A r c h i v e d  I n f o r m a t i o n

Looking at patterns in the relationships between supporting conditions and reform outcomes, the data as a whole provide considerable support for the basic hypothesis of the study. Generally speaking, schools that have implemented greater amounts of innovative practices also tend to have a greater number of the support factors in place. In particular, of the nine schools that implemented significant changes in two or more categories, eight of them also had in place high levels of at least four of the supporting conditions. Furthermore, of the eight schools that exhibited relatively limited amounts of reform, i.e., in one category at best, seven were characterized as having a high level of three or fewer supporting conditions (and in fact, six had zero or one). Thus, it certainly appears that extensive innovation regarding curriculum and instruction only takes place when a majority of these key supporting conditions are high. When they are lower, the schools in our sample have been much more constrained in their ability to generate meaningful reforms.

	School
	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
Power	*		*	*	*	*	*		*					*
Knowledge	*	*		*				*	*	*
Information	*	*	*		*	*			*					*
Rewards	*	*		*	*	*				*				*
Instructional Guidance	*	*	*		*	*	*		*					*
Leadership	*	*		*	*	*	*							*
Resources	*	*	*				*			*	*	*	*		*
Teaching for Understanding	*	*	*	*	*	*	*	*	*
Use of Technology	*	*	*	*	*		*			*	*	*
Educating All Students	*	*	*	*	*	*		*	*				*
Integrated Approaches	*	*	*	*		*	*	*

Another cut at the data serves to examine the potential importance of the individual support mechanisms, and also provides further evidence of the extent to which they serve as an interconnected set of factors supporting innovation. Table 3 indicates the frequency with which high levels of the supporting conditions are found in the high and low innovator schools. Considering first the nine schools with high levels of change in two or more outcome categories, two support variables scored high in seven of them, namely, power and instructional guidance. Clearly, if school-based management is to serve as a means through which schools will introduce important changes in how they operate, they must be given enough power to implement the kinds of changes seen as appropriate for their student community. Equally important, however, is the focus and impetus regarding the reform process that schools derive from a shared understanding -- rooted in curriculum frameworks, learning objectives, school visions or philosophies, etc. -- of the instructional direction of the school. Without such a common basis for reform, innovations are disjointed at best or not even attempted at worst.

Nine High Innovator Schools	Eight Low Innovator Schools
Power - 7	Power - 1
Knowledge - 5	Knowledge - 1
Information - 6	Information - 1
Rewards - 5	Rewards - 2
Instructional Guidance - 7	Instructional Guidance - 1
Leadership - 6	Leadership - 1
Resources - 4	Resources - 5

Information and leadership scored high in six of the schools with two or more categories of significant reform. These two variables, along with instructional guidance, seem to be key factors facilitating a high level of reform after power has been decentralized. Undoubtedly, these three elements are to some extent interrelated. For example, an important task of leadership is to share information broadly with the school's constituents; information regarding the school's goals and performance is necessary to develop a shared understanding of its instructional direction; and this common vision enables a broader range of individuals to assume leadership roles in the reform process. Hence, while sufficient power may provide schools with the autonomy needed to make desired changes, these additional three elements may constitute the foundation on which school participants can successfully base their efforts to implement these reforms. Without these supporting conditions in place, innovation is likely to be much more difficult to come by. This pattern is clear in our data, as the scores for all of these variables were low in all but one (School 14) of the low innovator schools (see Table 2).

High levels of knowledge and rewards were found in five of the highly innovative schools, in contrast to one and two, respectively, of the lower innovators. Interestingly, knowledge was the only high supporting condition in one of the schools that was a high innovator in three areas. As for rewards, the seventeen schools in our sample were rather limited regarding the extent to which performance was rewarded either formally or informally and the presence of viable accountability mechanisms. Even the schools that were coded high on the reward variable had usually not made great strides in developing a reward system that effectively based consequences on performance at either the individual or the school level. On one hand, the fact that even the use of relatively limited reward approaches shows up in five of the high innovators and only two of the lower innovators suggests that, if applied more extensively, rewards might motivate even more innovation. On the other hand, some schools appear to be willing and able to move forward with their reforms in the absence of a reward system tied to these efforts. To some extent, this may be because educators are by and large used to working without extrinsic rewards. However, many participants in the study indicated that the opportunity to be involved in school decision making and influence the direction of the school improvement process serves as its own reward. Thus, the development of a high-involvement system can intrinsically motivate people to invest the time and energy needed to improve curriculum and instruction, and it might also help leverage the effects of even low levels of extrinsic rewards. These interactive effects provide further evidence of the interconnectedness of these supporting conditions and their ability to reinforce each other in the reform process.

Finally, over half of the schools that were most successful in implementing curriculum and instructional changes did not have high levels of additional resources (i.e., beyond their basic budget allocation and standard entitlement moneys). Thus, it appears that such resources are not a key requirement for meaningful reform to take place. The importance of resources can be further assessed by looking at the pattern of findings among the eight schools that implemented fewer curriculum and instructional changes. Interestingly, five of these schools were coded as having high levels of resources, suggesting that the acquisition of external funding or other sources of material support does not by itself spur reform activity. Perhaps more important to reform than extra resources is organizational effectiveness in applying existing resources. This is undoubtedly enhanced by high levels of some of the other six conditions. Thus, in the four high innovator schools with a high level of resources, high levels of at least three other variables were also present (see Table 2). In contrast, in four of the five low innovator schools with a high level of resources, this was the only supporting condition present, with the fifth school demonstrating a high level of only two other variables. This pattern reinforces the conclusion that the other variables provide valuable support for the reform process, enabling the schools to benefit more from the additional resources they have acquired.

To summarize, all of these conditions, with the possible exception of resources, appear to be instrumental in facilitating innovations, as each was found relatively frequently among schools that had implemented many changes and not very often in schools where changes had been more limited. This makes it difficult to separate out any one or two factors as being the most critical, and supports the notion that these conditions constitute an interrelated set of system design features that support and reinforce each other in the process of introducing new and innovative practices. This is consistent with findings that organizational designs where employees are involved in performance improvement consist of a system of mutually reinforcing features (e.g., Lawler et al., 1992).

In contrast, even the hard work invested in generating changes in curriculum and instruction is likely to be limited in its impact in the absence of adequate systemic support. Among the schools in our sample, it was not uncommon for many individual teachers to have identified new practices and initiated use of them in their classrooms. However, those leading the reform process -- usually administrators and/or a governance council -- often failed to put into place mechanisms that would, for example, enable teachers to learn from each other, provide incentives for teachers to adopt new practices, or insure that individual innovations were part of a broader, well-defined strategy regarding the instructional direction of the school. Without a system designed to focus the reform effort and facilitate the diffusion of innovations school-wide, new practices frequently remained the sole purview of the teachers who had introduced them.

The data also suggest that a school does not necessarily require all of these conditions to be in place in order to be highly innovative. In fact, it appears that the potential for generating meaningful reform is quite high as long as the school is characterized by high levels of four or more of these elements. In any event, merely decentralizing power from the district to the school is not sufficient to insure that school-level reform will take place. This is the basic premise of the high-involvement model (Lawler, 1986) on which our research has been based, and our findings add further support to the validity of this model. Of course, we do not intend to imply that the four elements of that model, along with the three additional support mechanisms we examined, constitute an exhaustive set of system features that might be relevant to support school reform. However, since our data suggest that these seven can play an important role in successful implementation of significant reforms, it is worth considering each of them in greater detail, based on the experiences of the exemplary schools comprising our sample. In the discussion below, we focus particularly on the role the separate dimensions of each variable play in explaining how they support the innovation process.

Power. Two dimensions underlying the power variable are the extent to which decision making authority is decentralized to the school level and the extent to which a broad range of school-level constituents -- administrators, teachers, parents, community members, and students -- are in turn empowered for meaningful involvement in the decision making process. Both of these dimensions are important in differentiating high from low innovator schools. Regarding the amount of authority decentralized, the schools in our sample typically²⁸ faced some significant constraints in terms of their authority regarding personnel and budget decisions. In the area of curriculum and instruction, the high innovator schools felt considerable power and responsibility for determining how to deliver the curriculum. However, they also tended to be operating in the context of district or state curriculum guidance, indicating that some overall direction from the larger system was helpful in stimulating change.

As for the empowerment of school constituents, high innovator schools tended to have much higher levels of constituent involvement, in part by utilizing a variety of different decision making. Many of these groups were designed to facilitate interaction across the typical internal boundaries of departments and grade levels. Typical structures included governance council subcommittees that were open to membership by interested teachers or parents, and teacher teams that were actively included in the consensus-building process for school decisions. High school departments often played this role, but schools whose governance councils included staff members serving as representatives of their departments were usually less effective at promoting reform than councils using an alternative structure. Given a tendency for departments to "protect their turf" and thus resist changes they believe will impact their domain, governance structures that cut across traditional departmental boundaries are more supportive of the reform process. The most effective governance councils were those that served largely to coordinate and integrate the activities of the various decision making groups operating throughout the school. These councils provided the direction for the changes taking place and allocated resources to support them, focusing on the needs of the school as a whole rather than on the needs of individual academic departments or teaching teams. But the design and implementation of these changes resulted from the widespread involvement of most teachers and a number of active parents on multiple teams and/or subcommittees.

Knowledge. All of the schools in our sample used professional development activities to enhance staff knowledge and skills. Furthermore, high and low innovator schools did not consistently differ in terms of two dimensions of this variable, namely, the range of content areas covered in development opportunities or the extent to which professional development was strategically tied to the school's reform objectives. However, a greater proportion of the staff took part in development opportunities at high innovator schools. For example, training regarding decision making skills often was not limited to members of the school's governance council. Instead, many of these schools assessed their development needs and then planned and coordinated development activities to meet these needs through inclusion of a wide range of staff and sometimes even parents. In addition, staff at these schools participated in development opportunities on a more regular, ongoing basis, rather than only sporadically and infrequently (e.g., when school-based management was initiated). In key ways, then, the highly innovative schools focused their efforts to enhance staff knowledge and skills so as to support the initiation and implementation of reform activities.

Information. The high innovator schools typically did a better job of sharing school-related information among a broad range of constituents. This was true for both the internal and external dimensions, and oftentimes resulted from using more communication mechanisms. The low innovators often relied primarily on the distribution of council meeting minutes and school newsletters to disseminate information. Yet constituents at these schools sometimes indicated that they were not well-informed because they did not receive or read these documents, suggesting that simply distributing information is not sufficient. The high innovator schools, in contrast, took a more proactive approach to making sure that information was transmitted, relying as much on face-to-face means as on formal documents. The existence of multiple decision making groups at these schools facilitated this process, as these groups provided a useful conduit through which information could get out to all staff members. In addition, more kinds of information were regularly disseminated in high innovator schools, including information about innovation in other schools and about school performance. Finally, there was also more extensive solicitation of external input at the high innovator schools. The principals in these schools usually attended many different types of meetings at which external constituents were present, to discuss school activities and get their feedback.

Rewards. The high and low innovator schools differed to some extent on the reward variable, and by and large this difference appears to be tied to a single dimension. The evaluation of these schools tended to be based on performance more than at the low innovators. The same was not true for evaluation of teachers and administrators. As indicated earlier, reward systems that effectively tied consequences to performance were infrequently utilized in the schools in our sample, and this seems to be true for most schools, at least those in the U.S. Finally, no differences were apparent in the two groups of schools regarding the extent to which informal rewards were utilized. All in all, more effective use of evaluation and reward systems is a key area in which schools could improve.

Instructional Guidance System. Most of the schools in our sample operated according to a set of curricular guidelines developed at the district, state, and/or national (e.g., National Council of Teachers of Mathematics standards) level. Yet they had considerable leeway regarding the specifics of the curriculum they provided to their students and regarding the instructional approaches and materials they used. Many said that the frameworks specified the "what" of the curriculum but that the "how" was up to them. Most of the schools also indicated there was some form of school vision or other locally-developed statement that outlined the school's values and/or objectives. However, for the low innovator schools, it appeared that this vision was little more than a document that had been written and then filed away, as it did not have much influence on the activities of staff members. It often did not embody a shared instructional philosophy. These schools tended not to exhibit much shared understanding of or agreement with the instructional direction of the school.

The scenario was quite different in the schools with many innovations, where constituents were more familiar with the vision and saw it as an important and meaningful articulation of what the school was all about. For some of these schools, development of the vision was based on input from all constituents and emerged through a consensus-building process that naturally led to better understanding of the vision and commitment to it. Some schools had created some form of improvement plan that outlined the instructional direction for the school, which then served as a focus for the reform activities they initiated. In some cases, consensus regarding the instructional direction was achieved simply through frequent interaction regarding curriculum and instruction issues.

Leadership. The leadership variable was composed of three dimensions, and all of them help to differentiate the highly innovative schools from those with less innovation. The dimension which demonstrated the greatest difference between these two groups was the extent to which the principal facilitated participation in decision making. Principals at the high innovators were much more likely to promote widespread involvement and share information broadly. Second, the principals at these schools tended to take on more of a facilitator role. In addition to making sure all constituents were involved, they were often seen as very supportive of staff, were more readily described as a manager than as a leader, and focused on managing the decision making process. As a result, they had to effectively manage the relationships between individuals and groups, mediating among divergent interests and helping to resolve conflicts. In playing this role, they also had to strike a balance between, on one hand, exerting their opinions and desires too strongly and thus dominating school decisions, and on the other, being too uninvolved in the content of the decision. Principals at low innovator schools often erred on one side or the other, thus being perceived as either too autocratic or too laissez faire.

Finally, principals at the high innovator schools were more likely to develop and share leadership among a broad range of individuals throughout the school. While some principals in our sample took the lead in introducing ideas about new practices, in many cases teachers were the primary source of these new ideas. Teachers at schools on the leading edge of reform were more empowered to also take the lead in getting these innovations diffused throughout the school. As more and more staff members contributed to this process, a key role of the principal was to build a system that would support and coordinate the ongoing change process. Unfortunately, the use of these new practices tended to be limited at the low innovation schools to the teacher(s) who had introduced them into the school.

Resources. As suggested above, the acquisition of high levels of additional resources at the school does not appear to be necessary to implement meaningful reform, nor does it seem to ensure that such reform will take place. This was true of both kinds of resources included in this variable, namely, outside funding and partnerships with the community. Of course, this is not to say that such resources are not beneficial, and it makes sense for schools to be proactive in their efforts to acquire external funding and to develop relationships with the business community that will pay off in terms of resources available to the school. However, additional resources appear to have the most impact if applied in the context of a clear instructional direction. For example, the schools that best used their resources to facilitate the reform process were those that maximized the benefits by targeting their use to projects that were directly related to the school vision and reform objectives. Also, the process of change was facilitated when schools acquired and/or applied resources to enhance the process of staff development in areas that were most critical to support the school's reform effort.

²⁸ Throughout this discussion, the generalizations we make do not necessarily apply to all schools in the sample. However, they do reflect the general trends we observed among these schools.
-###-

[Part 4: Methodology]

[Part 6: Conclusion]