and worldwide. Parts A (general information) and C (suggestions) of the survey in- cluded open-ended questions only. Part B asked for a rating of the extent of problems experienced (on a scale from 1 to 4, with 4 indicating extensive problems), followed by a narrative response. Responses from Part C are incorporated into Part B or included in the section of this report presenting recommendations. SUMMARY OF RESPONSES TO PART A OF THE SURVEY Clusters Representing Five Types of Projects These projects have been classified into five clusters according to infor- mation provided in the topic/title and purpose sections of the survey (see Table D1).  Each of the 26 respondents was counted as one project, even if two or more related activities or subprojects were included in the responses. 

66 APPENDIX D TABLE D1 Titles/Summaries of the Internationally Collaborative Projects Classified into Five Clusters Respondent Comments A The Impact of Social and Cultural Adaptation of Juvenile Immigrants from the Former Soviet Union in Israel and Germany on Delinquency and Deviant Behavior CLUSTER 1 B I have done a number of collaborative projects with colleagues from Mexico, Guatemala, and Spain. In addition, I have done international work with postdocs and grad students from the U.S. who grew up in other nations (India, Turkey, England, Japan, Guatemala), but I am not focusing on that work here (although it is very important). CLUSTER 1 C International Collaborative Study of Ethnocultural Youth (ICSEY), a 13-country study of adolescents from immigrant and national (nonimmigrant) backgrounds CLUSTER 1 D The Effects of Improving Care Giving on Early Development, a project that studied training and training plus structural changes in orphanages in St. Petersburg, Russian Federation, and the effects on most aspects of caregiver behavior and children’s development. CLUSTER 2 E 1. Regulation of Stress Response in Neonatal Mice 2. Stress Response During Pregnancy and Birth Outcomes 3. Ontogeny of Circadian Rhythms of Corticosterone in the Rabbit CLUSTER 3 F I have carried out an ongoing collaboration related to biomedical as well as social functioning and adaptation of the oldest old people (age 80 and over) in Sweden. CLUSTER 3 G Collaborative Research on Hippocampus and Consolidation CLUSTER 3 H The Meanings of Learning, Achievement, and Motivation: A Study of Learning Beliefs and Behaviors in Three Cultural Milieu CLUSTER 1 

APPENDIX D 67 TABLE D1 Continued I Adolescents’ Interpretation of the “Social Contract”: A Seven- Nation Study —survey study of adolescents’ civic values and behaviors and attitudes toward the economy and the state CLUSTER 1 J 1. Civic engagement among youth in New York and Paris 2. Impact of HIV/AIDS on children’s development in South Africa CLUSTER 1 K Center for the Analysis of Pathways from Childhood to Adulthood (CAPCA), coordinated by the University of Michigan CLUSTER 5 L 1. International Study of Depression and Anxiety in Patients with Cystic Fibrosis and Their Caregivers 2. Translating and validating a disease-specific quality-of-life measure for cystic fibrosis in several countries 3. Evaluating the impact of lung transplantation on the quality of life of patients with cystic fibrosis, both those who go on a transplant list and those who choose not to be listed CLUSTER 3 M The IEA’s Civic Education Study: Adolescents’ Civic Knowledge and Political/Social Attitudes in 29 Countries CLUSTER 1 N Modulation of vulnerability to gastric ulceration by psychological context CLUSTER 3 O The Bucharest Early Intervention Project CLUSTER 2 P 1. Cross-national study of highly successful women with families 2. Cross-national collaboration on an intervention project to enhance critical thinking skills 3. Research project on sex roles and sex stereotypes performance in Turkey 4. Comparative analysis of U.S. and post-Soviet perspectives on selected topics in psychology CLUSTER 1 Q Neural Regulation in the High-Risk Infant CLUSTER 2 

68 APPENDIX D TABLE D1 Continued R Adolescent-Parent Relationships: A Cross-Cultural View CLUSTER 1 S 1. Creativity in Design 2. 3-D and 2-D Visualizations in Molecular Biology 3. Animations in Teaching Chemistry 4. Roles of Vividness of Landmarks and Paths in Learning Routes CLUSTER 4 T TEDDY—The Environmental Determinants of Diabetes in the Young CLUSTER 3 U A Microgenetic/Cross-Sectional Study of Matrix Completion CLUSTER 4 V The IEA’s TIMSS Video Studies—1993 to 2003 CLUSTER 4 W Long-Term Effects of Urbanization and Poverty on Health and Development in Johannesburg CLUSTER 5 X 1. Representing and Learning from Classroom Processes— Comparing Elementary Mathematics Instruction in China and the U.S. 2. Language and Symbolic Development—Comparing Mandarin Chinese and English CLUSTER 4 Y 1. Longitudinal Study on the Ontogenesis of Individual Competencies 2. Scientific Reasoning and Science Education CLUSTER 5 Z Democratic Decision Making and Values Education in Mainland China and Canada CLUSTER 1 

APPENDIX D 69 CLUSTER 1: DEVELOPMENTAL OR SOCIAL WITH HUMAN SUBJECTS IN ADOLESCENCE OR ADULTHOOD (10) CLUSTER 2: DEVELOPMENTAL WITH HUMAN SUBJECTS IN INFANCY OR EARLY CHILDHOOD (3) CLUSTER 3: PSYCHOPHYSIOLOGICAL OR MEDICAL WITH HUMAN OR ANIMAL SUBJECTS (6) CLUSTER 4: LEARNING AND EDUCATIONAL PROCESSES WITH HUMAN SUBJECTS OF SCHOOL AGE OR ADULTS (4) CLUSTER 5: LONGITUDINAL STUDIES OF HUMAN SUB- JECTS (3) It was decided that three projects were the minimum to form a cluster. The areas with two or fewer respondents (projects) included organizational and social psychology. This was because repeated requests to the principal investigators or co-principal investigators of two major projects did not elicit a response. Project clusters did not differ in the extent of problems reported (see Table D2). Generally the majority of projects were rated by the respondents as having few problems relating to the themes of the questions. Only one rating of 4 (many problems) was given. The modal rating was 1 (indicating few problems). If problems were noted, they were most likely to be practical issues (covered under Theme 4). It may be that the respondents judged the problems in relation to the benefits achieved (and minimized the problems in making their ratings). It may also be that those contacted for those proj- ects that had serious problems elected not to respond. Value Added by International Research (Theme 1) Clusters did differ in the specific expected “value added” from interna- tional collaboration (asked under Theme 1 of Part B of the survey). Cluster 1 projects sought collaboration to investigate contexts outside the country  Given the small N’s in each cluster, we did not conduct significance tests to compare these ratings. 

70 APPENDIX D TABLE D2 Ratings of Extent of Problems or Challenges by Project Ratings of Extent of Problems by Theme (4-point scale) T3 T4 T5 Abbreviated T2 Conduct Practical Data Titles Participants Research Issues Publications CLUSTER 1: DEVELOPMENTAL/ SOCIAL Mean 1.7 1.8 1.9 1.7 CLUSTER 2: INFANCY/ EARLY CHILDHOOD Mean 1.3 1.7 2.0 1.0 CLUSTER 3: PSYCHOLOGICAL/ MEDICAL Mean 1.3 1.5 1.7 1.8 CLUSTER 4: LEARNING AND EDUCATIONAL PROCESSES Mean 1.5 1.8 2.0 1.3 CLUSTER 5: LONGITUDINAL Mean 2.0 1.7 1.3 1.7 NOTE: 1 corresponded to a rating of no problems; 4 corresponded to a rating of extensive problems. of the investigators (using phrases like scientific, practical, or developmental context; generalizability of findings; quasi-experimental study of influences; and challenging narrow research findings). They were likely to seek policy implications as well as implications for practice. Cluster 2 projects tended to look at institutional conditions of early development that may place children at risk (often looking for implications for practice—for example, in adop- tion). Cluster 3 projects engaged in international collaboration largely to take advantage of skills or techniques available at specific universities or labs abroad or to find instances of a medical condition that is relatively rare. In Cluster 

APPENDIX D 71 4, two of the projects were experimental projects dealing with learning, while two were projects designed to study classroom processes in countries where students’ mathematics, science, or literacy achievement is ­especially strong (previously shown in international tests). Projects in this cluster would have implications for both educational practice and policy. Cluster 5 was formed from the studies that were longitudinal, in order to look for particular chal- lenges in following subjects over time in different ­national contexts. These projects might have fit better (in terms of value added) under the substantive clusters. The workshop was set up in a way to explore the value-added notion in different types of projects more thoroughly. Countries Included The range was 2 to 29 countries per project. Table D3 lists the coun- tries from which collaborators came; most projects included the United States in the comparison. Cluster 2 was concentrated in post-Communist countries, and several of the projects in Cluster 1 also included countries in this area. Other clusters were spread across countries. Countries where English is spoken widely by professionals are well represented. Low-income countries (es- pecially in Africa and Latin America) are poorly represented. One of the topics of discussion at the workshop might be how to fund and otherwise encourage participation from a wide range of countries in order to build scientific capacity. SUMMARY OF RESPONSES TO PART B OF THE SURVEY Participants and Personnel (Theme 2) The open-ended questions asked about original contacts with collabora- tors, levels of training and involvement by participants, whether more than one discipline or subdiscipline was involved, cultural differences in leader- ship style, informal groups, and problems with participants’ expectations. The overwhelming impression is that personal relationships (often but not always facilitated by international congresses or formal exchange programs such as Fulbrights) were important in initiating and sustaining projects. Often there was a kind of serendipity of networking or snowballing (where one participant recruited others). Existing behavioral or educational research organizations played a primarily positive role. Doctoral students 

72 APPENDIX D TABLE D3 Countries from which Collaborators Came Country No. of Collaborations Australia 6 Belgium (French) 1 Bulgaria 3 Canada 5 Chile 1 China 2 Colombia 1 Cyprus (Greek) 1 Czech Republic 4 Denmark 3 Estonia 1 Finland 5 France 4 Germany 7 Greece 2 Guatemala 1 Hong Kong 3 Hungary 3 India 1 Israel 4 Italy 2 Japan 2 Latvia 1 Lithuania 1 Mexico 2 Netherlands 3 New Zealand 1 Norway 3 Poland 2 Portugal 2 Romania 2 Russia 5 Slovakia 2 Slovenia 2 South Africa 2 Sweden 5 Switzerland 2 Spain 1 Turkey 1 UK (or England) 7 NOTE: Almost all projects had some component in the United States. 

APPENDIX D 73 and postdocs were of vital importance in a number of projects (sometimes as initiators of research later taken up by a wider network). A collaborative spirit was mentioned by several respondents and seems to have characterized most projects. Some respondents expressed enthusi- asm about what they had learned as American psychologists from taking the perspective of other researchers. These responses conveyed the idea that problems were part of the research process or that misunderstandings pre- sented opportunities to learn about the meaning of culture as it influenced research. Intrinsic motivation, such as getting new perspectives on problems identified in earlier research, was a common theme, as was mobilizing around a big idea (e.g., understanding how to foster democracy). A desire to look at the universality (or lack of universality) of research findings from North America was either explicit or implicit in many responses. Among the valuable attributes of collaborators were a positive and open attitude, commitment to consensus, patience and persistence, communica- tion of respect for other researchers and their views, willingness to challenge received wisdom, and a sense of humor. There was no substitute for reflec- tion on firsthand experience in the cultural setting and with researchers from that setting, according to several respondents. Some suggested either offering the type of collaboration training that many international busi- nesses have developed or making available “collaboration coaches” to help maintain a productive atmosphere in an international project (especially when many countries are involved or the participants are not well known to each other). In some projects there was quite a bit of asymmetry in the level of professional preparation of the researchers. The respondent who reported a range from 6th grade education through postdoctoral training saw this as a strength (perhaps because it was a project framed in cultural psychology with an aim of identifying different perspectives on everyday life events). In other projects where there was a range of levels of training, there were some difficulties (especially when familiarity with specific protocols or method- ologies of data collection was required or when one country’s participation was slowed by having few trained personnel in comparison to other coun- tries). Capacity building is clearly a need almost everywhere (though the particular capacities may differ). Several respondents spoke of initial mistrust among participants, which required conscious efforts at consensus building (in addition to the content- oriented communication required to code data and prepare publications). Not framing the work as “an American project” and avoiding “American 

74 APPENDIX D scientific imperialism” were important almost everywhere but especially in Eastern Europe and South Africa. In some cases an “antipsychology” bias was perceived (and successfully overcome). In a few projects either senior professors or medical school faculty sought to establish a hierarchy in which they could determine the direction of the research without listening to others’ views. There were tensions between male researchers and female researchers in a few projects. Some respondents mentioned differences in pacing and sensitivities to deadlines in different countries. Conduct of the Research (Theme 3) The open-ended questions asked how methodological decisions were made, whether existing or new methods were used, and how translation and cultural adaptation were dealt with and included checks on fidelity of implementation, sampling, and time schedule. The issue of cross-disciplinary collaboration arose in answers to both Theme 2 and Theme 3 questions. Disciplinary structures differ across countries, as do the methods used and the preferred strategies associated with given disciplines. The projects involved researchers whose primary identifications were psychology, sociology, education, measurement/sta- tistics, criminology, medicine, physiology, philosophy, communications, and ethnography. Some projects dealt with this by explicitly using a mixed method design, others by negotiating about what could be learned by using different methods or taking different perspectives on a issue. One issue was the choice between using the best measure or the most comparable measure across the participating countries. There were cluster differences here, with projects in Cluster 3 understandably most concerned about fidelity and precision in the implementation of standard research protocols. In the other clusters there appeared to be more flexibility in negotiating the instruments and coding (in some cases to meet the political sensitivities of a participating country). Arriving at common definitions of constructs was vital (but often time consuming). Nearly all respondents spoke of the need for a clear focus in research questions, extensive pilot testing, monitoring of procedures, and extensive communication throughout a project. Using a logic model in planning was mentioned. Starting with a relatively simple and well-circumscribed prob- lem, understanding it in two or three cultural settings, and then building from that success to enhance the scope of the research was suggested (rather than starting with a broad or diffuse idea to be explored in many countries). 

APPENDIX D 75 Another suggestion was that a project team could create an international core of instruments or methods (on which agreement could be obtained and to which all would strictly adhere) and international options (designed by the group of researchers but open for choice by participating countries or research institutes). Practical Issues (Theme 4) The open-ended questions asked about the funding infrastructure and its management, research regulation (including institutional review boards, or IRBs), incentives, bureaucracies, visas, and communication (face-to-face and electronic). There were some differences by cluster, with the intervention proj- ects (Cluster 2) and some of the projects requiring shipping of biological samples (Cluster 3) having special difficulties. In general it appears that the projects differed with respect to practical issues according to the project’s scope, whether the research was conducted under the aegis of a strong or- ganization with established international infrastructures and policies, and in which regions the research was conducted. The opinion was expressed that psychologists are too rarely involved in government-funded “big science” international trials. That said, small grants for seed money (often from home institutions) and flexible funding at later stages (especially for low-resourced countries) were also cited as important. Many of the projects operated on a shoestring; more than one respondent reported substantial outlays of personal funds and the need to piece together funds from different sources with different requirements and time frames. Uncertainty about funding also was a source of stress. Approval by IRBs or ethics committees (the term often used in Europe) differed in complexity. Difficulties arose when the rules or expectations in a participating country differed from those in the United States or when several universities were involved. One project developed a “template” for participants to use in applying for approval from IRBs or ethics committees. The opinion was expressed that some IRB members at North American institutions base their decision on assumptions about other countries that may be outdated. In some of the bureaucratic settings, lower-level personnel appear to have felt left out of the decision chain and responded by withholding per-  Some of these issues were dealt with under the previous two themes. 

76 APPENDIX D missions or declining to approve expenditures. In a few countries there was an expectation that the U.S. researchers would pay for everything. Some projects had practical problems based on the complexity and scope of the task undertaken (e.g., videotaping and coding a total of 700 lessons in science and mathematics) or on events beyond the researcher’s control (e.g., SARS, national political incidents, earthquakes) or because of difficulties in communication (e.g., deciding on analysis and deciding to what extent observed differences are related to culture or to method). Despite great advances in electronic communication over the past decade, regular face-to-face meetings remain a vital component of suc- cessful collaborations. Meetings of subgroups of participants were often held in conjunction with international congresses. Some mentioned the importance of long-term visits. However, using face-to-face meetings as the only venue for decision making has drawbacks if every participant cannot attend every meeting. Conference calls had drawbacks noted by several respondents. Electronic communication via e-mail was essential. This also facili- tated the participation of several researchers in the editing of a text before publication. Quite a number of projects did not appear to have dedicated Web pages used for dissemination of results (as they did not provide Web addresses). Use of electronic conferences can be useful at certain phases if carefully planned to address a relatively narrow agenda or set of decisions. This may be an area for future development. Data Access and Publications (Theme 5) The open-ended questions asked about data management, sharing and release, and decisions about authorship (and more generally the communi- cation of findings). There were differences in how authorship was credited that were often associated with the number of researchers involved in the project and their types of expertise (including their ability to communicate in the language in which the publications were to be issued, which was usually English). Several projects drew up specific guidelines on authorship for publications drawing data from more than one country. In one case, these were built on the National Institute of Child Health and Development’s Child Care Study in the United States, and in another, they were based on the policies developed over many years for all studies conducted by the International Association for the Evaluation of Educational Achievement, also known as 

APPENDIX D 77 IEA, which is an international consortium of research institutes headquar- tered in Amsterdam. Usually those guidelines drew on common practice (e.g., who wrote the first draft or took the initiative on the analysis). Individual researchers in almost all projects have been allowed to pub- lish the results from their own country where they wish and with whomever they choose. This is intended to stimulate publication in the local language, which is more accessible to the communities in which the research has been conducted (a value for many of the respondents). There is considerable variability in data release, ranging from nearly full access on a CD-ROM available on request or on the Web to restric- tions on the use of data only by the collaborating researchers. The former requires more documentation, which some projects cannot afford. Some social science projects use the Interuniversity Consortium for Political and Social Research (ICPSR) as a data archive, which might be suitable for some international psychology projects as well. There is also considerable variability in the extent to which the results of research have been disseminated to audiences of policymakers and prac- titioners whose work might be informed by relevant findings. This would probably require additional funding (and assistance from those who know how to write for these audiences). RECOMMENDATIONS BASED ON THE SURVEY In compiling these recommendations, the focus has been on those that would benefit from discussion at the workshop and on those that the U.S. National Committee for the International Union of Psychological Science might assist in implementing. • Meet with funders to encourage more funding and more flexible funding. For example, encourage a new set of small grants ($15,000 and up) focused on starting new projects or funding at the end of a project to support additional publications, release of data for secondary analysis, or publications in national languages. • Support a larger role for psychologists in federally funded interna- tional multidisciplinary research. In particular, set aside training funds. • Establish a U.S. fund for supporting international collaborations, especially involving younger scholars and those from countries where capac- ity building is especially urgent (perhaps in collaboration with professional organizations). 

78 APPENDIX D • Consider funding collaborations between U.S. and Canadian re- searchers around topics of common interest. • Consider how programs such as Fulbright Senior Scholar Awards could contribute to international collaborative efforts in psychology. • Offer training to senior and junior researchers in cross-cultural/in- ternational communication (as businesses do) to reduce the tendency to believe that “the way we do psychology in the United States is the only right way” and to minimize instances where investigators from other coun- tries perceive a lack of respect or sensitivity to cultural differences. Prepare mentors and make them available. (These would be persons not directly involved in the collaboration itself who know something about both the participating countries and the subject matter of the research). • Consider offering some U.S. National Committee for the Inter­ national Union of Psychological Science meetings as venues for discussion of projects (using as a model the Board on Comparative and International Studies in Education of the National Research Council, which served this function for international educational research in the 1990s). • Organize small group meetings (or workshops) at existing interna- tional meetings to plan research and provide funds to attend these meet- ings or short-term travel as a follow-up. Researchers are more interested in discussing collaborations relating to topics in their field than talking about international collaboration in an abstract or generic sense. • Develop models for explaining the contributions that international research undertaken in a collaboration framework can make and suggest follow-through on selected topics. • Develop a network to assist international scholars in preparing ar- ticles based on international collaborative research to meet the policies and practices of U.S. journals. • Consider models for developing and disseminating measures and methods for international collaborative research in selected areas. • Develop models for disseminating the results of internationally collaborative projects (executive summaries, policy briefs for different au- diences including those in participating countries, Web pages). It can be helpful to issue some publications at the midpoint of long projects in order to keep sponsors and researchers engaged.