This is an open access article distributed under the terms of the Creative Commons
Attribution License (“Research that produces nothing but books will not suffice.”
(Lewin, 1946, p. 35)
A Science for Problem Solving
Social psychology, like much of social science, was first conceived as a means of diagnosing and correcting social problems (Comte, 1856; Lewin, 1946). Because of this conception, early social psychologists such as Kurt Lewin, Stanley Schachter, and Solomon Asch found it necessary to conduct both laboratory and field experiments—using the principles developed in the lab to identify problems and design interventions in the world in which those problems naturally occurred (Lewin, 1946). Because psychological “laws do not tell what conditions exist locally, at a given place at a given time” (Lewin, 1946, p. 44, emphasis in original), founding social psychologists believed that diagnosing specific problems in the world was at least as important to the execution of psychological science as it is to engineering or medicine. Lewin, for instance, believed that social psychology’s only route to effecting societal change was a hybrid of basic and translational research, claiming that, “above all, [social psychology] will have to include laboratory and field experiments in social change” in order to be effective (Lewin, 1946, p. 36). In other words, if social psychologists have not translated their theory into the world in which it occurs naturally, then they are every bit as useful as a medical doctor who understands anatomy perfectly, but who cannot treat a headache.
The dialogic integration of laboratory and field research—or translational science—was both central to social psychology’s beginnings and to its mission. The goal was to create a science that was at once methodologically rigorous and practical to the world of naturally occurring social problems. As Lewin, who is often regarded as the progenitor of contemporary social psychology, explains:
It is important to understand clearly that social research concerns itself with two rather different types of questions, namely the study of general laws of group life and the diagnosis of a specific situation … To act correctly, it does not suffice, however, if the engineer or the surgeon knows the general laws of physics or physiology. He [sic] has to know too the specific fact-finding called diagnosis. For any field of action both types of scientific research are needed. (Lewin, 1946, pp. 36-37)
Lewin refers to psychology as a “field of action,” by which he means that it is fundamentally concerned with understanding the worlds in which we live in order to act upon them (towards a greater good). And, central to this ability, “as in medicine,” is the “skill and ingenuity” that “both diagnosis and treatment” require (Lewin, 1946, p. 44). In other words, social science efforts to effect social change are hobbled if they do not engage the social world directly. Why, then, are the most prestigious journals not full of translational research (or, what Lewinians referred to as action research; Asch, 1959; Citron, Chein, & Harding, 1950; Citron & Harding, 1950; Lewin, 1946), particularly when translational research was the hallmark of early work on race and other social justice oriented social psychology?
The present article outlines an answer to this question. In so doing, it also articulates the three largest obstacles facing translational research today, namely: the relatively higher financial costs, the difficulty pursuing theoretical advancements in translational research, and the time it takes to navigate logistical complications. After reviewing the historical origins of these three obstacles, we put forth one model of contemporary translational research with an eye towards ways in which the discipline can reanimate the tradition of social psychological work targeting societal change.
Whatever Happened to Translational Research? A Historical Perspective
Social psychology’s popularity owes much to its roots as a science dedicated to understanding social problems in the world. Kurt Lewin, who pioneered social psychology in the United States, believed that social psychologists should engage in two complementary endeavors: to unravel the universal laws of human behavior, and to apply (and to modify) these principles to specific contexts or societal problems of our times. Famously stating that “there is nothing more practical as a good theory” (Lewin, 1951, p. 169), he believed that solving problems as they exist in the world advances both practice and theory.
Indeed, social psychology was in Lewin’s time what we may call today a highly ‘translational’ field. From Lewin’s action research designed to target social change in issues as diverse as improving interracial relations and changing food habits, to Milgram’s and Asch’s concern with the evils that conformity can produce, to Kenneth and Mamie Clark’s (1950) “doll studies” famously referenced in Footnote 11 of the 1954 Brown v. Board of Education Supreme Court decision, psychological research did not hesitate to approach the burning issues of the times.
While still viewing Lewin, Milgram, and others as its founding figures, social psychology has changed remarkably since the mid-1950s, when psychological research began increasingly—and then almost exclusively—to move into the lab (Dovidio, 2001; Fiske & Taylor, 2013; Manis, 1977). Though an exhaustive historiography of translational research is beyond the scope of this article, it is sufficient for our purposes to note that the decline of translational research on social issues and the rise of experimental psychology approaches were both simultaneous and, perhaps, even related.
At the time when Leon Festinger (1957) was first articulating his theory of cognitive dissonance—the phenomenon that sparked the widespread use of laboratory methods in social psychology—the field maintained a foot in both experimental and field research (Cartwright, 1979; Duckitt, 1992). For instance, phenomena like cognitive dissonance were both inspired by and studied in their real world contexts in addition to in lab studies. However, as social psychology was struggling to promote itself as a science with a set of unified and rigorous methods of causal inferences, the favorability of experimental laboratory methods became increasingly enticing.
The result was an arresting unification of the field, with scholars at the height of the shift to social cognition remarking that its dominance made it “difficult to determine what aspects of social psychology could legitimately be regarded as ‘noncognitive’” (Manis, 1977, p. 550). Research on persuasion, emotions, the self, intergroup conflict, and nearly every other facet of social life became ripe for in-lab exploration, and brought clear, often counter-intuitive findings that delighted both other scientists and the lay public (Fiske & Taylor, 2013; Manis, 1977). In addition to creating a widely shared methodological toolbox, and popularizing the discipline, the rise of lab research had other attractive features. As Manis (1977) writes:
[This] approach to social psychology seems likely to retain its dominance in the coming years because of its inherent virtues, its compatibility with the cultural zeitgeist, and its well-developed theoretical and experimental paradigms that enable ambitious investigators to complete systematic research programs within a manageable budget of time and money [emphasis added]. These considerations (including the last-mentioned, matters of convenience) are important in any scientific movement, and it seems clear that they are, at present, quite favorable for further developments in the cognitive approach to social psychology. (Manis, 1977, p. 563)
Manis’s note refers to the relative ease of conducting laboratory research. For instance, in many of Festinger’s cognitive dissonance paradigms, individuals were simply brought into small lab rooms, given instructions, told to circle numbers on a piece of paper, and excused (Festinger, 1957). As Manis noted, moving from a research paradigm of translational research in which housewives must be recruited and convinced to cook intestines (Lewin, 1953) to one that might be described as a conversation between experimenter and participant (with some number circling thrown in) is appealing for anyone familiar with the hassles of field research. And, if the relatively economical paradigm is attractive to established researchers, it is surely more so to early career scholars eager to produce a volume of research quickly before tenure.
The past several decades, therefore, have seen the balance between laboratory and field research shift, with social psychologists increasingly moving to establish a culture that favors laboratory experimentation (Cartwright, 1979). The upside of this shift is that the past half-century has seen a tremendous increase in the production of excellent social psychology. The downsides, however, are myriad and militate against the very research that launched the field.
For instance, the behavioral aspect of psychological research was substituted for self-reported attitudes, feelings, and behavioral intentions. In other words, the field became “more interested in understanding the internal workings of the mind and brain rather than behavioral outcomes” (Baumeister, Vohs, & Funder, 2007, p. 396). This focus on internal processes is also evident in the field’s choice of operationalizations. As laboratory experiments dwarf field experiments, it may be said that social psychology has become a “science of self-reports and finger movements” (Baumeister et al., 2007, p. 396). Similarly, as research was driven into the laboratory, the field has increasingly retreated from the study of real world settings. Consequently, while some social psychologists previously considered developing an area of research that analyzed the facets of contexts (in much the same way we do for personalities; Brunswik, 1956), some contemporary scholars are left to wonder—literally—what has happened to the “social” in social psychology (Greenwood, 2004).
In addition to reductions in the mundane realism of social psychological research, the push to publish large numbers of papers combined with the tacit assumption that experimental methods and the laboratory settings allow for broad generalizations from narrow populations has led to the increased reliance on convenience samples—often college sophomores (P. J. Henry, 2008; Sears, 1986). As Sears has famously argued, the field saw a marked decline in the use of random sampling from the 1960s forward, and a corresponding increase in the use of convenience sampling. An analysis from the mid-1980s concluded that the most popular convenience samples possessed stronger cognitive skills, a weaker sense of self, less crystalized attitudes, more unstable peer groups, and were more likely to comply with authority—all important factors in the history of social cognition (Sears, 1986). In a related analysis, P. J. Henry (2008) finds that the reliance on student samples in the intervening twenty-plus years continues to raise “metatheoretical concerns” about the topics and conclusions we can draw from such samples. Moreover, Henrich, Heine, and Norenzayan (2010) argue that the field’s near exclusive focus on samples from “Western, Educated, Industrialized, Rich and Democratic (WEIRD)” societies may significantly undermine our ability to draw conclusions about human nature, because such samples are actually outliers in key behavioral domains. These authors review wide-ranging evidence that “WEIRD” samples are far from the mean in everything from visual perception (Segall, Campbell, & Herskovits, 1966) to positive views of the self (Heine & Hamamura, 2007; Mezulis, Abramson, Hyde, & Hankin, 2004) and moral reasoning (e.g., Haidt & Graham, 2007). As the field moves increasingly to online sampling via Mechanical Turk and other means, this may remedy an overreliance on college students, but introduce a new set of biases in experiments designed to reveal human universals (Buhrmester, Kwang, & Gosling, 2011; Paolacci, Chandler, & Ipeirotis, 2010).
Still, perhaps the most important transformation resulting from the dominance of laboratory studies involves the field’s publication demands and incentive structures. The shift to laboratory studies and convenience samples—especially ones compelled for academic reasons to participate in research—has made it relatively easy to conduct a significant volume of research in a short period of time. If, for instance, the translational researchers Festinger and Schachter (Festinger, Riecken, & Schachter, 1956) studied cognitive dissonance by observing the actual behavior of the followers of doomsday cults, their social cognitive colleagues could study the same phenomenon using a research assistant, a pencil, and paper. This radical increase in research efficiency has, over time, shifted the standards of productivity, with top journals requiring more studies and promotions becoming contingent on higher numbers of publications.
The shift in the culture of academic social psychology is, in turn, responsible for the obstacles to conducting translational research today. Specifically, because of the exceptional increase in research efficiency afforded by in-lab research methods (including a reliance on convenience sampling, non-behavioral measurements, and laboratory-based experiments) the costs of doing translational research are proportionally higher for scholars looking to translate their research to the field—particularly for early-career researchers.
With this historical context as a backdrop, the next section enumerates the three largest impediments to translational research: financial costs, the difficulty in translating “diagnoses” into theoretical innovation, and the amount of time logistical considerations take. The hope, of course, is that by identifying the obstacles it becomes easier for the field to address and remedy them—the theme of the fourth section.
Three Barriers to Translational Research
Given the way in which the move to laboratory studies has shaped the production of social psychological knowledge, it is not difficult to imagine how non-laboratory, translational social research might be at a competitive disadvantage. Any research that requires non-convenience sampling, observational and/or behavioral data, and/or cannot be shoehorned into a factorial experiment is simply (and demonstrably) “harder” to conduct than traditional laboratory research. However, a brief description of the three major impediments (cost, theory, and logistics) will be useful in reviewing successful models of contemporary translational research and outlining suggestions for its wider adoption.
The Cost of Translation
Put simply, few people other than college undergraduates can be persuaded to participate in research for free. And even if one is fortunate enough to conduct research on populations that may volunteer their time (e.g., high school students or volunteer workers), there are often costs associated either with traveling to the research site, setting up data collection procedures, or paying for the staff necessary to navigate significant logistic burdens that are uncommon in laboratory research. For instance, recent innovative research on the role of peer influence in reducing harassment in schools (Paluck & Shepherd, 2012) and on how interracial contact is experienced in educational settings (Al Ramiah, Hewstone, Voci, Cairns, & Hughes, 2013) would not have been possible without sizeable grants to the principle investigators. While research in neuroscience and psychophysiology also requires significant dollars, there is no section of the largest federal funding sources in the US (e.g., the National Science Foundation or the National Institutes of Health) or Europe that targets translational research. Additionally, individuals interested in studying peer influence and interracial contact have long been able to do so without such grants—provided that they conduct that research within a laboratory. Consequently, it is likely that, whatever costs one might incur from studying a phenomenon within a laboratory, they will be greatly enlarged for a translational version of the same project. These financial costs could be offset by the professional rewards for undergoing such expensive research. In the next section, however, we reveal that they (largely) are not.
The Difficulty with Advancing Theory
Particularly in the wake of the transition to laboratory research, the field of social psychology has valued the theoretical contribution of research far more than any particular application of it. Though research on some topics, such as stereotype threat (Steele & Aronson, 1995) and innovations in intergroup conflict (e.g., Richeson & Shelton, 2003), receive significant praise both within the field and outside of it, the privileging of “basic” research over “applied” is pervasive (Baumeister et al., 2007). Consequently, research that fails to advance the field’s theoretical understanding of a given topic—usually by identifying a narrow theoretical mechanism for a behavior—is at a competitive disadvantage.
While this standard of scientific excellence is difficult to achieve in and of itself, it is far more difficult outside of the controlled environment of a laboratory. Consequently, despite the extra time and expense that translational research frequently requires, it is often still more difficult to produce the kind of scholarship that prestigious social psychology journals publish. And, while there are specialty journals (such as Basic and Applied Social Psychology) that cater to translational research, they are not cited with the frequency of journals that award no extra consideration for the work of translating psychological research into the field. The result, then, is that translational research is expensive, seems less scientifically significant, and is more difficult to place in top journals. And, as we discuss below, it also takes more time.
The Time to Translate Logistics
In much the same way that convenience sampling is almost always cheaper than random sampling or sampling specific populations, convenience sampling is also … more convenient. That is, it takes less time to recruit from a ready-made pool of college participants or a near limitless pool of low-wage online participants than it does from any other population. And, because of this alone, translational research is more time consuming than most basic research. However, in addition to the hassle of arranging for a broader (or narrower, as the case may be) sample, there is another significant demand on translational researchers’ time: managing the logistics of the project.
Whereas the logistic concerns of basic research can often be left up to graduate students, working with primary schools (Al Ramiah et al., 2013; Paluck & Shepherd, 2012), healthcare providers (Hagiwara et al., 2013; Penner et al., 2009, 2010), or police departments (Correll, Park, Judd, Wittenbrink, Sadler, & Keesee, 2007; Eberhardt, Goff, Purdie, & Davies, 2004; Goff, Epstein, Mentovich, & Reddy, 2013; Goff, Jackson, Nichols, & Di Leone, 2013) requires a larger time investment for all individuals involved. Because access to a population and/or context is both more important to translational research and more tenuous than it is in basic research contexts, managing logistics invariably means managing relationships as well, requiring regular communication between researchers and practitioners. Additionally, translational research frequently requires learning to navigate new legal requirements, a new institutional culture, and even providing additional deliverables to the research partner (e.g., a final project report or trainings) in order to facilitate the project.
Finally, each of these potentially time-consuming additions to the research process will likely be new and unique to researchers for each project. Consequently, the learning curve for each translational project is far steeper than traditional laboratory research. And, given the unique circumstances of each research partnership, the learning curve may remain steep with each new project regardless of how much a researcher has learned from the last one. The lags in communication, false starts, and trial by error that are universal in the scientific process are, consequently, often magnified as scholars attempt to translate from the laboratory to the field. A scholar who has spent a career learning how to make college participants demonstrate subtle racial biases may still find it difficult—and time consuming—to learn how to make hospital directors and police chiefs comfortable with the idea of having their organization’s racism “diagnosed.”
Consequently, the more translational research one does, the more time is spent on logistic concerns as compared to one’s colleagues. Because of the competitive disadvantage this produces in a culture that values numbers of publications so highly, this represents another significant professional impediment to translational research.
Still, despite these impediments, translational psychology appears to be making something of a comeback. This change can be attributed to an increasing number of practitioners and social policy makers who are interested in the implementation of social psychology principles to their cause. Similarly, the rise of behavioral economics—a kind of translational social psychology—has revealed the potential value of answering social psychological questions in the context that most urgently raise them. At the same time there has been a renewed understanding in the field that translating psychological research provides much needed opportunities to replicate, generalize, and refine laboratory findings. The next section reviews one model for addressing some of these impediments: the Mind Science Justice (MSJ) model. In reviewing the MSJ model, we intend to demonstrate one possible set of solutions to the obstacles translational research faces, some of which have been used more broadly.
A Model of Translational Justice in Policing Research
As noted above, the recent rise in translational science’s status has neither been accomplished without significant funding nor absent theoretical contributions—and subsequent publications in top journals. That is, contemporary translational research has neither identified a way to conduct research without money nor convinced journals to publish socially applied research. Consequently, the time and logistical concerns of translational research are the natural targets for reducing the burden on translational researchers. In articulating the MSJ research model as employed by the Center for Policing Equity (CPE), we will also highlight both difficulties and solutions that are common to social psychological research conducted in translational contexts.
Basics of the MSJ Research Model
The CPE is a research and action think tank that works with police departments to conduct original research in the interest of improving equity in police organizations and the delivery of police services (http://www.policingequity.org). The goal of the organization is to use the tools of social science—particularly social psychology—to equip police departments and communities to reduce identity based injustice (e.g., racism, sexism, homophobia, etc.). These stated goals serve as an umbrella to the scholarly missions of the organization, which are simultaneously to “give away” social psychology to law enforcement (Miller, 1969, p. 1017: “Our responsibility is less to assume the role of experts and try to apply psychology ourselves than to give it away to the people who really need it.”) and to provide the discipline with a model of successful translational research. These goals are shared by a large number of research think tanks and advocacy organizations, which led CPE’s leadership to develop the MSJ model, since adopted and/or borrowed by the American Values Institute, the Laboratory to Combat Human Trafficking, and John Jay’s Center on Race, Crime and Justice, among others.
Establishing an organization capable of achieving these goals requires resolving a host of domain-specific complications (e.g., teaching social psychologists the inner workings of multiple data management software systems). In addition, creating an organization designed to maximize research efficiency and minimize logistic difficulties, revealed a set of five broad challenges. Specifically, how to (1) address concerns surrounding legal liability; (2) streamline relational concerns; (3) satisfy the transactional concerns of research partners; (4) offset researcher productivity concerns; and (5) allay partner and researcher reputational concerns. In addressing these issues, the MSJ research model seeks to produce a more robust set of principles with which to approach translational research.
The Five Challenges
Legal Liability
Whether an MSJ partner provides educational data (Al Ramiah et al., 2013; Cohen, Garcia, Purdie-Vaughns, Apfel, & Brzustoski, 2009; Cook, Purdie-Vaughns, Garcia, & Cohen, 2012; Paluck & Shepherd, 2012), healthcare data (Hagiwara et al., 2013; Penner et al., 2009, 2010), or law enforcement data (Correll et al., 2007; Eberhardt et al., 2004; Goff, Epstein, et al., 2013), research partners are frequently quite concerned about the legal liability attached to sharing data with researchers. For instance, police departments working with CPE were primarily concerned with three potential consequences. The first was that police executives in some jurisdictions were concerned that releasing police data to anyone meant a department was voiding any right to keep those data confidential in case of subpoena or public records requests. In response to this, CPE worked with university and department lawyers to craft a legal agreement—a Memorandum of Understanding (MoU)—that protected departments against this possibility and indicated that legal counsel at participating universities would aid police departments in keeping records confidential.
A second area of concern was that, once data were made available to researchers, they could be subpoenaed or otherwise leaked. To address this concern, university and police attorneys crafted a legal agreement that committed university lawyers to fighting the possibility of a confidentiality violation. Standard psychology human subject protocols alleviated most other fears of data leakage.
Finally, the third concern was that a research finding (e.g., evidence of a department having high levels of racial bias) could leave a department vulnerable to litigation. This concern was by far the most difficult to resolve. Because CPE’s goal is to help law enforcement while developing a better science and better scientific understanding of the issues, the solution was to allow for a “right of first refusal.” This policy entails informing law enforcement executives beforehand of major research results, providing them with the opportunity to be anonymous in the publication of the results (e.g., the San Jose Police Department, for instance, would become a “large police department on the West Coast”), and affording them a reasonable time frame (currently 30 days unless otherwise specified) in which to make that determination. During this period, they can choose to implement solutions, inform the press of these initiatives, or do nothing. And all of this was memorialized in the same binding MoU.
Together, these solutions alleviated the bulk of partner law enforcement’s legal concerns. Importantly, with a sample MoU, and an experienced team of dedicated university lawyers, new researchers need not re-invent the wheel. Consequently, the legal concerns of new partner departments or surrounding new projects are already addressed by a standard agreement. Although partners in other domains are likely to differ in their specific concerns, the same approach to legal liability can be helpful for other researchers working with sensitive data, which is why it is an important feature of the MSJ model for translational research.
Relational Concerns
Because translational research so often requires access to resources, participants, or contexts not typically available to traditional academics, it is often necessary to engage in a research partnership with non-academics. Moreover, because these partners control access, maintaining a positive relationship with that partner is crucial to the successful completion of the research. Consequently, the MSJ model requires that regular communication be prioritized in the execution of a project.
In the case of CPE research, law enforcement executives are understandably wary of researchers possibly revealing the moral failings of the organization for which they are responsible. Moreover, there is a cultural tendency among law enforcement professionals to value those who are immersed in their jobs (Adlam, 1982; Lorr & Strack, 1994), which requires researchers to be in regular communication with law enforcement practitioners (weekly or monthly is usually a requirement) and to be able to respond to their partners quickly if questions arise. Taking two weeks to reply to an email will not necessarily strain a long-distance research collaboration between social psychologists, but it will almost certainly doom one with a police or sheriff’s department.
Additionally, given law enforcement’s history of racial conflict and subsequent litigation, it is also important to begin with at least two affirmative assumptions about the partner. First, it is important to begin with an assumption that everyone involved wants to do the right thing—that is, that the research partners are not bigots. The affirmative assumption on the part of the “racism experts” (or “sexism experts” or “social justice experts” more broadly) regarding this issue allows for practitioners to avoid the understandable defensiveness that frequently occurs when racism is discussed—particularly in the context of policing as with the CPE.
Second, it is important that the researchers share the affirmative assumption that ridding a department of racism is both a worthy goal and a difficult one. One benefit of this assumption is that it is supported by significant social psychological research (Dasgupta & Greenwald, 2001; Devine, 1989; Dovidio, 2001). Another is that executives often feel as if others assume that the answer to questions of racial mistrust is simply “not to be racist,” and, as a result, feel that the existence of racism is seen as proof of either their own racism or a lack of will to fix it. Any employer, however, is constrained in the quality of the services it provides by the quality of the job applicants it receives. In other words, because law enforcement hires humans, they will hire humans whose biases reflect the biases of the community it serves. Consequently, the job of trying to eradicate bias within a department is at least as complicated as trying to do the same in society at large.
In the case of CPE, because law enforcement culture can be insular (Loftus, 2009; Waddington, 1999), a willingness to be present, an authentic desire to help the department, and the affirmative assumption that they are trying to do the right thing allay reasonable fears, facilitating the research process. While these may seem like obvious relationship-maintenance strategies, they are each beyond the considerations psychological scientists need to afford to human participants that volunteer online or sign up for a study “in exchange for partial course credit.” Additionally, it can be difficult for anyone to hold questions of intent in abeyance when confronted with stark racial inequities, making the simple “benefit of the doubt” a more effortful gift in these research contexts than in other relationships. In recognition of these concerns, the MSJ model requires that full-time staff be tasked with facilitating regular communication and providing training to new researchers in relational approaches to research partners.
Transactional Concerns
Having attended to basic relational concerns, many translational research relationships are also transactional—that is, the partner wants something more than a positive relationship in return for granting research access. Consequently, the MSJ model requires that research partners receive something of value in exchange for what they give in perceived risk to the organization and in the substantial time and energy it takes to permit research access (e.g., hours spent processing and redacting data, allowing space for research, communicating with line officers about the importance of participating in research, etc.). To achieve this end, CPE researchers typically produce a comprehensive and actionable research report that is separate from any peer-reviewed publication. Although producing such reports can be time-consuming, doing so is worth maintaining the mutually beneficial relationship and having a positive effect on actual department policy—which, in turn, can facilitate new research opportunities. Consequently, the CPE employs staff and postdoctoral scholars dedicated to assisting principle investigators in providing these reports in a timely manner.
Productivity Concerns
Though the MSJ model attempts to streamline the research process, working with non-academic partners is still time consuming, and there is no way to delegate critical thinking to others. Consequently, the MSJ model includes two modest structures designed to further offset concerns with productivity. First, researchers are encouraged to share data collection responsibilities with other research teams, spreading out the workload and, potentially, sparking new research collaborations. Second, staff members are often tasked with collecting large datasets that allow researchers the ability to investigate multiple hypotheses and, ideally, produce multiple papers from a single data collection. Though these are modest productivity offsets, they are surely worthwhile, if only as insurance policies against the messy nature of field data.
Reputational Concerns
With legal, relational, transactional, and productivity concerns partially addressed, the last concern that the MSJ model addresses is the reputational concerns of both researchers and law enforcement partners. While researchers may be concerned with perceptions of their objectivity, partner executives are concerned, again, with the perception of the moral authority of their departments.
In order to remedy researchers’ concerns, the MSJ model forbids accepting money from research partnerships.i This ensures both the perception and reality of research objectivity. The MSJ model also requires that the MoU includes language guaranteeing that researchers are given access to all relevant data necessary to answer their research question, and that partner departments do not attempt to tamper with public reports or research publications.
In order to remedy research partner’s concerns—beyond the legal protections described above—the standard MSJ model MoU forbids researchers from talking with the press about the research they are doing within a city during the tenure of the project, unless explicitly cleared to do so by law enforcement. The MSJ model also includes researcher agreements that encourage scholars not to serve as expert witnesses in litigation against a department on the subject they researched within that department, though this is not a precondition of research access.
A Final Note on the MSJ Model
Having negotiated solutions to common concerns of translational research partnerships, the MSJ model encourages research centers to provide legal counsel, staff resources, and relationship training in the service of this non-traditional research model. The goal is to limit the negative impact that participating in translational research might have on scholars whose intellectual trajectory might lead them towards work in the vein of Lewin, Asch, and Schachter. In so doing, it may be possible to encourage both an increase in translational research within social psychology and a reframing of its import.
Conclusions
Despite the dominance of the social cognitive laboratory paradigm, social psychologists are rediscovering translational research. In an era when “big data” and “evidenced-based practices” shape an increasing percentage of important societal spheres such as education, law enforcement, healthcare, and public policy, the time would seem right for this re-discovery. And, in fact, several factors external to the discipline are likely to nudge the field in that direction.
Major granting agencies in the United States, including the National Science Foundation (NSF) and the National Institutes of Health, have tightened their requirements regarding the “broader impact” of proposals—making intervention research more likely to get funded. For instance, the NSF recently increased the evaluative importance of “broader implications” in the social psychological research it funds in an effort to encourage social psychologists to translate theoretical innovations to the social worlds they can most improve (Dovidio, Penner, & Albrecht, 2013). Similarly, with the rise of behavioral economics, increasing numbers of policy leaders are relying on the principles of social psychology to shape day-to-day practice as well as long-term policy agendas. From the use of CompStat in police deployment (a system that uses crime data [e.g., time and location] to optimize police deployment and enforcement; V. E. Henry, 2002) to the use of social psychologists and behavioral economists in presidential campaigns and dynamic policy implementation strategies (Benedictus, 2013; Carey, 2012), translational social psychology—despite its recent scholarly absence—boasts plentiful avenues in which to make direct societal change.
Consequently, we offer four modest suggestions to further facilitate the growth of this original portion of the science: First, create infrastructures designed to address many of the logistic constraints. For instance, the American Psychological Association, the Federation of Associations in Behavioral and Brain Sciences, and NSF could devote resources to providing infrastructures of the kind the CPE has done for research on mind sciences and policing. The field would also benefit from more proactive and widespread sharing of successful models (e.g., sharing CPE’s sample MoU). Second, reward innovators in translational science. Early-career and established scholars can be acknowledged for excellence in “diagnosis,” as Lewin might put it. In addition to specific recognition of individuals, journals and departmental evaluation committees could adopt specific metrics for positively weighting translational research (or studies within a program of research) in order to offset the professional costs of doing the work. Third, create a community of scholars who value translational research. The goal is that these scholars can serve as reviewers of grants and papers that come out of the work. Fourth, engage policymakers, with the goal of demonstrating the power of this kind of research. The present administrations in both the White House and 10 Downing Street have demonstrated a willingness to rely on behavioral scientists in making policy and campaign decisions—particularly ones with translational experience. Consequently, though the obstacles are numerous and severe, there is once again a population hungry for “diagnoses.” If the field of social psychology can see fit to provide that science, the field would be well positioned to effect evidence-based societal change.