3
Teams in Complex Environments

THE CENTRALITY OF TEAMS

Throughout history, teams have been the fundamental unit of organization in the military. Wars are fought and won or lost by people working together in small groups located at every level of the command hierarchy: the infantry squad in the field, the aircraft mechanics in the hanger, the technicians in the engine room of a nuclear submarine, and the Joint Chiefs of Staff in the boardroom. Team effectiveness has always been and continues to be essential to the success of military missions.

Given the centrality of teams, it is not surprising that the behavior of teams has been studied for a long time. Recently, major strides have been made in understanding of team effectiveness. At least seven major reviews of the research on teams were conducted between 1990 and 2000 (see Kozlowski and Ilgen, 2006). In addition, there has been comprehensive and quantitatively focused research on team decision making under stress by Salas and his colleagues (Cannon-Bowers and Salas, 1998) and qualitatively focused work of Hackman and his students across multiple types of teams (Hackman, 1990). Recent reviews (Ilgen, Hollenbeck, Johnson, and Jundt, 2005; Kerr and Tindale, 2004; Kozlowski and Ilgen, 2006; Salas, Stagl, and Burke, 2004) document the continued advances in understanding team behavior.

The interest and research activity is due, in part, to a shift in the way work is organized in civilian as well to military organizations—away from the isomorphic linking of individuals to jobs to that of defining jobs at the team level. This shift was brought about by increasing workplace complex-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 29
3 Teams in Complex Environments THE CENTRALITy OF TEAMS Throughout history, teams have been the fundamental unit of organi- zation in the military. Wars are fought and won or lost by people working together in small groups located at every level of the command hierarchy: the infantry squad in the field, the aircraft mechanics in the hanger, the tech- nicians in the engine room of a nuclear submarine, and the Joint Chiefs of Staff in the boardroom. Team effectiveness has always been and continues to be essential to the success of military missions. Given the centrality of teams, it is not surprising that the behavior of teams has been studied for a long time. Recently, major strides have been made in understanding of team effectiveness. At least seven major reviews of the research on teams were conducted between 1990 and 2000 (see Kozlowski and Ilgen, 2006). In addition, there has been comprehensive and quantitatively focused research on team decision making under stress by Salas and his colleagues (Cannon-Bowers and Salas, 1998) and qualita- tively focused work of Hackman and his students across multiple types of teams (Hackman, 1990). Recent reviews (Ilgen, Hollenbeck, Johnson, and Jundt, 2005; Kerr and Tindale, 2004; Kozlowski and Ilgen, 2006; Salas, Stagl, and Burke, 2004) document the continued advances in understanding team behavior. The interest and research activity is due, in part, to a shift in the way work is organized in civilian as well to military organizations—away from the isomorphic linking of individuals to jobs to that of defining jobs at the team level. This shift was brought about by increasing workplace complex- 2

OCR for page 29
0 HUMAN BEHAVIOR IN MILITARY CONTEXTS ity and its consequences: demands for expertise, flexibility, and adaptability. Furthermore, military teams are increasingly composed of people who differ in culture, ethnicity, gender, age, and other externally apparent individual characteristics. Consider, for example, American infantry soldiers who receive combat training in small units that are relatively homogeneous in age and language and who have been both raised and trained in the United States. Then these soldiers suddenly find themselves in teams of Iraqi po- lice trainees, patrolling the streets of Baghdad with teammates who can- not speak their language and have little or no shared experiences, either culturally or in tactical training. The mix of technical and interpersonal skills, along with those of temperament and adaptability needed to perform in such teams is critical to soldiers’ survival as well as the success of the mission. From a technical rather than a social perspective, teams are increasingly imbedded in technological systems in which the boundaries between human and nonhuman functions are inextricably confounded. Cockpit crews be- lieve that they fly aircraft; designers and engineers believe that the technol- ogy flies aircraft. Both are right and both are wrong. Given the complexity of interacting systems in today’s aircraft, precisely assigning the percentage of responsibility to humans or machines is impossible. DEFINITION Definitions of teams most similar to those in the military typically assume that they are composed of two or more individuals who interact socially; possess one or more common goals; perform tasks valued by the organization in which the team is embedded; possess some interdependen- cies with respect to workflow, goals, and outcomes; have roles in the team that may or may not differ in status and responsibilities; and are embedded in a larger organizational system. Similar characteristics were articulated by Salas, Dickenson, Converse, and Tannenbaum (1992) and appear in much of the research literature on work teams (e.g., Alderfer, 1977; Argote and McGrath, 1993; Hackman, 1992; Hollenbeck et al., 1995; Ilgen, 1999; Kozlowski and Bell, 2003; Kozlowski, Gully, McHugh, Salas, and Cannon- Bowers, 1996; Kozlowski, Gully, Nason, and Smith, 1999; and Kozlowski and Ilgen, 2006). This definition of teams includes important contextual conditions. Specifically, the teams exist to perform tasks, and they are rarely free-standing units but are embedded in larger organizational systems. Implicit in the inclusion of task performance as a defining characteristic of teams is the fact that teams exist to serve some purpose and that the ef- fectiveness of the team is important. The model that has guided research on team effectiveness for the last 30 years is the input-process-output (I-P-O) heuristic model usually attributed to McGrath (1964). Inputs are resources

OCR for page 29
 TEAMS IN COMPLEX ENVIRONMENTS available to the team from its members (e.g., their skills and abilities), tech- nological assets, and other factors in the team’s environment. Processes are the activities of the team that convert the inputs into outcomes (e.g., coor- dinated effort of team members). Hackman (1987) defined the outcomes domain through partitioning into three relatively independent categories: performance, which is measured in task-relevant units; factors that affect team member needs (e.g., status, feelings of accomplishment, income); and team viability, the factors that hold the team together. Most research ques- tions fit within a framework of investigating the effects of some factors that can be seen as inputs on team processes and, in turn, on one or more of the three output criteria of performance. Today the I-P-O paradigm still dominates research on team effective- ness. However, this theoretical structure has been expanded in two impor- tant and critical ways: by time and by levels of analysis. In terms of time, process is, by definition, dynamic. Team process constructs are the result of team member cognitions, emotions, and behaviors that occur in the life of the team. Marks, Mathieu, and Zaccaro (2001) made explicit the dynamic nature of many of the constructs classified as “process” by introducing “emergent states” as a way of differentiating dynamic elements in the link between inputs and outcomes. Kozlowski et al. (1999) raised similar ideas in the framework of developmental cycles in the life of teams. With respect to levels of analysis, the well-accepted notion that people (human systems) are nested within teams and that teams are, in turn, nested within organizational and technical systems (Olson and Olson, 2003) is now not only simply acknowledged: the implications of this nesting are being more clearly articulated both theoretically and methodologically. One immediate result of adopting a levels-of-analysis perspective is more careful scrutiny of the implications of within-team variances. Before researchers adopted a levels-of-analysis perspective, such team-level con- structs as self-efficacy were typically defined as the sum or average across team members; within-team variance was assessed primarily for purposes of evaluating the accuracy of the team score. Now, however, researchers consider more carefully whether or not the construct itself requires conver- gence across members in order to be valid. For example, team self-efficacy is the extent to which the members of a team believe the team is efficacious performing its task. The agreement of the team members is typically assessed only to decide whether the mean represents the team members’ beliefs. Yet, in considering the most extreme responses of team members, a bimodal distribution of some other configu- ration might be of interest in determining team self-efficacy. Theoretically meaningful algorithms for indexing team self-efficacy are getting more at- tention as multiple levels of analysis are taken into account.

OCR for page 29
2 HUMAN BEHAVIOR IN MILITARY CONTEXTS BASIC TEAM PROCESS AND EMERgINg STATES Given that teams are themselves multilevel systems of people and their environments, it should not be surprising that the team literature addresses constructs at multiple levels. Researchers have taken the constructs of cog- nitive, emotional/motivational, and behavioral processes of individuals and raised them to the team level. Two team-level cognitive analogs are mental models and transitive memory. The former refers to team members’ shared mental representation of knowledge or beliefs relevant to key elements of the team’s task environ- ment (Klimoski and Mohammed, 1994). Transitive memory is shared infor- mation about team members’ knowledge, roles, skills, and abilities (Wegner, 1995). Mental models have been shown to have a positive effect on team performance, and they can be influenced by both training (Blickensderfer, Cannon-Bowers, and Salas, 1997) and leader behavior (Marks, Zaccaro, and Mathieu, 2000). Although having shared mental models is important in any context, it is particularly important in military contexts: team members may have differ- ent perspectives but think they share the same ones as their teammates. For example, when there are joint operations with military units from different nations, there may be differences in small unit tactics, and the troops may not realize all the consequences of those differences. When a person trained in one army finds himself or herself in a unit of another army, the person may assume that others in the unit share his or her perspective, which leads to increased chances of errors and misunderstanding, particularly under the stress of combat. Although both mental models and transitive memory are typically limited to team members’ beliefs and perceptions, these domains can be expanded with information technology (e.g., through computer-supported cooperative work). Extending team memory, adding decision-making ca- pacity with intelligent systems, and totally changing the meaning of distance and time through synchronous and asynchronous interaction in space and time in virtual teams are radically changing the nature and results of cogni- tive processes in teams (Olson and Olson, 2003). For example, the concepts of shared memory and shared representations are related to the concept of distributed cognition. This view represents a shift in the study of cognition from being the sole property of an individual to its being a property of groups, material artifacts, and cultures (Hutchins, 1995; Suchman, 1987). This viewpoint is increasingly gaining acceptance in cognitive science and human-computer interaction research. Its importance in the military is evident as robots, unmanned aerial vehicles, and other automated systems with the capacity to learn from their own experience are imbedded in

OCR for page 29
 TEAMS IN COMPLEX ENVIRONMENTS teams whose mission requires both human and nonhuman interaction for success. Distributed cognition has two central points of inquiry, one that em- phasizes the inherently social and collaborative nature of cognition and one that characterizes the mediating effects of technology or other artifacts on cognition. Clearly, this model of cognition can be closely interwoven with the study of teams, especially those in environments characterized by high levels of technology. A number of affective, mood, and emotional constructs that are im- portant for individual effectiveness have also been found to affect team effectiveness, including satisfaction or identification with the team and commitment. Interestingly, the effect of these factors is greater on member outcomes or team viability outcomes than it is on performance criteria of team effectiveness (see Kozlowski and Ilgen, 2006). These factors also affect critical team processes of cooperation, coordination, and collabora- tion, and so are worthy of study in the team context as well as in the areas described elsewhere in this report (see Chapter 5, Chapter 6, and Chapter 7). From the perspectives of both internal team and the setting in which teams operate, multicultural factors are also becoming increasingly im- portant influences on team process. Understanding how culture influences cognitive and emotional processing in ways that affect team functioning is at an early stage; research on this topic will yield results in the long term 10-20 years (see, e.g., Gelfand, in this volume). For example, as attitudes change within a culture toward minority racial, ethnic, or religious groups, they might affect the ability of the military to train and maintain effective team functioning. Such changes in attitudes and in society and within the military itself have been documented by sociologists (see, e.g., Segal, 1989; Segal and Segal, 2004) and historians. When teams share what are commonly called values, it is frequently labeled team climate. There is significant empirical support for the con- sequences of team climate on team effectiveness (Ostroff, Kinicki, and Tamkins, 2003). These value and motivational factors that develop in teams play a role not only in the teams themselves, but also with respect to es- tablishing commitment of the team members to the larger military context. With a voluntary military, staffing is always critical. Interpersonal bonds that are established in teams have the potential of influencing soldiers’ long- term commitment to a career in the military. Although it is likely that the construct of “climate” influences team effectiveness in technologically rich environments, understanding of how it works there is limited. Another set of team processes addresses linkages among team members or between team members and their tasks. Cohesion, conflict, cooperation, and coordination are processes that, by definition, involve interactions

OCR for page 29
 HUMAN BEHAVIOR IN MILITARY CONTEXTS among two or more team members. There has been a great deal of research on each of these, addressing both their effects on team effectiveness and on the factors that cause these processes (for some meta-analytic reviews, see Beal, Cohen, Burke, and McLendon, 2003; DeDrue and Weingart, 2003; and Gully, Devine, and Whitney, 1995). Technology that takes into account the psychology, sociology, and anthropology of teams will be much more effective than technology that requires teams to adapt in ways that conflict with the ways people usually work together. The impact that technology has on a team’s effectiveness can be either devastating or enabling. Behavioral and social science research can help ensure that it is the latter. INFLuENCINg TEAM EFFECTIvENESS Understanding team processes is absolutely essential for building a foundation for understanding team effectiveness. It is also necessary to understand the conditions that allow team processes to operate in positive ways in one or more of the three domains of effectiveness: performance, meeting members’ needs, and improving team viability. These domains can also be characterized by the issues of staffing; training and development; leadership; and the design of the organization. Staffing Key functions of staffing are recruitment, selection, and placement. Well-developed and validated policies and practices for these functions have been developed for individuals, and much of what is known at the individual level holds for teams when the pool of knowledge, skills, and abilities of individuals is expanded to include teamwork-relevant knowl- edge, skills, and abilities. The staffing paradigm as it relates to placement at the team level is different than at the individual level because team members may be added and dropped over time, creating the need for dynamic models. There are some models, such as Schneider’s (1987) attraction, assimilation, and at- trition model, that describe staffing over time, but most other models are mainly analogs of individual models. Team-level staffing also must take into account the fact that critical constructs, such as adaptability, cannot be understood by looking simply at the adaptability of individual members: the adaptability of a team involves individual and job design inputs and emerges over the life of the team in ways that are not captured well at the individual level (Burke, Stagl, Salas, Pierce, and Kendall, 2006). One important staffing issue is that of composition with respect to diversity. There have been two hypotheses on the effects of diversity, one

OCR for page 29
5 TEAMS IN COMPLEX ENVIRONMENTS proposing positive effects and the other proposing negative effects. The first is that diversity among team members expands the resource pool of the team with respect to task behaviors and is helpful for performance. The second is that diversity of culture, values, gender, and other easily identifi- able differences is likely to generate teamwork processes that reduce team effectiveness. An extensive review of the research on diversity (Mannix and Neale, 2006) does not support either hypothesis: diversity does not show a main effect on task performance in either direction. Rather, whether or not the diversity effect of the expanded resource pool is positive depends on the nature of the tasks to be performed. Future work is needed to better map the task and diversity space, keeping in mind the normative demands for creating and maintaining diverse teams. Team Training and Development The team training literature is a rich one, and there is considerable evi- dence that training can have a significant positive impact on team effective- ness. Klein et al. (2006) and Salas, Nichols, and Driskell (2007) looked at team training objectives, evaluating them against specific training objectives as well as what they called team building. The training was an intervention for already existing teams and monitored by the teams themselves. The au- thors found that more than 23 percent of the variance in team effectiveness was predicted by the team training and building programs. Simulated exercises have been frequently used to train teams, and these methods have been found to be very useful, particularly in complex, high- reliability environments in which avoiding errors is extremely important. Another form of training, cross training, requires team members to learn the tasks of all of their team members, to the extent technologically possi- ble, in order to increase their ability to back up others who are overloaded, to monitor others’ behavior, and to detect potential problems for others. Back-up behavior is particularly critical in military settings where sudden spikes in demand for high levels of effort and performance by one or two team members may require others in the team to come to their aid. Also, when team members are incapacitated or reassigned, those who remain in the team must cover for the missing person(s) until the team is able to return to a fully staffed level. Training that improves teamwork has been applied in a number of settings, many of them using simulations and other technologies. Meta- analyses have revealed that simulation-based training can indeed improve team performance. In particular, training for adaptability and coordination using principles first developed for the National Aeronautics and Space Administration as part of cockpit resource management has been adapted

OCR for page 29
6 HUMAN BEHAVIOR IN MILITARY CONTEXTS to other team settings, such as emergency rooms and surgery teams in hos- pitals, with good results. Leadership Discussions of leadership often partition factors that influence effective leadership into traits (stable characteristics of the leaders themselves), states (skills that could be trained), and situational conditions in which some states and traits are better than others. Today it is recognized that all three of these factors interact and change over time: there are relatively stable leadership propensities that do matter, but skills can be developed over time, and they are a function of both traits and experience. Furthermore, some leadership characteristics are more suited to some situations than others. According to Kozlowski and Ilgen (2006, p. 107): “[E]merging meta-ana- lytic findings provide a useful indication of the potential value of leadership in the promotion of team effectiveness.” Team leaders can directly influence the mental models, transactive memories, and perceptions of team climate and, through them, they can affect team members’ behavior and effectiveness. Leaders influence per- formance regulation in teams through goal setting, strategy development, and feedback and by influencing individual and team rewards as well the contingencies between behavior and outcomes. Leaders often have influence on team members’ roles, the technological resources available to the teams, and the structure of tasks and interpersonal relationships. Finally, a recent meta-analytic review of research on leader behavior patterns and styles by Judge and Piccolo (2004) found that both transactional leadership (leader behaviors that influence follower rewards) and transformational leadership (similar to charisma, inspiring people to higher goals) positively influenced team effectiveness. Anecdotal, popular reports are full of stories of military leaders who are believed to have had major impacts on their team’s performance in battle. The consistency between these reports and the literature based on behavioral science research adds credence to the general nature of the effect of leadership on team performance. Moreover, leaders’ influences through transformational leadership behaviors are likely to go beyond the context of the team to that of the commitment of team members to careers of mili- tary service. Organizational Structure and Design Structure is important for team effectiveness both within the team and as part of the larger organizational system. Research on structural effects is often focused at the interpersonal level of coordination, collaboration,

OCR for page 29
 TEAMS IN COMPLEX ENVIRONMENTS influence, lines of authority, communication patterns, and other mechanistic linking mechanisms for tying individuals to tasks and to each other. These factors also have an effect at the level of units throughout an organization. Structure at a higher level is addressed in terms of organizational charts, the language in use (functional and divisional), or models (Petri-nets). All of these structural frameworks have been used at various times with teams and have been shown to alter team behavior. Yet, without denying the importance of them or the overall importance of structure and design, it is clear that most of these constructs are not independent of each other. This lack of independence has been recognized, and it is likely that a more concise conceptual space for structural frameworks will be developed in the near future. Structure can sometimes impede performance. For example, the U.S. Army routinely conducts after action reviews (AARs), in which a training or actual exercise is critiqued, not by an outside investigating body, but by the individuals in the unit. Such reviews are conducted as collegial discus- sions, insofar as possible, and are not structured by rank. Sergeants can criticize lieutenants, and lieutenants can comment on captains’ decisions. These exercises provide extremely useful feedback. The AAR is effective in part because it permits input from people with different perspectives. In combined operations such as are currently being conducted with the Iraqi “friendly” forces, it is essential that everyone participating in the exercise also participate in the AAR. However, the AAR violates both the hierarchical structure of the Army and the even more rigid hierarchical structure of many Middle Eastern and Asian cultures. Since feedback is very important for the improvement of performance, appropriate ways to conduct AARs that will include feedback from non-American personnel need to be developed. Although the word “structure” implies rigidity, it does change over time. Social changes, such as a sudden—or even slow—breakdown of respect for authority or credentials, can cause dramatic changes in an or- ganization in any society. Since such respect is crucial to the function of military teams, research by social historians and sociologists is essential to understanding the consequences of social changes. The demand for teamwork among geographically dispersed team mem- bers has led to the development of collaborative tools—“teamware” and other software systems—designed to facilitate work in virtual space. A number of commercial programs exist for working with these systems. Those programs have been informed by very promising research on virtual teams, but at this time more work is needed to understand better how the characteristics of these systems alter team effectiveness (Kiesler and Cummings, 2002). Teams are ubiquitous in today’s organizations, including the military,

OCR for page 29
 HUMAN BEHAVIOR IN MILITARY CONTEXTS and their effectiveness is absolutely critical to the performance and viability of the organizations in which they are embedded, as well as to the well-be- ing of the people who are part of them. A large research literature focuses on understanding team behavior and functioning; it informs the design of organizations, leaders’ behaviors in them, and policies and practices related to staffing teams. The complexities of teams as dynamic systems nested in organizations and changing over time present major challenges for future research that are just beginning to be addressed both theoretically and methodologically. These complexities are the focus of long-term research (10-20 years). It is important that the momentum of research on teams continue because it is likely that work will increasingly be organized around teams rather than individuals. The challenges of diverse memberships, rapidly changing demands, and dispersion over space and time will also increase. A two-pronged, interdependent approach is needed: work on better under- standing the processes that influence team effectiveness and work on the factors that influence the organizational policies, practices, structures, and leader behaviors that influence key team processes to improve their effec- tiveness. The two approaches must inform each other and advance in an iterative fashion, integrating their insights and results, and studying them in organizations of all types. They pose questions that must be addressed in both short- and long-term research. As military team members work ever more closely together and yet are dispersed in space and time, their knowledge and skills must be as thor- ough and comprehensive as possible, and they must be finely tuned to the requirements of the mission. These objectives in turn require the best train- ing available and lots of practice: both training and practice are expensive commodities and are currently strongly oriented to technology. These fac- tors will influence the kinds of behavioral and social science research that promise the best payoff, in both the short and longer term.