Staff Development and School Change
by Milbrey Wallin McLaughlin & David D. Marsh - 1978
A secondary analysis of the Rand Change Agent Study describes four broad factors underlying the understanding of staff development: institutional motivation, implementation strategies, institutional leadership, and teacher characteristics. These factors are used to suggest specific staff development activities. (Source: ERIC)
The desultory status of staff development as education's neglected stepchild is changing. Until recently, there was little interest in the professional development of experienced teachers. Teacher-training institutions were preoccupied with preservice education and local school districts were struggling to accommodate burgeoning student enrollments and build new schools. But now issues related to staff development have moved to center stage. One reason for the new status of staff development is the recognition that many of the "Great Society" education reform efforts fell short primarily because planners seriously underestimated teacher-training needs. In retrospect, it was unrealistic to expect that classroom teachers could bring about significant change in the services provided to such special student groups as the disadvantaged and the bilingual without substantial in-service education. In the absence of such training, it is not surprising that the result of many reform efforts was, disappointingly, more of the same. An important lesson of the so-called "Decade of Reform" (1965-1975) is that even the "best" educational practice is unlikely to fulfill its promise in the hands of an inadequately trained or unmotivated teacher. We have learned that the problem of reform or change is more a function of people and organizations than of technology.
A second reason for the current interest in staff development is that the last decade's period of unprecedented growth has been followed by an equally dramatic decline in pupil enrollment. The market for new teachers is practically nonexistent andfor the first time in many yearslocal school districts find themselves with a stable, and tenured, staff. Thus, teacher-training institutions are confronted with the need to move from a focus on preservice education; local school districts can no longer rely on "new hires" to bring fresh ideas into district classrooms and must face the problem of how to upgrade the skills of the teachers they already have.
However, the only consensus that appears to exist about staff development is that what we have now is ineffective and a waste of time. The general feeling is that most staff-development programs have benefited neither teachers nor students.
If effective staff development is not an isolated workshop or an evening extension course, what is it? Rand's study of federal programs supporting educational change looked closely at the local process of changeand at the factors that support teacher growth.1 The Change Agent study deals with a number of issues that are central to the design and implementation of staff-development programsfor example, what motivates teachers to acquire new skills? What helps teachers to retain these skills? What can the principal do to support and sustain teacher change? What is the role of the central administration in the efforts of classroom teachers to improve their practices? This essay draws on the Rand Change Agent study to suggest issues that will be central to rethinking both the nature and the role of staff-development programs.
THE RAND FINDINGS
The Rand study examines staff development in the context of broader changes in schools associated with various types of federally funded projects. The study is rich in implications for in-service education for several reasons. The study used "outcome" measures that correspond directly to the anticipated results of in-service education. These outcomes include change in teacher practices, pupil growth, and the retention of teacher change in the form of continued use of project methods and materials following termination of federal funds. The study also included many of the process variables considered in in-service education such as teacher commitment and involvement, staff reward structures, skills training and classroom follow-up, and the role of the principal or school climate in teacher growth and the maintenance of change. Finally, in the view of the study's respondents, "successful change" and "staff development" were essentially synonymous.
The Change Agent study identified four clusters of broad factors as crucial to the successful implementation and continuation of local change efforts. These clusters are: institutional motivation, project implementation strategies, institutional leadership, and certain teacher characteristics. This section will discuss each of these clusters of factors and examine its relationship to the extent of project goals achieved, the extent of teacher change, the extent of student growth, and the continued use of teacher methods and materials following termination of federal funds.2
Institutional motivation is the first cluster of factors that the study found to be critical to project outcomes. A school district may undertake a special project, and a school or teacher may agree to participate in the project, for very different reasons. A district may initiate a change-agent project to address a high priority needor it may start a project to ameliorate community pressures, or to appear "up-to-date," or simply because the money is there. Similarly, teachers participate in a special project effort because they are "told to," or because it is their own idea, or because of collegial pressure, or because they see the project as an opportunity for important professional growth. The institutional motivations that characterize a planned change effort significantly influence both project implementation and the extent to which project methods and strategies are eventually incorporated into regular school or district practice.
Not surprisingly, the commitment of project teachers is very important. The Rand Change Agent study found that teacher commitment had the most consistently positive relationship to all the project outcomes (e.g., percentage of project goals achieved, change in teachers, change in student performance, and continuation of project methods and materials). The importance of teacher commitment to the achievement of project goals is axiomatic: Project success is unlikely unless teachers want to work, hard to make it happen.
Though few disagree that teacher commitment is a necessary ingredient to project success, there is debate about the extent to which the commitment of teachers can be affected by policies or program strategies. A number of practitioners and plannersperhaps turned somewhat cynical by a parade of disappointing change effortshave come to believe that teacher commitment is essentially "immutable": Some teachers are eager to change and learn new practices, and some simply are not. The policy implications of such a perspective are discouragingthat is, efforts to improve educational practice should be limited to those teachers who evidence strong initial interest and motivation. The Change Agent study suggests a much less deterministic view. Both the fieldwork and the survey analysis suggest that teacher commitment is influenced by at least three factors: the motivation of district managers, project planning strategies, and the scope of the proposed change-agent project. How does each of these factors affect teacher commitment?
MOTIVATION OF DISTRICT MANAGERS
The attitudes of district administration about a planned change effort were a "signal" to teachers as to how seriously they should take a special project. The fieldwork offers numerous examples of teachersmany of whom supported the project goalswho decided not to put in the necessary extra effort simply because they did not feel that district administrators were interested. In the absence of explicit district support for their efforts, many teachers felt that the personal costs associated with a change effort were not in their professional self-interest and were unlikely to make a difference in the long run. Consequently, few of the projects that were initiated by the district primarily for opportunistic or political reasons were effectively implemented; none were officially continued after the end of federal funding. As one respondent put it, "The superintendent had better believe in the projectgive it his personal backing and support. Teacher confidence is essential; teachers should see in the beginning that top administration believe [in the project] and are committed to it."
The Change Agent study further suggests that administrative support must be generated at the outset. Given the multiple demands facing district administrators, local project planners cannot expect to secure this interest and backing once the project activities are under way. But even were this possible, the initial disinterest of the district leadership will likely leave a negative legacy upon the commitment and attitudes of project teachers.
The commitment of both district administrators and teachers was also influenced by the way the project was planned initially. Four general patterns of project planning characterized change-agent projects and had very different implications for staff commitment and project outcomes. One could be called a "top-down" strategy. In this case, project plans were made almost entirely in the central office and announced to would-be implementors. A second pattern, "grass-roots planning," was just the oppositeplans were devised by teachers or school-based project staff with little involvement of district administrators. A third planning strategy was one of essentially "no planning." A project plan and project funds were imported into the district with little or no involvement from district staff at any level.
"Collaborative planning" characterized the fourth general planning strategy. In this mode, project plans were made with equal input from teachers and district managers. Although this style was rarely characterized by conscious notions of "parity," participants at all levels in the system were treated as partners in the process of planning for a special project effort. Of these four planning strategies, a collaborative planning style was necessary to both the short-term and long-run success of a planned change effort.
Top-down planning strategies typically resulted in indifferent implementation and spotty continuation even when district officials were committed to project goals and serious about the change effort. Top-down planning usually met with indifference or resistance from the school staff. Teachers felt that such projects were not "theirs" but the central administrators' and had little personal investment in project objectives or success.
Top-down planning strategies often resulted in disappointing projects for another reason: Central office staff were insufficiently aware of the needs and practices of particular schools, classrooms, and teachers. One teacher made this revealing comment:
This project hasn't worked out and its main effect has been to cause a close, well-organized faculty to turn to distrust each other. This was the result of forcing a program on a school, using an outside coordinator unfamiliar with the school and faculty, and not having the full support of teachers. I personally felt the project ideas were good and could have worked if the teachers in our school had been involved in the planning.
Or, as another teacher in an unsuccessful project complained, "The project was planned and designed without the knowledge and consent of the teachers at the school . . . the planner had hardly ever been to our school." Top-down planning generally fails even with the best of intentions both because it cannot generate the staff commitment necessary to project success and because this planning style does not incorporate the special knowledge and suggestions of the staff who will be responsible for project implementation.
The second planning strategy, grass-roots planning, was only a little more successful. Projects that were conceived and planned at the school level with only cursory review by central office staff often evidenced high initial teacher commitment. But, in the absence of explicit support from district managers, that commitment waned over the course of project implementation. Project teachers found it difficult to sustain initial enthusiasm and motivation when there was little indication that district officials also cared. In the long run, grass-roots projects generally disappeared as completely as top-down projects. Though teacher-initiated project practices and methods could be found in isolated pockets of the district, without district support, even successful projects withered away because of factors such as staff turnover, indifference, or lack of information on the part of building principals.
Only the fourth (and more time-consuming) planning strategy, collaborative planning, generated the broad-based institutional support necessary to effective implementation and to the continuation of successful practices. Projects adopting this planning style actively engaged both teaching and administrative staff from the preproposal period through implementation, thereby gaining consensus and support from teachers, principals, and central office personnel. Evidence of the Change Agent study on this point gives the lie to the conventional wisdom that teacher-initiated projects are usually more successful than are those conceived downtown. "Who" originated a project did not matter. For example, teacher change and the continuation of project strategies were not significantly different in schools that had originated the project from what they were in schools that had been asked to participate. What did matter was "how" project planning was carried out, regardless of the source of the idea.
SCOPE OF CHANGE
A third factor that influenced teacher motivation is the scope of change proposed by a project. The Rand study found that the more effort required of project teachers, and the greater the overall change in teaching style attempted by the project, the higher the proportion of committed teachers. Complex and ambitious projects were more likely to elicit the enthusiasm of teachers than were routine and limited projects. The reason for this, we believe, is that ambitious projects appeal to a teacher's sense of professionalism. Evidence from the Change Agent study indicates that a primary motivation for teachers to take on the extra work and other personal costs of attempting change is the belief that they will become better teachers and their students will benefit. Fieldwork observation and interviews with practitioners suggest that the educational promise of an innovation and the apparent opportunity for professional growth are crucial factors in generating teacher commitment.
The importance of professionalism, or intrinsic motivation, to teacher motivation (and thus to project success) raises questions about the utility of extrinsic rewards as a project strategy. Many groups have proposed extrinsic rewardscredit on the district salary scale, extra pay, and so onas possible solutions to the "problem" of teacher motivation. Although the Change Agent study did not consider this issue comprehensively, it did examine the effects of extra pay for attending staff training sessions. Sometimes this strategy was used "to get the teachers to go along" with a project, or to "sweeten the pill." Teachers who received extra pay for training (about 60 percent of the sample) were less likely than others to report a high percentage of project goals achieved. These teachers also reported less improvement in student performance, especially academic performance, than did other teachers in the study.
These findings support the idea that instrinsic professional rewardssuch as those implicit in the proposed scope of changeare far more important in motivating teachers. To this point, a number of project directors commented that although the teachers appreciated the extra pay, the pay alone did not induce teachers to work hard to learn new skills if professional motivation was absent. As one teacher remarked, "I'll go [to the training session], and I'll collect my $30, but I don't have to listen."
PROJECT IMPLEMENTATION STRATEGIES
The second critical factor affecting the outcomes of local change efforts was the project implementation strategy. Among the most important choices made during the initial planning period were those about how to put the project into practice. Local planners had considerable discretion in selecting project implementation strategies. For example, similar reading projects utilized very different staff-training strategies and project-governance procedures. The most important of those local choices were those that determined the ways in which the school staff would be assisted in acquiring the new skills and information necessary to project implementation staff-development strategies.
Project strategies that fostered staff learning and change had two complementary elements: (1) staff-training activities', and (2) training-support activities. The study found that well-conducted staff training and staff-training support activities improved project implementation, promoted student gains, fostered teacher change, and enhanced the continuation of project methods and materials. These training and support variables alone accounted for a substantial portion of the variation in project success and continuation.
This in itself is not a surprising finding. After all, teachers have to acquire new skills or behavior if project-related changes are to influence student performance and if project strategies are to be continued. But the very different contributions made by these two activities staff training and support activities furnish important insights for staff-development planners.
Staff-training activities were typically skill specificinstruction in how to carry out a new reading program or introduction to new mathematics materials, for example. As such, this component of a project's implementation strategy corresponds most directly to the traditional focus of many in-service teacher education efforts. Skill training usually involved project workshops, and could occur prior to project implementation, during the first year of project operation, or after the first year. By themselves, staff-training activities had strong, positive effects on the percent of project goals achieved, and on student performance in the areas of both achievement and behavior. However, skill-specific training had only a small and not significant effect on teacher change and on the continuation of project methods and materials. In other words, skill-specific training alone influenced student gains and project implementation only in the short run.
This finding is puzzling at first glance. After all, if staff-training activities significantly and positively affected student performance during the period of project operation, why didn't this effect continue after special funding was terminated? The Change Agent study results suggest a straightforward explanation for this apparent anomaly. Skill-specific training activities only have transient effect because, by themselves, they do not support staff learning and teacher change. Skill-specific training enabled teachers to implement new project methods and materials under the aegis of special-project operation. But this implementation was often mechanistic and did not necessarily constitute teacher assimilation of the new techniques and procedures. Thus, when the supports of the funded project operation were removed, teachers discontinued using the practices that apparently enhanced student performance because they had never really learned them in the first place. Skill-specific training, in short, can affect project implementation and student outcomes, but it does not affect the longer-term project outcomes of teacher change and continuation. Staff-support activities are necessary to sustain the gains of how-to-do-it training.
Projects pursued a number of activities to support teacher assimilation of the skills and information delivered in training sessions. In particular, the study examined the contribution of classroom assistance by resource personnel, the use of outside consultants, project meetings, and teacher participation in project decisions. Taken together as a support strategy, these activities (when they were seen as useful by the school staff) had a major positive effectas did staff trainingon the percentage of project goals achieved and on student performance. But in contrast to staff-training activities, these support activities also had strong positive and direct effects on the longer-term project outcomesteacher change and continuation of project methods and materials. Well-conducted staff-support activities not only reinforce the contribution of staff training, but they also make their own important contribution to promoting teacher change and to supporting staff assimilation of project practices.
Training is essentially an information transferproviding teachers with necessary techniques. But, as the first phase of this study found, the process of implementation is a process of mutual adaptation in which teachers modify their practices to conform to project requirements and project technologies are adapted to the day-to-day realities of the school and classroom. Staff-support activities, in particular classroom assistance from resource personnel and project meetings, can provide the feedback project staff need to make these modifications. Through these support activities, skill-specific training can be "individualized" for project teachers in terms of timing and content modification.
Staff-support activities can also aid teachers in understanding and applying complex new strategies in ways that standard trainingin terms of both form and contentusually cannot do effectively. For example, even a carefully planned staff-training program usually cannot anticipate the nature or the timing of project staff-assistance requirements, especially as they relate to particular classroom problems. Likewise, staff often cannot perceive what they need to know until the need arises. For both reasons, the needs of project staff are not always predictable or synchronized with scheduled training sessions. The utilization of local resource personnel or consultants to provide "on-line" assistance can help remedy these inevitable deficiencies. However, it is important to note that the quality of this assistance is critical. The study found that the amount of classroom assistance from local resource personnel did not matter when teachers perceived their help as useful or very useful. But the frequency of classroom visits did have an effect when it was not perceived as helpful. Numerous visits to the classroom by district or project staff were counterproductive when teachers did not feel they were being helped. This assistance actually interfered with project implementation.
Similarly, it was better for projects to use no outside consultants than to use poor onesand much better than to use poor ones often. Good consultants helped by providing concrete practical advice to project teachersshowing them how to adapt project methods or materials to their own situation. Good consultants assisted teachers in learning how to solve problems for themselves, rather than by solving problems for them. Ineffective consultants often furnished advice that was too abstract to be useful. In making a recommendation for improving project implementation, one teacher advised, "Be sure consultants know [the project] goals and some specific things to tell the teachers and not a lot of worthless generalizations and theory." Another teacher remarked, "I found most [of the consultants] to be completely lacking in their exposure to, familiarity with, and willingness to come in and work with young children. Many were good philosophically, but not practically, in the day-to-day approach and follow-up."
Ironically, even "good" consultants actually diminished project outcomes in some cases. Consultants often unintentionally preempted staff-learning opportunities and prevented teachers from learning to implement project strategies for themselves. One superintendent attributed the failure of a project to this factor: "The first year, teachers came in from other communities and worked with our teachers. The following year, our teachers were alone and it was impossible to fully implement the program." The negative effects of consultants that appear in the Change Agent data can be interpreted both as a result of too little and too much help from consultants.
Frequent project meetings were another support strategy that aided teacher efforts to adapt project precepts to their classrooms and assimilate new strategies. Project meetings provided a forum whereby teachers could learn from one another's experience. Project meetings also supported the affective needs of teachers as they attempted to implement change. As one teacher commented, "Regular monthly meetings are absolutely critical for reinforcement and building interpersonal relationships for co-workers." However, like consultants and classroom assistance, if meetings were not perceived as useful, they had a detrimental effect on project operations. Frequent meetings that were not judged useful by teachers were strongly associated with less successful projects in the survey sample. The fieldwork suggested that meetings were unproductive when they dwelled primarily on details of project administration and record keeping and rarely included opportunities for staff to share their problems and reports on process. Such meetings did little to enhance classroom implementation, and teachers found them irritating.
All of these project activities to support trainingclassroom assistance, outside consultants, frequent meetingscontributed to project outcomes in yet another and equally critical way. These support activities were necessary for the development of clarity concerning the goals of the project and the implications of project strategies for ongoing classroom practices. It is important to note that clarity was not the same as programmatic specificity, nor were these factors necessarily related.
The analysis showed that specificity of goals had a major effect on implementation: The more specific the teachers felt the project goals were, the higher the percentage of goals the project achieved, the greater the student improvement attributed to the project, and the greater the continuation of both project methods and materials. Program planners and grants-makers, hoping to enhance special project outcomes, have placed increasing emphasis on the careful specification of project objectives. However, such programmatic specificity is only one component of the broader specificity so important to project success. Even more important is the second component of specificitynamely, conceptual clarity, or the extent to which project staff are clear about what they are to do and understand the rationale underlying project activities.
Programmatic specificity is fundamentally a project-design issue and, by itself, does not guarantee staff clarity about project operations. There were many projects in the study in which even clearly stated project objectives made little sense to project staff in terms of their day-to-day responsibilities. Furthermore, programmatic specificity of the type advocated by grants-makers and planners is often not feasible for many ambitious projects, such as projects that focus on change in classroom organization. Conceptual clarity may be fosteredbut cannot be assuredby specific project-goal statements or by the use of packaged materials or by lectures from outside consultants. The conceptual clarity critical to project success and continuation must be achieved during the process of project implementationit cannot be "given" to staff at the outset. Frequent staff meetings and timely classroom assistance by resource personnel are strategies that provide staff with this practical understanding concerning the project goals and methods promulgated in training sessions and project designs.
The effectiveness of both training and training-support activities was enhanced by another project-implementation strategyteacher participation in project decision making. Decisions and choices during implementation would not be necessary if projects were always carried out as originally planned. Most projects, however, particularly successful projects, underwent modification in their initial plans and objectives, and these adaptations were almost always positive improvements.
The strong, positive effect of teacher participation on the percent of project goals achieved suggests that teacher inputs can significantly improve implementation. Teachers, because of their day-to-day involvement with project operations, are in a much better position than district specialists or even the project director to identify problems and recommend feasible solutions. To this point, one elementary school principal advised, "Give the classroom teacher a strong role in planning any project that he or she is going to be working with. Then listen and change when things do not go as planned on paper."
Teacher participation in decisions about the project had an important instrumental value: Teacher suggestions improved the implemented project, and staff participation in reviewing and modifying project procedures significantly enhanced staff clarity. Teacher participation also made an important affective contribution to project implementation, namely, development of teachers' "sense of ownership."
Institutional leadership is a third important factor for the successful implementation and continuation of a local change project. Indeed, the Change Agent study suggests that project planners need to enlarge their notion of the leadership critical to project outcomes. District planners invest considerable time in identifying competent, enthusiastic leadership for special project efforts, because their concern about leadership typically focuses on the project director. Research underscores the importance of project director leadership to project outcomes. The Change Agent data show that the more effective the project director (in the view of teachers), the higher the percentage of project goals achieved, and the greater the student improvement observed as a result of the project. An effective project director has significant instrumental value to project implementationa director's special skills or knowledge can foster staff understanding of project goals and operations, minimize the day-to-day difficulties encountered by classroom teachers, and provide the concrete information staff need to learn during the course of project operations.
The data also indicate, however, that effective project leadership plays only a short-term and circumscribed role in the outcome of local change-agent projects. The effectiveness of a project director had no relationship to project continuation or to teacher change. Both the fieldwork and the survey analysis point to other components of school district leadership as critical to these important longer-term project outcomes.
The support and interest of central office staff was, as suggested earlier, very important to staff willingness to work hard to make changes in their teaching practices. Though a skilled and enthusiastic project director may be able to effectively implement a special project in the absence of explicit support from "downtown," project staff are unlikely to continue using project strategies unless district administrators express interest.
The attitude of the building principal was even more critical to the long-term results of a change-agent project. The support of the school principal for a special project was directly related to the likelihood that staff would continue to use project methods and materials after special funding is withdrawn. Furthermore, principal support positively affected project implementation. The Phase II survey asked teachers to indicate the attitude of their principal toward the project. Few of the projects in which the principal was perceived to be unfavorably inclined toward the project scored well on any of the study's outcome measurespercent of goals achieved, teacher change, student improvement, or continuation. Some projects with neutral or indifferent principals scored well, particularly in the percent of goals achieved, but these projects typically focused on individualized instruction or curriculum revisionactivities that could occur almost completely "behind the classroom door" and in which highly effective project directors could compensate for lukewarm principals. Projects having the active support of principals were most likely to succeed, and to be continued.
Why is the principal, not the project director, so important to long-term project outcomes? At the end of federal funding, the principal must take a stance toward the project and make a variety of decisions that explicitly or implicitly influence what happens to project methods and materials within the school. In particular, the principal is chiefly responsible for establishing the school's educational policies and philosophy. A project that is in agreement with the school's general operating style would be more likely to be sustained than one that was not. For example, the fieldwork examined an open-classroom project that operated in a very traditional school as part of a district-wide project. Once the umbrella of project authority was removed, the principal made it clear to project teachers in the school that he wished to see their classrooms returned to the traditional pattern; he also strongly discouraged nonproject teachers who expressed interest in trying some of the project ideas in their classrooms. In the same district, however, the principal at another school strongly supported the open-classroom approach. After the project ended, this principal encouraged the use of project methods in other classrooms and allocated discretionary money to purchase the necessary materials.
In short, the building principal gives subtle but nonetheless strong messages concerning the "legitimacy" of continuing project operations in the schoola message that teachers cannot help but receive and interpret in terms of their professional self-interest. Support from the principal is also important to the longevity of special project strategies because of the staff turnover experienced by most schools. If project methods are not to dissipate over time, the principal will have to familiarize new teachers with project concepts and techniques. As one superintendent observed. "A large turnover in staff [makes it hard to] sustain volunteer activities. If you get a principal who isn't in agreement with project philosophy, it can be difficult to keep a program in a school."3
One way in which principals demonstrated their active support for project activitiesas well as gained the information necessary to promote continuation of project strategieswas to participate in project training sessions. Involvement in project training updated their classroom skills and knowledge, and equipped them to lend advice and a sympathetic ear to project teachers. But equally as important, the attendance of principals in project training imparted some important messages to teachersnotably, their personal commitment and their view that the project was a school effort in which everyone was expected to cooperate and work hard. In this way, principal attendance at project sessions helped undermine the "deficit" model that sometimes colors staff-training activities and builds resentment of the project as something done "to" teachers.
In summary, the quality of the leadership available to project staff was critical to the successful implementation and continuation of change-agent project efforts. However, it was not enough simply to provide the special project with a skilled and enthusiastic project director. The efforts of even a talented director were likely to be ephemeral unless central-office leadership supported the efforts of project staff and unless the school principal actively engaged in project activities. Evidence from the Change Agent study suggests that the task confronting planners in establishing effective leadership for project activities must be construed in broad, institutional terms, not in narrow, special-project terms.
To this point, the study found that the school climate was as important as the principal as an influence on continuation of project methods and materials once the federal funding was terminated. The Rand data indicate that good working relationships among teachers enhanced implementation and promoted continuation of project methods and materials. Good working relationships and teacher participation in project decisions were correlated: The development of the one helped the development of the other. And, in addition, the quality of the school's organizational climatewhether teachers felt their school was a good school to work in, had esprit de corps, was efficient, and was managed effectively by the principalinfluenced the quality of project relationships. The correlation between participation in project decisions and good staff working relationships draws attention to the implementation strategies chosen for the project; in particular the influence of the general school climate a background factor not directly related to project operationsunderlines the significance of district site selection. Good project working relationships could develop in "average" schools when teachers participated in project decisions; and, conversely, "good" schools could develop good project working relationships without teacher participation in decisions. However, projects combining a supportive organizational environment with a strategy of teacher participation in project adaptation were most able to implement effectively and continue their innovations.
The fourth general factor that the Rand study found had major influence on the outcome of planned change efforts is teacher characteristicsthe attitudes, abilities, and experience teachers bring to a special-project effort. A "conventional wisdom" has developed concerning the effects of various teacher attributes: that older teachers are less willing to change, that the best ideas come from younger teachers, that teachers with high verbal ability are more able to achieve gains in student performance, and so on. Such beliefs suggest that the personal characteristics of project teachers could have significant import for the outcome of planned change efforts. The study collected information on several teacher attributes most often cited as significantly influencing both student performance and the outcome of innovative projects: age, educational background, verbal ability, years of experience, and sense of efficacy.
Three teacher attributesyears of experience, verbal ability, and sense of efficacyhad strong and significant, but very different, effects on most of the project outcomes. Specifically, the number of years of teacher experience was negatively related to all of the dependent variables, with the exception of teacher continuation of project techniques, where there was no relationship. In other words, the more experienced the project teacher, the less likely was the project to achieve its goals, and the less likely was the project to improve student performance.
These relationships in large part are attributable to the fact that the more experienced teachers also were less likely to change their practices as a result of project participation. Moreover, both the fieldwork and the survey analysis suggest that teacher tenure has a curvilinear relationship to project outcomes. That is, teachers seem to "peak out" after five to seven years of teachingeither maintaining their level of effectiveness (in the best cases) or actually becoming less effective. For many teachers in the Rand study, the passage of time on the job seemed to diminish their capacity to change and to dampen their enthusiasm for innovations and for teaching. This "calcifying" effect seemed less an intrinsic characteristic of teachers or of the teaching role than testimony to the way schools are managed and the way professional development activities are provided for staff.
In particular, the professional-development needs of experienced teachers are different from those of new teachers. For example, the workshop approach that may be useful for teachers still mastering the classroom craft is not sufficiently relevant or challenging to more experienced teachers. After several years in the classroom, teachers want to explore new areas and take more responsibility for their professional growth.
District-wideor even school-widein-service education activities that only elaborate on present practice usually are seen as a waste of time by experienced staff. But few schools or districts explicitly address the professional-development needs of their tenured staff. Thus, it is not entirely surprising that experienced teachers sometimes feel there is little challenge left for them and "turn off from teaching.
A second teacher characteristic, verbal ability, was significantly related to only one outcome measuretotal improvement in student performance.4 However, when student performance was broken down into its cognitive and affective components, the data indicate that teacher verbal ability affects only achievement; it apparently had no significant effect on student affective development.
The most powerful teacher attribute in the Rand analysis was teacher sense of efficacya belief that the teacher can help even the most difficult or unmotivated students.5 This teacher characteristic showed a strong, positive relationship to all of the project outcome measures. Furthermore, the effects of a sense of efficacy were among the strongest of all the relationships identified in the analysis. Teacher sense of efficacy was positively related to the percent of project goals achieved, the amount of teacher change, total improved student performance, and the continuation of both project methods and materials. Teachers' attitudes about their own professional competence, in short, appear to have major influence on what happens to change-agent projects and how effective they are.6
An important question for planners is the extent to which a teacher's sense of efficacy can be affected by project-design choices, or whether teacher perceptions about their competence is simply a "given." The study did not measure this teacher attribute before project activities began, and so cannot report "before and after" findings. However, the information that was collected furnishes important insights for planners of in-service education. First, teacher sense of efficacy was not significantly related to years of experience or to verbal ability. In other words, a highly verbal, experienced teacher is no more or less likely to feel a sense of efficacy than are other teachers. Second, teachers having a high sense of efficacy tended to be part of projects that placed heavy emphasis on the staff-support activities discussed earlier. That is, projects that involved teachers in project decision making, that provided timely and ongoing assistance in the classroom, and that had frequent staff meetings were more likely to have teachers with a high sense of efficacy than were projects with narrowly defined goals, that had little teacher participation, or that relied heavily on the use of outside specialists to implement the project.
An obvious question is whether low efficacy and high efficacy teachers were "selected into" these different project types. Did project directors, based on their assessment of a sense of competence on the part of project teachers, encourage or discourage teacher participation in the project? Or, did "low efficacy" teachers tend to avoid projects in which they would have to play a major role? Though such self-selection undoubtedly occurred to some extent, the fieldwork suggests that project training-support activities functioned to enhance teacher efficacy. Staff-support activities seemed to promote teacher efficacy in several ways. They provided timely assistance to teachers and a forum in which teachers could talk through project strategies in terms of their own classrooms and thus feel confident in utilizing new ideas. Furthermore, they allowed peer encouragement and development of a sense of ownership in project activities. Staff-support activities provided teachers with crucial collegial support in their efforts to change and grow.
Projects that sought staff participation and involvement in project decision making also conveyed the message to teachers that school administrators viewed them as competent professionalsable to make important decisions about project activities and objectives. A "Pygmalion effect" of sorts may operate in projects where teachers are given a responsible role. Teachers who are given an opportunity to make decisions about project activities and take responsibility for the substantive direction of their decisions soon acquire such skills.
In summary, the Rand study confirms much of the conventional wisdom concerning the importance of teacher characteristics to the outcome of a planned change effort. But, more importantly, this study also suggests ways in which project-design choices and district leadership can influence these important factors.
The Rand study presents a fundamentally different view of staff development or in-service education from that typically found in the literature or in practice. The study moves away from a traditional view of staff development as a concern about the governance, financing, staffing, delivery, and reward structures for "those workshops" or as a problem of technology transfer. Instead, the Rand study emphasizes learning for professionals as part of ongoing program building in an organizational context. This view of staff development is one of the most important implications of the study.
As part of this view of staff development, the Rand study suggests a number of new assumptions to guide the design and implementation of staff-development activities. First, the study suggests that in terms of knowledge about the practice of teaching, teachers often represent the best clinical expertise available. For example, in teaching as in other clinical settings, the appropriate strategy to resolve problems is unclear. To this point, the state of educational research is such that it is difficult to reach consensus concerning the value of most any set of teaching strategies. For teachers, the learning task is more like problem solving than like mastering "proven" procedures. Consequently, outside experts and tightly structured training are relatively less helpful than they are in technology-dominant activities such as industry. The instrumental value of involving classroom teachers in identifying problems and solutions is clearly expressed in the Rand study.
Second, the Rand study describes the process by which an innovation comes to be used in a local setting as adaptive and heuristic. This mode of implementation exemplifies the professional learning process for the projects in the Rand study. In a sense, teachers and administrative staff need to "reinvent the wheel" each time an innovation is brought into the school setting. Reinventing the wheel helps the teachers and administrative staff understand and adjust the innovation to local needs. Learning occurs throughout this adaptation process as staff come to understand their own needs for additional information. Even clarifying the purpose of the innovation is a learning process in itself. Conceptual clarity about project goals in the study evolved as staff learned to understand the implications and nuances of the innovation.
Besides having training that is individualized according to learning rate and learning style, the study found that training and staff support play different roles in this professional learning process. Skill training typically provides the cognitive information and general skills, which then must be adapted within individual classrooms. Outside consultants typically had a difficult time in meeting the learning needs of staff within the training and support staff framework. Whether intentional or not, consultants typically upstaged staff.
A third and related assumption communicated by the Rand study is that professional learning is a long-term, nonlinear process. In the study, innovation sometimes took one or several years to achieve full implementation. Over this time, teachers and administrative staff needed to learn what innovation was needed and what the innovation ought to look like in their particular school setting. They needed to learn what help was needed to implement the innovation as well as the new ideas and skills contained in the skill training. They also needed to learn how to apply these skills or ideas in their classrooms and even how to retain these skills or materials once federal funding had terminated. The continuation of project methods and materials can be seen as a learning problem where the methods and materials are used under new conditionsthe absence of the supporting structure provided by federal funds.
A fourth assumption suggested by this study concerns viewing staff development as part of the program-building process in schools. In the Rand study, the process of adoption of a specific innovation helped define the program-improvement goal for teachers, administrators, and project staff. It helped to coalesce the commitment and energies of these groups around the implementation of this innovation and focused the resources and expertise needed to complete the learning process. The three groupsteachers, administrators, and project staffdid not necessarily see the goals of the project in congruent terms for several reasons. Moreover, as described above, conceptual clarity evolved as people came to understand the innovation more thoroughly. Consequently, the innovation process helped to focus the energy, commitment, resources, and expertise for staff development even though the perception of the project evolved over time and differed, to some extent, by role group. But it was important that professional learning be related to ongoing classroom activities. Staff-development activities undertaken in isolation from teachers' day-to-day responsibilities seldom had much impact.7
Viewing staff development in the context of program building also helped to shift staff development from a deficit model where teachers are seen as needing in-service because they lack professional skills. In general, the deficit model of staff development is characterized by the view of other educators that teachers need staff development because they lack the necessary skills to teach successfully. This characterization has several elements that need to be understood if the deficit-model approach to staff development is to be changed. First, the deficit model is a collective view supported by members of diverse role groups such as principals, school district administrators, university professors, state department of education officials, and legislators. This leaves teachers with the belief that everyone is critical of them. Secondly, these outside groups bring to bear administrative regulations, credential requirements, university degree requirements, and state law as a network of reinforcement for their belief: The critical view of other educators is being powerfully communicated to teachers. Thirdly, teachers have typically been excluded from any discussion of their "deficit" or any discussion about how to carry out its removal and, finally, the deficit model has been built based on the dogmatic belief of other educators that they know, and can justify, their statements about what constitutes good teaching. Though educational research developed over twenty-five years has not resolved the dilemma of what constitutes good teaching, deficit-model outside experts or central office specialists often act as though they know.
A number of features of staff development as part of program building at the school site help to displace the deficit-model view of staff development among teachers and administrators. In the Rand study, staff development became part of a program-improvement process where many role groups needed new skills: Teachers were not the only group involved in project-skill development activities. Such an approach helped to spread and lessen the psychological risks of change. Teachers also were major decision makers about the innovative process; outside groups were no longer deciding what teachers ought to know. Moreover, teachers perceived that change was possible on a broader scale because the implementation process itself brought about changes in administrative structures, curriculum, and instructional-materials strategies as well as in teacher behavior. In fact, integrating teacher improvement with other aspects of school change such as curriculum development and administrative reform increased the effectiveness of the staff-development efforts.
In this context, the intrinsic motivation associated with teacher professionalism was fostered. This is in sharp contrast to traditional thinking about staff development where the debate seems to focus on which form of extrinsic motivation should be used to get teachers to attend one-shot workshops or other almost ritualistic forms of staff development. In the context of staff development to support the innovation process, the Rand study found that the teacher motivation issue was no longer one of persuading teachers to attend staff-development meetings to remedy their deficiencies.
A fifth broad assumption suggested by the Rand study concerns the importance of seeing staff development in the context of the school as an organization. Within the most successful projects, the project was not a "project" at all, but an integral part of an ongoing problem-solving and improvement process within the school. In a sense, good staff development never ends. It is a continual characteristic of the school site.
The importance of the organization context was apparent in all phases of the change process. Projects that sought federal dollars to address an important concern in the entire school were much more successful than were those that were more opportunistic in their search for money. Successful implementation was also dependent upon the organizational climate and the leadership of the school as well as upon implementation strategies that provided staff with new skills and information but also provided classroom follow-up and other staff support. Extensive teacher participation and a critical mass of school staff needed to bring about change were also necessary ingredients.
Yet it was at the continuation phase of the project where the linkage between staff-development activities and the organizational characteristics of the school became most important. In successful projects, continuation was a dynamic process that helped maintain changes in teacher practices. Active involvement of both the principal and school district leadership was vital to the maintenance of these changes. Contrary to the common belief that the availability of district funds is the main factor in determining whether successful innovations are retained, the Rand study found that district and school-site organizational factors were more important than were financial factors. From the initiation to the effective institutionalization of an innovation, organizational factors play a vital role. The implementation and especially the long-term effectiveness of in-service efforts are very much influenced by these same organizational forces.
In summary, the Rand study suggests that effective staff-development activities should incorporate five general assumptions about professional learning:
Teachers possess important clinical expertise.
Professional learning is an adaptive and heuristic process.
Professional learning is a long-term, nonlinear process.
Professional learning must be tied to school-site program-building efforts.
Professional learning is critically influenced by organizational factors in the school site and in the district.
These assumptions support a view of staff development emphasizing learning for professionals as part of program building in an organizational context.
The broad view of staff development given by the Rand study has a number of implications for teachers, teacher organizations, school and school district leadership, and universities. For teachers, this approach to staff development implies long-term teacher responsibilities, collaborative planning, and implementation of significant change in schools. In many ways, this is the positive opportunity teachers have asked for for some time. Yet in the past, the invitation for teachers to participate in collaborative planning and implementation of significant change has been a mixed message. Teachers were invited to participate without having significiant decision-making power and without time being given for them to participate meaningfully. Moreover, school district administrators and colleges often set up a host of bureaucratic regulations that made authentic teacher participation quite difficult. The current financial, legal, and political tension within school districts means that the current invitation for teacher participation is a complex one. The spring of each school year is better symbolized by dismissal letters and budget cuts than it is by opportunities for new collaborative planning around program improvement.
Teachers will also have to overcome a resulting tendency to feel that they are the victim of external forces. Currently, teachers seem impatient with the long and often arduous process of collaborative planning, learning, and adaptation necessary to make innovations successful. Ironically, teachers currently want to give priority to their role as classroom teachers when the Rand study suggests that their role as collaborative planners has become increasingly important in the context of creating ongoing problem-solving capabilities within schools.
The Rand study also suggests that more experienced teachers may need a different approach to their professional growth than is contained in staff development as part of the implementation of innovations. Some experienced teachers continue to grow personally and professionally, and their contribution as teachers is well respected. In general, however, the Rand study suggests that teachers with many years of experience find it more difficult to bring about change in their own teaching behavior and to maintain the use of new teaching strategies and new teaching materials over time. Lortie8 also found that many older teachers had shifted a good deal of their energies to family or other outside-of-school interests, either out of frustration or weariness. We believe that a more personal approach to professional growth may be important for more experienced teachers. This personal approach should emphasize new cognitive frameworks for looking at teaching practice and at their effectiveness as teachers. The apparent mutability of a teacher's sense of efficacy suggested by the Rand study suggests that experienced teachers need not peak out, but can continue to learn and grow.
The Rand study has several implications for the teacher center movement. Our view of staff development emphasizes professional learning as part of program building in an organizational context. Current thinking about teacher centers emphasizes many of the qualities and many of the characteristics of professional learning that we have described above. Yet, some of the writing about teacher centers lacks sufficient concern about program building within schools or the organizational context for staff development, which we feel may hinder the ultimate effectiveness of teacher centers as vehicles for staff development. However, teacher centers have a number of features we find very attractive. The insistence on extensive teacher participation, the call for practical training that is perceived as useful by teachers, and the linkage of teacher behavior and curriculum materials are all strongly supported in the Rand findings.
The Rand research has a number of implications for the roles of principals and school district leadership. The Rand study gives new meaning to the role of instructional leadership for school principals. This instructional role usually connotes activities such as clinical supervision with individual teachers or conducting staff meetings at the school site. The Rand research sets the role of the principal as instructional leader in the context of strengthening the school-improvement process through team building and problem solving in a "project-like" context. It suggests that principals need to give clear messages that teachers may take responsibility for their own professional growth. The results also emphasize the importance of principals and school district leadership giving special attention to the task of continuation of teacher change and innovation at the school. Administrative involvement for such continuation includes early support for the continuation phase of the innovation cycle, administrative participation during the implementation of the innovation, and attention to the organizational as well as financial consideration for program continuation. We have learned that innovation is more a learning process than a systems design problem. Administrators enamored with a systems design or technology transfer notion of change will find little encouragement in the Rand research. On the other hand, the role of administrator was not merely managing the educational enterprise in a static fashion. Successful local projects were part of a dynamic, problem-solving organizational framework headed by a committed administrator.
The Rand research also points to the need for staff development for principals and district administrators. Ironically, these groups have been ignored in federal legislation concerning local educational reformin part because staff development for administrators would make the federal government appear to be taking too heavy an intervention role in local district affairs. The staff-development needs of middle-level managers are usually ignored in most school districts as well. The Rand study suggests that staff development for principals is critical. It is needed to strengthen their ability to carry out the many facets in the innovation process in the context of building an ongoing problem-solving capacity at the school.
Finally, the ineffectiveness of outside consultants in the implementation process raises serious questions about the roles that universities can play in school-based staff-development programs. It is clear that packaged in-service programs, especially those offered without extensive classroom follow-up and teacher participation, are not likely to be effective according to the Rand research. In turn, however, universities could play several creative roles. First, they could prepare administrators and other school leaders who are able to carry out the innovation process as described in this study. Secondly, in their preservice teacher education programs, they could prepare teachers to play the secondary role of collaborative planner within a problem-solving dynamic organization. In any role that universities are to take in support of school-based staff-development programs, it is clear that they need to be part of the ongoing developmental process at the school. This means that they will need to be a part of the collaborative planning and implementation process at the school site. They would need to provide concrete, timely training that is perceived as useful by the teachers and be willing to help in the classroom follow-up process. University faculty would also need to be credible in the school setting and themselves be willing to undergo an adaptation process as they take on these new roles. In short, the Rand study suggests that universities will have to implement significant change themselves if they are to be effective partners in school district staff-development efforts.
1 Rand has just completed, under the sponsorship of the United States Office of Education (USOE), a four-year, two-phase study of federally funded programs designed to introduce and spread innovative practices in public schools. This study of federal programs supporting educational change is often referred to as the "Change Agent Study." The first phase of the study addressed those factors affecting the initiation and implementation of local "change-agent" projects. The second phase of the study examined the institutional and project factors that influenced the continuation of innovations after special federal funding terminated.
The study collected extensive information from superintendents, district federal program officers, project directors, principals, and teachers about the local process of change. In the first phase of the study, 293 local projects were surveyed and fieldwork was conducted in 24 school districts. The second phase of the study involved a survey of 100 projects in 20 states and fieldwork in 18 school districts.
The results of the first phase of the study are summarized in Paul Berman and Milbrey Wallin McLaughlin, Federal Programs Supporting Educational Change, Vol. IV: The Findings in Review (Santa Monica, Calif.: The Rand Corporation [R-1589/4-HEW], April 1975). The findings of the second phase of the Change Agent study are reported in Paul Berman and Milbrey Wallin McLaughlin, Federal Programs Supporting Educational Change, Vol. VII: Factors Affecting Implementation and Continuation (Santa Monica, Calif.: The Rand Corporation [R-1589/7-HEW], April 1977).
Also see Making Change Happen? a series of articles on the Phase I study in the Teachers College Record 7, no. 3, February 1976.
2 These five dependent measures as discussed in this article (percentage of project goals achieved, total teacher change, total student performance gain, continuation of teacher methods, and continuation of project materials) were continuous variables derived from the teacher questionnaire. Berman and McLaughlin, Federal Programs Supporting Educational Change, Vol. IV, chap. IV, describes these measures and their validity in detail.
3 The enthusiasm of principals is also an important element in introducing project methods to additional school sites. A superintendent commented, "This project has really been sustained through the direction and enthusiasm of principals. They were tremendously enthused, at first particularly, and so the project spread to other schools."
4 Teachers' verbal ability was measured by a self-administered Quick Word Test consisting of a fifty-question, multiple-choice, vocabulary-type test. The test, developed by Edgar F. Borgatta and Raymond Corsini (Quick Word Test [New York: Harcourt Brace Jovanovich, nd]) as a measure of verbal abilities, has high reliability and is correlated highly with more complex measures of intelligence. See Berman and McLaughlin, Federal Programs Supporting Educational Change, Vol. VII, p. 137.
5 The measure of teachers' sense of efficacy was based on two questions. One asked whether the teacher felt that "when it comes down to it, a teacher really can't do much because most of a student's motivation and performance depends on his or her home environment." The other asked whether the teacher thought that "if I really try hard, I can get through to even the most difficult or unmotivated students." Responses to these two questions were combined into a single measure of efficacythe extent to which the teacher believed he or she had the capacity to affect student performance. Ibid.
6 A Rand study of the School Preferred Reading program in Los Angeles drew heavily on the instrumentation and design of the Change Agent study and reached similar conclusions. Specifically, it concluded that "the more efficacious the teachers felt, the more their students advanced in reading achievement." See David Armor et al., Analysis of the School Preferred Reading Program in Selected Los Angeles Minority Schools (Santa Monica, Calif.: The Rand Corporation [R-2007-LAUSD], August 1976). This study used, as the dependent variable, the change in individual students' scores on a standardized reading test.
7 See, especially, Appendix A in Peter W. Greenwood, Dale Mann, and Milbrey Wallin McLaughlin, Federal Programs Supporting Educational Change, Vol. III: The Process of Change (Santa Monica, Calif.: The Rand Corporation [R-1589/3-HEW], April 1975).
8 Dan Lortie, School Teacher (Chicago: University of Chicago Press, 1975).