Action Research - Action Research in Organization Development

Action Research in Organization Development

Wendell L. French and Cecil Bell define organization development (OD) at one point as "organization improvement through action research". If one idea can be said to summarize OD's underlying philosophy, it would be action research as it was conceptualized by Kurt Lewin and later elaborated and expanded on by other behavioral scientists. Concerned with social change and, more particularly, with effective, permanent social change, Lewin believed that the motivation to change was strongly related to action: If people are active in decisions affecting them, they are more likely to adopt new ways. "Rational social management", he said, "proceeds in a spiral of steps, each of which is composed of a circle of planning, action and fact-finding about the result of action".

Lewin's description of the process of change involves three steps: Unfreezing: Faced with a dilemma or disconfirmation, the individual or group becomes aware of a need to change. Changing: The situation is diagnosed and new models of behavior are explored and tested. Refreezing: Application of new behavior is evaluated, and if reinforcing, adopted.Figure 1 summarizes the steps and processes involved in planned change through action research. Action research is depicted as a cyclical process of change. The cycle begins with a series of planning actions initiated by the client and the change agent working together. The principal elements of this stage include a preliminary diagnosis, data gathering, feedback of results, and joint action planning. In the language of systems theory, this is the input phase, in which the client system becomes aware of problems as yet unidentified, realizes it may need outside help to effect changes, and shares with the consultant the process of problem diagnosis. The second stage of action research is the action, or transformation, phase. This stage includes actions relating to learning processes (perhaps in the form of role analysis) and to planning and executing behavioral changes in the client organization. As shown in Figure 1, feedback at this stage would move via Feedback Loop A and would have the effect of altering previous planning to bring the learning activities of the client system into better alignment with change objectives. Included in this stage is action-planning activity carried out jointly by the consultant and members of the client system. Following the workshop or learning sessions, these action steps are carried out on the job as part of the transformation stage.The third stage of action research is the output or results phase. This stage includes actual changes in behavior (if any) resulting from corrective action steps taken following the second stage. Data are again gathered from the client system so that progress can be determined and necessary adjustments in learning activities can be made. Minor adjustments of this nature can be made in learning activities via Feedback Loop B (see Figure 1). Major adjustments and reevaluations would return the OD project to the first or planning stage for basic changes in the program. The action-research model shown in Figure 1 closely follows Lewin's repetitive cycle of planning, action, and measuring results. It also illustrates other aspects of Lewin's general model of change. As indicated in the diagram, the planning stage is a period of unfreezing, or problem awareness. The action stage is a period of changing, that is, trying out new forms of behavior in an effort to understand and cope with the system's problems. (There is inevitable overlap between the stages, since the boundaries are not clear-cut and cannot be in a continuous process). The results stage is a period of refreezing, in which new behaviors are tried out on the job and, if successful and reinforcing, become a part of the system's repertoire of problem-solving behavior. Action research is problem centered, client centered, and action oriented. It involves the client system in a diagnostic, active-learning, problem-finding and problem-solving process. Data are not simply returned in the form of a written report but instead are fed back in open joint sessions, and the client and the change agent collaborate in identifying and ranking specific problems, in devising methods for finding their real causes and in developing plans for coping with them realistically and practically. Scientific method in the form of data gathering, forming hypotheses, testing hypotheses and measuring results, although not pursued as rigorously as in the laboratory, is nevertheless an integral part of the process. Action research also sets in motion a long-range, cyclical, self-correcting mechanism for maintaining and enhancing the effectiveness of the client's system by leaving the system with practical and useful tools for self-analysis and self-renewal.