came to an end (Lippitt, 1949). This exchange led to a promising new pattern in which researchers reported on their observations and the participants listened, reflected, and shared their own interpretations of their own behavior.
Attendance at the evening sessions soared in subsequent days, with almost all participants attending, and this led to the researchers’ conclusion:
Group members, if they were confronted more or less objectively with data concerning their own behavior and its effects, and if they came to participate nondefensively in thinking about these data, might achieve highly meaningful learnings about themselves, about the responses of others to them, and about group behavior and group development in general. (Benne, 1964, p. 83)
Lewin seemed to know instinctively that this was a potentially powerful finding, remarking that “we may be getting hold of a principle here that may have rather wide application in our work with groups” (quoted in Lippitt, 1949, p. 116). The training group (or T-group) was born.
The following year, 1947, the first T-group session took place at the National Training Laboratory in Bethel, Maine. T-group sessions were designed to last 3 weeks and comprised approximately 10 to 15 participants and one or two trainers. In open and honest sessions in which authenticity and forthright communication were prized, group members spent time analyzing their own and others’ contributions, as well as the group’s processes. Regardless of whatever process the groups followed, the common objective of each T-group was to create interpersonal change by allowing individuals to learn about their own and others’ behavior, so that this education could be translated into more effective behavior when the participants returned home. As the word spread about the effectiveness of the T-group laboratory method, managers and leaders began to attend to learn how to increase their effectiveness in their own organizations. Attendance was aided by a BusinessWeek article in 1955 that promoted “unlock[ing] more of the potential” of employees and teams (“What Makes a Small Group Tick,” 1955, p. 40). By the mid-1960s, more than 20,000 businesspeople had attended the workshop (which had been reduced to a 2-week session), in what may be considered one of the earliest fads in the field of management (Kleiner, 1996).
The research that Lewin began has had a significant influence on OD and leadership and management research. His research on leadership styles (such as autocratic, democratic, and laissez-faire) profoundly shaped academic and practitioner thinking about groups and their leaders. His influence on his students Benne, Bradford, and Lippitt in creating the National Training Laboratory has left a legacy that lives on today as NTL continues to offer sessions in interpersonal relationships, group dynamics, and leadership development. The fields of small-group research and leadership development owe a great deal to Lewin’s pioneering work in these areas. Though the T-group no longer represents mainstream OD practice, we see the roots of this method today in organization development in team-building interventions (a topic addressed in detail in Chapter 11). Lewin’s research also influenced another tradition in the history of organization development—action research and survey feedback.
Action Research, Survey Feedback, and Sociotechnical Systems
Lewin’s objective at MIT was to develop research findings and translate them into practical, actionable knowledge that could be used by practitioners to improve groups and solve their problems. Lewin called this model action research to capture the idea that the research projects at their core always had both pragmatic and theoretical components and that rigorous scientific methods could be used to gather data about groups and to intervene in their processes (Cunningham, 1993). Two important developments during this time were a survey feedback process and the field of sociotechnical systems.
Survey Feedback
While Lewin and his colleagues were developing the T-group methodology, an effort was taking place at the University of Michigan, where a Survey Research Center was founded in 1946 under the direction of Rensis Likert. In his Ph.D. dissertation at Columbia in 1932, Likert had developed a 5-point scale for measuring attitudes (a scale known today as the Likert scale). One of the first “clients” brought to Michigan was that of the Office of Naval Research, which was “focused on the underlying principles of organizing and managing human activity and on researching techniques to increase productivity and job satisfaction” (Frantilla, 1998, p. 21). The contract with the Office of Naval Research provided needed and important funding for Likert’s work on management practices in particular, culminating in a 1961 book, New Patterns of Management, which reported the results of his funded research. (These findings are discussed in the next section.)
The Survey Research Center’s goal was to create a hub for social science research, specifically with survey research expertise. Sensing an opportunity to improve their organizations, derive economic success, and develop a competitive advantage, some organizations proposed survey research projects to the center but were denied because the center aimed to focus on larger projects of significant importance beyond a single organization and to share the results publicly. These two criteria (addressing questions of larger significance and making the results known to other researchers and practitioners) formed the core of the action research process. One such project that met these criteria was a survey feedback project at Detroit Edison.
Members of the Survey Research Center conducted a 2-year study at Detroit Edison from 1948 to 1950. The survey of 8,000 employees and managers was administered to understand perceptions, opinions, and attitudes about a variety of aspects of the company, such as career progression and opportunities for advancement, opinions about managers and colleagues, and the work content and work environment itself. The survey also asked supervisors specifically about their opinions about managing at the company, and invited senior leaders and executives to offer additional perceptions from the perspective of top management. The researchers sought to understand not only how employees at Detroit Edison felt about the organization but also how the results of this project could be used to understand, instigate, and lead change in other organizations. There were four objectives of the research project:
1 To develop through first-hand experience an understanding of the problems of producing change
2 To improve relationships
3 To identify factors that affected the extent of the change
4 To develop working hypotheses for later, more directed research. (Mann, 1957, p. 158)
Following the initial data collection, feedback was given to leaders and organizational members about the survey results. Mann (1957) described the process of sharing this feedback as an “interlocking chain of conferences” (p. 158) in which initially the results were shared with the top management, assisted by a member of the research team. At this meeting, participants discussed the results, possible actions, and how the results would be shared with the next level of the organization. Next, each of those participants led a feedback discussion with his or her team about the research results, also conducting action planning and discussing how the results would be shared with the next level. This pattern continued throughout the organization. At each level, the data relevant to that specific group were discussed. Mann noted that the leaders in each case had the responsibility of presenting the data, prioritizing tasks, taking action, and reporting to their supervisors when they had reached an impasse and needed additional assistance to produce change. The researchers observed that this series of feedback meetings had a very positive influence on initiating and leading change in the organization, but they had been unable to substantiate this observation with data.
In 1950 that changed with a second study conducted in eight accounting departments at Detroit Edison that had participated in the first survey. In four of the eight departments, after the initial feedback meeting, no action was taken based on the survey results (two intentionally as “control” departments; two due to personnel changes that made it impossible to continue to include them in the experiment). In the four departments that did take action, managers developed action planning programs that differed significantly from one another. Some programs took as long as 33 weeks, while others took 13; some departments met as frequently as 65 times, while others met as few as 9. Some department action programs