Research Profiles: Prolegomena to a New Perspective on Innovation Management
Sandia National Laboratories, USA
Southern Illinois University, USA
University of Maryland, USA
Despite the increasing importance of the management of research for innovation, the range of differences among types of research, as well as projects and programs, is not adequately captured in current theories of either project or organizational innovation. This chapter offers preliminary discussions for a new perspective about alternative styles of managementfor different types of research, whether basic, applied, product development, manufacturing, quality control or marketing. Based on these discussions, the chapter proposes a framework for a new perspective of innovation management, called Research Profiles, which is derived from a literature review and extensive field research. This new perspective delineates four research profiles on the basis of two dimensions of research objectives and two dimensions of research tasks. In matching the research objectives and tasks, we identify inherent dilemmas that managers must address and this developing perspective suggests some appropriate research management approaches.
Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Despite the central importance of scientific and technological research, including product development, for national competitiveness and security, at present there is not an adequate theory about the appropriate managerial styles needed to address alternative kinds of research objectives at the research project, program or inter-organizational level. Organizational innovation theory stemming from Burns and Stalker (1961) typically focuses on the entire organization and, we would suggest, one organizational model (the organic organization), rather than recognizing the existence of different kinds of research work. More critically, the organic model does not include either the concept of complexity (Brown & Eisenhardt, 1995; Hage, 1999) or external networks of expertise, which are precisely the ones that are increasingly important in the growth ofknowledge network communities (Mohrman, Galbraith, & Monge, 2004; Shinn,
2002) , and the spread of inter-organizational relationships (Alter & Hage, 1993; Hagedoorn & Duysters, 2002; Powell, 1998; Powell, Koput, & Smith-Doerr, 1996; Van De Ven & Polley, 1992). Indeed, in the organizational innovation literature there is only one study that examines the structure and performance of research laboratories and it does not include external relationships of any kind (Hull, 1988).
Although we are beginning to see an increasing number of studies of research labs (Brown, 1997; Joly & Mangematin, 1996; Jordan, Streit, & Matia - sek, 2003; Menke, 1997), inter-organizational alliances (Gomes-Casseres, 1996) and a few studies of research consortia (Browing, Beyer, and Shelter, 1995), the fact remains that none of these studies have connected the measurement of scientific and technological research objectives, to the nature of the research tasks and their appropriate managerial styles. The research literatures cited above stand largely in isolation, often ignoring other kinds of research work. Specifically, the level of the project is overlooked, which is a smaller unit than the organization, the whole organization, and inter-organizational networks of various kinds. Indeed, what makes a proposed theory of management styles necessary is the considerable range in the ways scientific and technology research is organized. While many small research projects funded by the National Science Foundation (NSF) and the National Institutes of Health (NIH), such as those found in academia, tend to be the standard structure, a considerable amount of research is conducted in large-scale organizations and programs, such as mission agencies like the National Aeronautics and Space Administration (NASA) and the National Oceanic and Atmospheric Administration (NOAA), as well as large scale inter-organizational research programs such as the Human Genome Project. For the same reason, the new and growing literature on projects (Brown and Eisenstadt, 1995) overlooks what might be called “Big Science” as represented in the research conducted at the large national and international laboratories such as Argonne in the US and CERN in Switzerland.
Further, as Clarke (2002) has discussed in comprehensive detail, the management of a large number of researchers is very different from the typical management issues involved in contemporary firms or public bureaucracies. Among other differences are the oft-cited assertions that researchers are more motivated by intellectual curiosity than monetary compensation, the longer and more uncertain time horizons for successful objectives, and, perhaps most importantly, work that is seldom standardized and difficult to evaluate.
While a theory about the diversity of research management styles would necessarily differ from more general theories of organizations, the logic in the construction of our perspective is basically the same. First, one must specify particular kinds of research objectives and identify the potential trade-offs. Then one must also distinguish different kinds of research work and tasks. Finally, the management styles appropriate for the linking of
the typology of research tasks with a typology of research objectives at the levels of project, program and inter-organization networks have to be determined. In this chapter, we present our argument for a diversity of research management styles in three sections. First, we provide a more detailed justification of the need for our perspective on of management styles. Second, we specify a typology of research work and a typology of research objectives and provide a theoretical linkage between the two. Finally, we offer our proposed view of Research Profiles and discuss the managerial styles necessitated by the kinds of management challenges that are presented in each profile’s combination of research tasks and research objectives.
Essentially, we identify two lines of argument for justifying a new perspective of management styles. We take as a given that the central importance of scientific and technological research, not only for competitiveness but many other national goals, speaks to the need for greater attention. First, the closest appropriate specialty, namely organizational innovation research, needs to be altered in the light of new theoretical and conceptual developments (see Hage, 1999). Second, the literature on scientific and technological research is missing a conceptual apparatus that would allow for the accumulation of findings across a myriad of studies.
The first argument is based on the fact that most current and past innovation research continues to be dominated by Burns and Stalker’s (1961) seminal work on the organic model. In many ways, this model has become outdated. For example, the model is largely focused on new product development by engineers, rather than scientific and technological research. This model also largely ignores the concept of complexity, which has proven to be fundamental when combined with the organic model and the strategy of risk-taking in explaining differences in innovation rates (Hage, 1999). Finally, as argued previously, the organic model is not designed to handle the varying sizes of units in which research is conducted, from the very small project within an organization to the very large mission agencies such as NOAA and NASA or a research consortium such as SE - MATECH (Browning, 1995) or various strategic alliances (Gomes-Casseres, 1996).
More recently, the importance of both the organizational learning perspective (Grant, 1996; Kim & Wilemon, 2007) and the theory of idea innovation networks (Hage & Hollingsworth, 2000; Kline & Rosenberg, 1986) are implicitly calling attention to the importance of research as a mechanism for learning. Further, these perspectives suggest the importance of inter-organizational networks that connect two or more ofthe six arenas of research: basic, applied research, product development, manufacturing research, quality research, and commercialization research. In fact, the latter literature necessitates a considerable revision of the organic model, specifically to allow for external relationships.
In short, the organic model, and organizational innovation research in general, needs to be linked in various ways to the research project literature (Brown & Eisenhardt, 1995; Hobday, 1998; Shenhar, 2001) and the abundant literature on inter-organizational relationships cited above. One of the reasons for this is the dramatic changes that have occurred because of the explosive growth and globalization of R&D and the way in which science and technology are evolving (Hage & Hollingsworth, 2000). Furthermore, the linkages between arenas (for example basic and applied) of research are usually at the project/program level and therefore research projects and the problems of how to manage them should be the analytical focus.
The second line of reasoning that justifies the need for a theory of research management styles is that much of the social studies of science literature have emphasized concrete categories rather than general dimensions that would allow one to accumulate evidence about how best to manage research. For example, the outcomes of research are typically measured in concrete categories, such as papers, patents, peer review assessments, and citations. As we discuss below, we propose a reconceptualization of these ideas so that they can be used as general dimensions, such as the degree of radicalness ofthe outcome of the research or the scope ofthe research outcome, that can be applied to basic or applied research, product development research, manufacturing research, and so on. Similarly, research work is often described by such terms as scientific and technological research or the specific content, i. e. physics, chemistry, or biology. Again, there is a need for general dimensions of research tasks that can be applied across these various situations such as the degree of complexity or diversity in the research team and the size ofthe research program or inter-organizational network. Until these general dimensions are developed and linked, it is not possible to identify the correct managerial practices for linking the nature ofthe research task with the kind of desired outcome. In summary, the increasing importance of research and these two lines of reasoning provide a compelling case for the importance of our perspective of Research Profiles based on tasks, objectives, as well managerial styles and managerial challenges.
What characteristics might one desire in a categorization of research management styles? Ideally, one should encompass dimensions that tap into the fundamental dilemmas, tensions and problems of conducting research. Since our interest is in making meaningful distinctions, we want to isolate multiple dimensions of research objectives and tasks. Another concern is to connect choices about objectives as much as possible to the existing literature and, in particular, to the literature on innovation.
To provide new insights about research management styles across a wide range of different kinds of research projects, programs and inter-organizational networks, we conducted an inductively-based exploratory study, funded by the U. S. Department of Energy (DOE), to help identify the critical factors facing the research workers and managers. Utilizing the Competing Values Framework as an organizing framework for understanding the pursuit of research (Cameron & Quinn, 1998; Quinn & Rohrbaugh, 1983), a number of focus groups were conducted with scientific researchers to more specifically identify attributes of a research environment. The focus group discussions uncovered a number of unique tensions in managing research, which demonstrated not only that competing values exist in the research environment, but that these values differ depending on the research objectives and tasks. For example, a tension mentioned often in these discussions was between researchers’ desire and need for autonomy and management’s desire and need to focus research and meet deadlines. In their study of productive climates for scientists, Pelz and Andrews (1976), found similar tensions such as these prevalent and important to consider in managing scientists because oftheir significant impact on performance. The findings of the exploratory study and subsequent surveys suggest that the diversity of research projects, programs and organizations can be sorted according to two primary dimensions for research objectives and two primary dimensions for describing the nature of the research work or task.
We define the research objectives dimension as Degree ofRadicalness in the Scientific or Techno - logicalAdvance on a continuum from incremental to radical. In slight contrast to the Schumpeterian notion of the degree of radicalness, which focuses primarily on the competitive impact of an invention (Dahlin & Behrens, 2005), we would argue that the degree of radicalness in research includes the degree of change in the state of the art, the centrality of the research problem, and the discovery of a pattern that upsets existing theory or a technology that creates a market niche. For scientific research, the task environment is the knowledge world or “the state of the art,” that is, how much is known, and what is considered to be an important scientific concern or requirement. Radical advances in science sometimes occur when a central problem is solved, such as the identification of the structure of DNA (Judson, 1979). Sometimes this also happens when a major discovery is made, such as the observation of the first candidate black hole in 1971 (Cygnus X-1), or when a research finding challenges an existing theory, such as the discovery of skeletal remains in the Americas estimated to over 1,000 years older than it was theorized that the Americas were colonized or the discovery in China of fossils of dinosaurs with feathers.
The second dimension of research objectives is the Scope of Focus, a continuum from narrow to broad, defined by the number of variables or processes or components or the number of levels or systems involved, or the extremeness of the environments of the work. In product development, this includes the number of performances affected as well as the need to change supply chains and distribution chains and create idea innovation networks or strategic alliances, as discussed in great detail by Shenhar (2001) and Hobday (1998). The challenge is adapting these concepts of scope from the product development and industrial innovation literature for scientific and technological research.
As we have already observed, the question of the amount of the scientific advance can involve multiple outcomes, that is, the number of variables or processes that are being researched at the same time. Some disciplines have a systemic quality, that is, a large number of variables have to be considered at the same time. Further, not all scientific problems can be approached with small research teams; some ofthem require a large scale focus. While progress in mapping and sequencing the human genome could have proceeded with small research teams, it took a large-scale inter - organizational program to coordinate a range of efforts so that the time needed to complete the entire genome was appreciably lessened. Indeed, the systemic quality of some types of research is frequently overlooked as a critical dimension to the scientific problem. One example of what might be called a systemic problem is research on the weather, which encompasses both oceanic and atmospheric systems. NOAA was created in 1970 to unify and coordinate the government’s research efforts on various aspects of the global environmental system, including the National Weather Service (NWS). Altogether, the scope ofNOAA’s mandate necessitates quite expensive and specialized equipment and teams to collect the relevant data, including satellites, ships, buoys, and planes. Indeed, one could argue that the choice to conduct large scale data collection represents one form of a broad scope focus.
Just as science has a systemic quality in some areas that cannot be easily divided into small projects, there are technical systems that necessitate a large program of product development that may last many years. Leifer and colleagues
(2000) observe in a series of case studies that this is a common characteristic of radical product innovation. Further, the research involves not simply the program, but also supporting technologies as well. For example, research on high speed trains, hydrogen-fueled cars, or fusion research
also necessitates accompanying research on infrastructure, distribution and delivery systems.
Taken together, these two dimensions of research objectives, or intended outcomes, can be cross-classified to generate a typology of research obj ectives (see Table 1). In this manner, the choice of the relative emphasis on the radicalness of the scientific discovery or technological advance and focus scope generates four distinctive kinds of strategic choices. In general, the research of Shenhar (1993; 2001) on engineering proj ects is suggestive of how these two dimensions of research objectives can be operationalized in scientific research, where both the idea of scientific or technological uncertainty and systemic scope are in effect. In science and technology, it is the combination of objectives such as superconducting at room temperatures that reflects a radical advance.
Table 1. A typology of research objectives
Research tasks can also be characterized by two dimensions. The first, and most obvious, dimension is the relative size of the research project, measured by the number of researchers, the number of different instruments, the number of technicians that are involved, and the number of teams and organizations involved. Furthermore, as is obvious, as the costs increase--the movement from 10 million to 100 million to one billon—there is a need for significant changes in the organization and management ofthe research. These changes in organization roughly parallel the movement from research project to research program or research organization, such as a national laboratory, to mission agency or inter-organizational network, as in a research consortium. At the extreme is the cost of a space shuttle, with a price tag of over 20 billion dollars expended over ten to twenty years (Geles, 1999).
Significant work can be accomplished in small projects, such as the theoretical advances involving DNA (the double helix structure) and RNA (the messenger), which were radical advances in scientific knowledge, involved relatively small and complex research teams working within a circumscribed knowledge community (see Judson, 1979). In contrast, the mapping of the human genome necessitated a large research program that required the coordination of multiple research teams and inter-organizational relationships since there were over 3,000 genes. In systemic research, it is the large number of variables or properties of the system that require large scale programs to test and develop theories. In large organizations, programs of research are frequently housed in separate divisions because they focus on a specific area of research.
This dimension of size is similar to the fundamental distinction in the Competing Values Framework between flexibility and controlled structures, with the idea that larger projects are those that are more likely to be controlled. But the dimension of size also encompasses a number of consequences that create tensions about coordination and control mechanisms that inevitably impact on project autonomy that are discussed at greater length below. Some would classify the cost
of a project as an indicator of radicalness of the innovation (McDermott & O’Connor, 2002), but we would argue this should be kept quite separate from measures of revolutionary breakthroughs in science or in technology because the cost of these projects is variable. Beyond this, the issue of the cost leads naturally into the more critical question of the size of the research project.
The other dimension of research projects is the complexity or diversity of research on a continuum from specialized to very diverse, as represented by the variety of scientific and engineering disciplines involved. This dimension is equally well established in the literatures on organizational innovation (Hage, 1999) and contingency theory more generally (Lawrence & Lorsch, 1967), as well as in science more generally (Mote, 2005). In our research at Sandia National Laboratories, we have found that many research projects had only six to ten people but in some cases, this represented six or more departments; in other instances, it reflected only one or two departments (Jordan, Hage, Mote, & Hepler, 2005). Thus, even in relatively small research projects, there can be a considerable variation in the degree of diversity in the composition of the researchers and technicians. Above, we have stressed the importance of the technicians and the equipment as one aspect of the size of the project. These factors also represent an element of complexity that is frequently ignored. For example, one of the more interesting aspects of complex research projects is the variety of research equipment that is utilized. In Latour and Woolgar’s (1979) well-known study of the research laboratory, they found ten different instruments used for the purposes of measurement.
Diversity or complexity in the number of researchers and technicians can change across time. Furthermore, as research findings develop, one begins to recognize the need for still other kinds of equipment or of other kinds of expertise, that is, knowledge areas. The changes in the knowledge composition ofthe research project or fluidity over time are yet another measure of complexity. Again, as our research on Sandia National Laboratories research projects and programs demonstrates, typically each year, new scientific and engineering specialties were added and in some cases, others were dropped. In other words, complexity has not only a static dimension but a temporal one as well.
This dimension of structure typically measures whether expertise exists within an organization or within external organizations. Since complexity is essentially a measure of the knowledge pool of the research effort, it is to be expected that not all of the necessary skills and attributes needed are to be found in the same research unit or even within the same research organization, even in large organizations such as the national laboratories or mission agencies. The literature has consistently demonstrated that as the complexity ofthe research effort increases, there is frequently the need for expertise outside the research organization (Alter & Hage, 1993). In this manner, the search for additional expertise or knowledge fosters a greater external focus within the research project or program.
The two dimensions of size and complexity (see Table 2) yield four distinctive types of research projects: (1) small complex research projects; (2) large complex research programs or research organizations; (3) large specialized research programs or research organizations; and (4) small specialized research proj ects. Earlier, we emphasized the importance of adding the external, inter-organizational dimension to the variations on the organic model. Small complex research projects are usually connected to knowledge or practice communities, but typically in a more informal manner, and perhaps one inter-organizational relationship. In contrast, large complex research programs are more likely to be connected to set of inter-organizational relationships and maybe even a consortium, as in global alliances (Gomes-Casseres, 1996) or research consortia such as SEMATECH (Browning, 1995). Again, these are general dimensions with a considerable range of variability.
Degree of Complexity or Diversity
Size of Research Effort
Specialized research projects
Complex research projects; knowledge communities
Specialized research programs or research organizations
Complex research program or research organization; inter-organizational relationships
Although we have employed the word “choice”, there still remains a question of the latitude that managers have in selecting particular research tasks. For instance, the choices may be dictated to them by the environment, whether because of the agendas of funding agencies, control exerted by the state, or certain socialization practices that create a distinctive world view (DiMaggio & Powell, 1983). An example is that the culture surrounding the peer panel review process in general creates a bias toward what is often termed “normal science,” that is, toward incremental advances in knowledge (Braun, 1998).
Crises can also affect the choice of research objectives and tasks. For example, the cyclical nature of environmental concerns has been manifested in rising and falling pressures on public research laboratories and private companies and shifting choices of research objectives and tasks from incremental ones, such as minor improvements in gas mileage, to more radical ones, such as the hydrogen fuel car. As this illustrates, not all pressures are necessarily directed towards radical advances in knowledge. For instance, the pressing need for immediate improvements in national security after the events of September 11, 2001 in the United States might necessitate the need to place greater emphasis on incremental advances, that is, to quickly transfer current technology into security applications (Trajtenberg, 2006).
But the most interesting constraints on choice emerge from the nature of the research problem. Natural systems that are at the extremes of scale, from the very small (subatomic) to the very large (outer space), require quite expensive equipment, numerous technical personnel and many researchers to study them. It is the case that the research problem might be capable of compartmentaliza- tion so that it can be pursued as smaller research projects. Obviously, when this is possible, the pursuit of a smaller research project presents itself as a viable research choice. But if one wants to study the system in its entirety, then either a large research organization, mission agency or a research consortium becomes necessary.
The influence ofthe external world is not the only pressure that affects the choice of the research objectives and tasks. Other, perhaps more critical, pressures are the different kinds of management dilemmas that accompany each choice. These need to be discussed in some detail because they reflect important management challenges and provide examples ofhow management style could make a difference.
Two very common terms in the management literature, especially in discussions of innovation, are the terms “risk” and “uncertainty”. Both of these apply to the choice ofthe research objectives and reflect dilemmas, particularly in the notion of how much risk one should absorb and how much uncertainty. While the terms are frequently
used interchangeably in the literature, we would propose making the following distinctions: risk is a measure of the degree of radicalness in the research objective, while uncertainty is a measure of the scope of the focus on possible outcomes. A review of the literature on the management of innovation indicates that when one shifts to the language of uncertainty, the issue becomes simply a matter of counting the number of uncertainties. In other words, there are not only technological uncertainties but market uncertainties and others, although the role of the market in many areas of basic and applied research are not readily clear (Clarke, 2002). And while it follows that a large number of uncertainties, almost by definition, translates to high risk, the converse is not necessarily true. Indeed, there might only be a few technological uncertainties but still the choice to pursue a radical outcome would entail high risk. For example, Shenhar (1993) and Hobday (1998) suggest that technological uncertainty is related to the degree in which new devices, knowledge or techniques are embodied in a product. But it could be the case that one seeks a radical research outcome, yet still utilize existing technology to do so.
In short, we propose the idea of using risk and uncertainty in somewhat different ways, while still recognizing that the combination of many uncertainties with high radicalness presents the most difficult management tasks. In this manner, we are suggesting that each dilemma flows from a different set of issues than that found in new product development and industrial organization. Attempting to make a significant advance in science or technology development obviously carries a large risk of failure. Uncertainty, however, flows more from the scope of the focus that one is desiring to achieve because as the implicit number of unknowns increases, it becomes more and more uncertain as to which set is the most critical. Stated another way, when the focus is on understanding an entire system, either large or small, especially at multiple levels, there are a large number of potential unknowns or avenues of research.
One might assume that the simple solution to reducing risk is by pursuing a greater number of smaller, incremental research projects. In one sense, this is a familiar dilemma faced by managers of research in risk-averse, budget-conscious organizational environments. But there are times when progress can only be made by taking a radical approach, and “failures” can still result in a scientific advance in the sense that learning took place. In this regard, one is reminded of Thomas Edison’s remark the he did not fail, he simply found ways that did not work. Similarly, one might assume that the simple solution for reducing uncertainty is to pursue a small-scale project that is highly focused on a smaller range of unknown factors, but this can result in losing what many might call the “big picture”, a problem in any systemic science or multiple component technology.
Managerial dilemmas are not restricted simply to the choice of research objectives but also exist in the selection ofresearch tasks. Research projects of a large size require a significant investment in management control and coordination. Such projects also typically need a substantial support staff and support systems for required services, such as accounting, human resources, and library resources. Further, larger teams also need effective leaders who can allocate resources and maintain communication and focus among team members. Finally, management also must define and communicate clear goals and strategies in order to align large groups with strategies. Indeed, the success of a large research project often depends on management correctly positioning the research to fulfill a need or fill a niche. Overall, a unifying system-wide scope makes it possible to set specific goals and track progress. The dilemma is that in increasing size and, correspondingly, internal coordination and control, there is less and less research autonomy and more bureaucracy.
Clearly, the distinction between autonomy and internal control is one ofthe basic structural dilemmas. Therefore, given the tight connection between size and managerial control, we have included the degree of coordination of the research project or program on the vertical dimension of the typology of structure. The desired independence and autonomy of academics are well known, as is the fact that researchers are motivated as much by the recognition oftheir work and the intrinsic pleasure of doing their research as by extrinsic rewards. The tension between autonomy and control also manifests itself in another way, which is the need to search for expertise as complexity increases. In so far as one locates this expertise outside the organization, another structural dilemma emerges because it conflicts with organizational autonomy given that some external control comes with the external expertise. In this regard, coordination difficulties across organizational boundaries are a dominant theme in the inter-organizational literature.
We note that there is a certain irony in this dilemma. Those projects that are already more complex because of their revolutionary strategy are precisely the ones that are most likely to recognize the need for other pools of knowledge. This flows from their aspirations and thus makes the strategic choice so critical, as Zammuto and O’Connor (1992) argued in explaining the adoption of the radical process technology of flexible manufacturing. Meeus and Faber (2006) also observe this is true on the inter-organizational side of the structure as well.
In summary, we have identified a number of dilemmas attached to both the choice of research outcome and the choice of research task within the constraints of how the context—economic, political and scientific—constrain the choices that are made. And as we combine the choice of outcome with the choice of task, these dilemmas multiply, a subject we turn to in the next section.
A central idea in contingency theory is that structure must follow from strategy, thus the two basic dimensions of research objectives discussed above must dictate how the research tasks should be structured. The creation of general variables that is accomplished with the typologies describing both the research objectives and the research tasks allows one to develop several hypotheses about the match between task and objective or intended outcome, the equivalent at the project or program level to the match between structure and strategy. The heart of our perspective is connecting the typology of research objectives with the typology of research tasks, which results in four kinds of research profiles. And since each choice of outcome and of task also has a set of related dilemmas, it is also the case that there are four general areas of managerial problems.
As we stated above, the matching of research task with research outcome is the heart ofthe theory of management styles. The construction of the new perspective is based on a series of hypotheses that emphasize the range of each of the dimensions used to identify research objectives and research tasks. Of these hypotheses, we present two, both of which draw heavily on contingency theory.
1. The greater the emphasis on the radicalness
ofthe outcome in science or technology, the greater the need to emphasize the complexity or diversity of expertise in the research proj - ect, program or inter-organization network.
The more that managers choose to pursue a radical breakthrough in science, then the research project itself must become complex, particularly in terms of different types of knowledge. As discussed before, this flows from the need to synthesize different perspectives as a way of achieving the radical breakthrough. In this vein, Shenhar’s (2001) study of engineering projects provides at least some evidence for the fit between the choice of a revolutionary strategy and the need for expertise; projects coded as most uncertain had the highest proportion of academic degrees.
Research findings on the positive relationship between complexity and industrial innovation defined as new products or services have been accumulated over a forty-year time period. Damanpour (1991) summarized the first thirty years of work in a meta-analysis that explored the relative importance of complexity vis-a-vis other variables while controlling for a number of alternative possibilities. Hage (1999) updated this study and noticed the absence of research on complexity and radical innovation in science. For instance, in discussions about the importance of the discovery of the double helix structure of DNA, it is often overlooked that Watson had been trained as a biologist and Crick as a physicist (Watson, 2001). In a study of the Institut Pasteur, the relationship between complexity and radical innovation holds quite strongly at the level of project (Hage & Mote, 2008).
2. The greater the emphasis on a broad scope of focus, the greater the need to increase the size of the research project, program or inter-organizational network.
As the scope of research broadens, the specific expertise that is needed may not reside within the research organization and must be sought outside the organization, either informally, as in knowledge or practice communities, or formally, as in inter-organizational alliances or consortia. The search for expertise suggests that the research unit increases in size as a consequence, moving along the continuum from project to program to inter-organizational network. The movement along this continuum is driven by the notion that scope is a measure of the number of variables involved in the research and the choice to study multiple levels, components of systems or entire systems. Many ofthe national public laboratories in the U. S. were created to work on difficult and intractable problems. For example, the origins of both LosAlamos and Sandia National Laboratories reside in the Manhattan Project and the drive to construct the first atomic weapon. In response, the Europeans created CERN in 1954 to reverse the brain drain of physicists to the United States and to restore and rebuild European research in high-energy physics.
In general, these types of large national laboratories and mission agencies are quite different from the smaller academic research projects funded by the National Science Foundation (NSF) or National Institutes of Health (NIH). Furthermore, within these national laboratories and mission agencies, additional distinctions can be made between problems that can be approached with small projects and those that require a more extensive program of research. Ifthe project focuses on a system, then almost by definition, one needs a large-scale research program to handle all the aspects of the system, if not a mission agency or national laboratory. Oceanographic ships, space shuttles, and radio astronomy observatories are all examples of quite expensive equipment needed to study particular natural systems, which in turn means quite large support staff and a large number of technicians for their operation.
Another aspect of scientific research is what might be called the difficulty of the research problem, that is, studying phenomena under extreme conditions. For example, consider the case of conducting research on flora and fauna at the bottom of the ocean or the difficulty of studying black holes at the other end of the universe. Again, the presence of extreme conditions often necessitates a large scale program of research.
With these two hypotheses, we can delineate four distinctive kinds of Research Profiles, each with their own distinctive management challenges and dilemmas. We now turn to a discussion of these four Research Profiles.
Combining the two dimensions generates four ideal-type Research Profiles, each of which can be used to define a particular management style. In addition, because we have suggested that there were a series of dilemmas associated with the choices of the degree of radicalness and the degree of scope and similarly with the degree of complexity and the size of the project, we want to explore ways in which these dilemmas can be managed.
Table 3. Research profiles: Strategic and structural dilemmas and associated management challenges
In Table 3, we have combined the dilemmas associated with the strategic structural choices on the assumption that structure follows strategy as we have hypothesized above. Within each quadrant are potential solutions to the dilemmas associated with each Research Profile. It is critical that one interpret this figure as suggesting dilemmas in all four quadrants. While it is easy to argue that high risk or many uncertainties pose managerial problems that necessitate solutions, it is much harder to recognize that too little risk or too few uncertainties also pose managerial problems that require solutions. In particular, the combination of little risk and few uncertainties, which is part of the Small Incremental Research Profile, raises issues about the loss of competitiveness of the research organization in which the project of this nature are located, as well as lack of understanding even by research sponsors, of the importance of some of this work.
Two important general qualifications need to be made about the four Research Profiles. First, the dimensions on which the Research Profiles are based have considerable range. In other words, one must remember that the underlying dimensions are the degree of radicalness, the broadness of the focus, the degree of complexity, and the size of the project, program, or inter-organizational network. Of course, this means that the specific managerial problem discussed occurs in varying quantities accordingly. Second, we recognize that each of these dilemmas exists to a certain degree in each quadrant. For example, if we focus on the Small Radical Research Profile, the two primary managerial problems are to encourage risk-taking and integration of diverse perspectives. But these same issues exist in the Large Radical Research Profile, only now they are overshadowed by the larger number of uncertainties and the greater amount of coordination needed to conduct research. In contrast, in the Small Incremental Research Profile, the motivational problems are more centered on the issue of keeping abreast of the discipline or the competition. As a consequence, a very different set of managerial solutions are needed because of the greater research autonomy. In the Large Incremental Research Profile, the same motivational problems exist but now they emerge at the level of research teams within the program and thus necessitate a different set ofmanagerial solutions.
In this respect, we associate dilemmas and Research Profiles where the problems tend to be accentuated.
To help set the stage for future research on this perspective, we identify one managerial problem associated with the choice ofthe research outcome and one problem associated with the choice of the research task. Many of these examples arose from the discussions in the focus groups that were conducted in the development of the research environment survey instrument discussed earlier (Jordan & Streit, 2003; Jordan, Streit, & Binkley, 2003; Jordan, Streit, & Matiasek, 2003)
It goes without saying that most, ifnot all, researchers would like to have a radical breakthrough in their scientific or technological research, and many often tend to think that their research is geared toward that end. However, as we have suggested, it is important to view this along a dimension with considerable range. In our work at Sandia National Laboratories, we identified several examples of small projects (under 10 million funding over five years) with complex research teams that achieved radical breakthroughs, such as the project that developed a particulate trap for diesel engines that reduced particulates by 400 percent as well a project focused on the development of semi - autonomous, modular robotic control technology that greatly decreases the necessity of human control. Within NOAA, we identified a project focused on new compression methods for satellite data which achieved compression rates of almost a factor of 3 over a current compression standard (Mote, Jordan, & Hage, 2007).
Two managerial problems associated with this Research Profile are encouraging researchers to take intellectual risks and to integrate them as a team. In other words, the motivation issue is how to encourage a team of researchers to think and act boldly. The managerial solution associated with the choice of pursuing radical objectives is to create enough time to think and to explore, encourage the willingness to take risks and also have an atmosphere of challenge. A potential solution is to provide sufficient flexibility, with the resources and freedom to pursue new ideas. This suggests that it is necessary to give researchers in this profile enough time while resisting the pressure of deadlines that may be coming from those higher in the hierarchy.
As we have discussed, increasing the radicalness of the outcome requires that one increase the complexity of the research team, adding multiple disciplines, specialties within them, and technicians and their specialized equipment. But here we face a critical managerial problem associated with this Research Profile: How does one integrate across the various disciplines and equipment so that there is access to tacit knowledge? Furthermore, as Nooteboom (1999, 2000) has observed, although one creates radical innovation by increasing cognitive distance, the tendency is for people to communicate less as the cognitive distance increases. Clearly, the managerial solution is to develop a number of mechanisms to encourage the sharing of tacit knowledge (Judson, 1979), that is to encourage integration across diverse perspectives. Furthermore within the context of science and smaller research projects that address only a component of a larger system or area, we would suggest that this requires more than the classical mechanisms suggested by Lawrence and Lorsch (1967), because this involves ensuring the cross-fertilization of ideas and managing the external collaborations with various knowledge communities.
As one moves from a project that costs less than ten million dollars over a five year period to a program that costs more ten million dollars, and even upwards of a billion as in the example of NASA’s space program, a new set of problems emerge that necessitate a different managerial style. Indeed, the increase in scale magnifies the problems described above, but alters them in interesting ways.
One of the earliest examples of a large-scale program of research is the famous Manhattan Project that produced the first atomic weapon. Since that time, large-scale research programs have been initiated to pursue a number of high-profile topics, with recent examples being the Human Genome Proj ect and the National Nanotechnology Initiative (NNI). On a somewhat smaller scale, Sandia National Laboratories has initiated a handful of relatively large projects under the category of Grand Challenges, which are designed to be first to solve a major technical challenge. One recent example has been a research project on the feasibility of wide-scale, mainstream use of light emitting diodes (LED), which requires advances in both semiconductor design and manufacturing.
In this Research Profile, not only does one want to encourage taking risks and to integrate diverse perspectives, but one also has to be concerned about handling the number of uncertainties and the multiple research teams, including teams that cross organizational boundaries. What are the solutions? To maintain an emphasis on a radical breakthrough in a larger scale program of research, one must put forth a clear and consistent vision that provides much of the motivation for the research teams. Part of this vision is the investing in future opportunities that may provide the breakthrough in either the knowledge or technology needed to advance. The problem of handling the many uncertainties is dealt with by developing a clear set of project goals and strategies so that the many different re search teams and organizations that are cooperating together in this effort can understand their responsibilities. The clarity ofthe vision, such as Project Apollo’s explicit goal to land a man on the moon and return him safely to Earth, is critical for the success of these large scale and complex efforts. Initially, the process of determining the right research strategy and achieving a consensus on that strategy can be both time and resource intensive. Hence, continued funding over long time periods is essential. Therefore, projects and programs in this profile often endure well beyond the shorter time horizon of smaller projects, as Leifer et al (2000) have observed.
But an increase in scope of focus also creates new difficulties with managing the complexity of multiple research teams in networks of organizations, all of which have to be integrated. And in large programs and research organizations, it is important to maintain coordination and control over the research efforts. As research autonomy is reduced in this manner, so too are the opportunities to create a radical breakthrough. One of the ways in which this reduction in autonomy impacts on the creativity of the research is the inherent conflict between creativity and maintaining progress and schedules—an important issue in large complex programs. While measuring and tracking progress is an essential part of the solution to this managerial problem, it must be done in a way that does not force researchers to focus on shorter term, incremental, measurable products, benchmarks, or milestones, or take too much time away from research.
The managerial problems shift dramatically in small scale research projects designed for incremental improvements in science and technology, which, of course, are the most common ones in academia, government and in industry. Interesting examples from our work with NOAA include the ongoing, incremental improvements in the calibration of satellite data, as well as the identification of a new type of hurricane that maintains its intensity for a longer time period. A project from our work at Sandia National Laboratories also helps to illustrate this profile. This research proj ect seeks to understand the nature of soot formation from unsteady, “dirty” flames; previous research primarily focused on the properties of steady flames generated by simple, pure hydrocarbon fuels. As these examples demonstrate, as well as many of the research projects funded by the NIH or the NSF, the research is not necessarily aimed at radical advances but at incrementally pushing the knowledge boundaries.
Since the projects in this profile are relatively smaller, the effectiveness depends much more upon the individual researcher and, at most, a small team. Therefore, one of the managerial problems in this profile is to ensure continued professional development, particularly with regard to maintaining and improving skills. Left to their own devices, researchers often tend to focus on the same set of problems across much of their professional life. In this regard, a kind of human capital decay sets in because not enough new issues are being considered to really continue to master a particular specialty. Professional development, such as continued education, seminars, and conferences, becomes a key mechanism for increasing their human capital. Individual growth is also encouraged when there is an atmosphere of cooperation rather than competition among diverse research projects in an organization. Under these circumstances, individuals are more willing to interact and talk with others about their research, particularly scientific challenges they might be facing and need help in overcoming.
Unlike the Small Radical profile, the problem here is potentially too low a level of aspiration rather than one that is too high. Again, the managerial solution to this problem is in various ways to indicate to the individual that they are valued and their work is valued even if it does not need to take intellectual risks or a broad focus. This can be accomplished by providing respect for the individual and valuing their ideas and opinions which can be achieved by allowing autonomy in decision-making, as well as more latitude in choosing areas of research. While this may appear to be quite simple, it is, in fact, often more difficult because typically the rewards are given to those scientists and research teams that make the big breakthroughs. Therefore, part of the solution includes providing greater recognition to the contributions of researchers who have not experienced similar achievements.
The combination of less radical and broad scope of focus research is frequently found in the various national laboratories and mission agencies funded by federal governments. This profile is characterized by large-scale and functionally specialized teams. For example, NOAA maintains separate research programs on both the atmosphere and the oceans, each of which encompasses a number of integrated research projects. Further, the development and implementation of a new weather satellite typically entails a substantial program that includes teams from a number of mission agencies, such as NOAA and NASA, the military and private contractors. This type of research is focused largely on incremental advances and poses different managerial challenges than research efforts focused on radical advances, such as hydrogen fuel-based cars or unmanned aerial vehicles.
Paradoxically, the large scale of research in this profile poses two managerial problems that are also observed in small, specialized projects discussed above, that is, professional development and individual respect. Given the larger scale of research, however, the existence of these two problems in this profile reflects the need to overcome the impersonality of layers of bureaucracy and, hence, suggests different management strategies. A major difference in which the handling of motivation in this profile differs from that above is that rather than value the individual or the few individuals in the team, here the problem is proper recognition of each team. Unlike the Large Radical Profile, which also emphasizes teams and teamwork, this profile emphasizes the pursuit of incremental advances, primarily on components of an overall system and use of functionally specialized teams. It goes without saying that coordination among these teams relies not only on management that is well informed about the research and rewarding and recognizing individual merit, but also on good internal allocation of funding. Cultivating well-informed, technically adept management within the research project is always difficult, for it require s skill in administration and management of people as well as knowledge and skills in the research area. This is particularly important given that the research expertise may lie in separate departments not under the immediate control of the program manager. Indeed, it is this kind of profile that led to the creation of matrix authority structures and, as is well known, this particular structure has not always functioned well because employees find it difficult to serve two masters, discipline (or department) and project (Tushman & O’Reilly, 1999; Tushman & O’Reilly, 1996). For all of these reasons, the need for clear cut measures of technical progress for each team is essential so that progress can be measured carefully and appropriately.
As indicated in the Large Radical profile, one of the major issues for large projects is the lack of research autonomy. Shenhar (2001) provides a vivid description of how managers in this profile attempt to bureaucratize the coordination of the disparate parts of the research program or teams. But, in fact, this solution is what creates many of the problems within this research profile. In addition to the impersonal effects, bureaucratic control also has negative impacts on technical management in requiring time for administrative tasks that can detract from time spent on research. Indeed, in this situation, team leadership becomes very desirable as a solution (Hage & Mote, 2008).
In summary, our argument is that the management of different types of research, including basic, applied, product development, and quality control, necessitates an adaption of existing theories of innovation and R&D management in the current literature. The Research Profiles should help with portfolio level decisions because it proposes general categories of research objectives and tasks that can be applied across various arenas of research from basic to manufacturing research to inform strategy decisions and to judge progress and effectiveness. Thus, our perspective about the diversity of research projects, contained in the Research Profiles, has an incredible spread of potential impact. Of course, a test of this assertion is the number of insights that one obtains from the perspective and below we identify and discuss three such insights.
One of the insights is the revision of the organizational learning perspective. In this respect, we would argue that the Research Profiles represent four different kinds of learning. The strategic choice of pursuing a radical breakthrough is probably going to lead to more learning than the pursuit of an incremental strategy. This corresponds to the notion of competency destroying innovation and competency enhancing innovation, which largely looks at the impact on other organizations. The Research Profiles perspective suggests that the organization that has the radical breakthrough is learning a great deal and building its capabilities in the process. As has been indicated in our discussion of each of the four profiles, whether the organizational structure and management style match the research strategy will determine if the desired scientific and organizational learning actually takes place. In each instance, different obstacles to learning present themselves. A second insight is the revision of organizational innovation theory, particularly the emphasis on the organic model, and the integration with the inter-organizational literature. The diversity of research projects shifts the attention in organizational innovation to the study of where the research is largely accomplished, while recognizing that there is not just the organic model but four profiles of research activity.
The third insight is the discussion of inherent dilemmas and problems. Rather than perceive low risk and low uncertainty as presenting no managerial dilemmas, we have suggested that they do. And unlike traditional contingency arguments that assert that the correct fit ameliorates any managerial problems, we are suggesting quite the opposite. Even with a good fit, there are still managerial problems associated with each of the four types of Research Profiles. These problems flow from the dilemmas that have been delineated. We have suggested what some of the solutions to these dilemmas are and then indicated that these define specific management styles. This is clearly not the usual perspective in either contingency theory, or management theory in general, but we would argue it is critically important given the central importance of research for both economic competitiveness and national security.