Article courtesy of Phi Delta Kappan, May 1982

The Impact of IMPACT II

By Dale Mann, professor and chairman of the Department of Educational Administration, Teachers College. Columbia University, New York. N.Y.

IMPACT II--an experimental teacher to teacher network--aimed to change the nature of classroom innovation in the New York City schools. The program, which cost about 27 cents per child, showed that innovation works best from the bottom up.

In 1979 the Exxon Education Foundation created a teacher to teacher networking program, IMPACT II, to improve schooling in New York City. IMPACT II has since fulfilled that mission to a significant degree;¹ it has also provided a model for other cooperative ventures between private and public agencies aimed at the improvement of classroom instruction.

A former mayor once described New York City as an insoluble problem disguised as a marvelous opportunity. There is truth to this description, at least as far as the Exxon Foundation project was concerned. Some of the obstacles to the improvement of classroom instruction seemed almost overwhelming: an enormous education bureaucracy that seemed likely to smother new ideas, a corps of older teachers who might be reluctant or even unable to learn new methods, a strong union whose rules could stifle innovation, achievement testing programs that might restrain creativity, and tight budgets that seemed almost certain to pinch initiative. The cumulative weight of these obstacles should have crushed any program founded on the simple idea that teachers could help other teachers.

Conventional wisdom holds that schools are unimaginative institutions that can be changed only through projects that outsiders design and impose. IMPACT II sprang from a different view, one that propels the growing practice of "quality circles" in industry: Those who do the work should control the change. Simply put, teachers should direct classroom innovation.

The IMPACT II Strategy

The Exxon Education Foundation made grants to the New York City Board of Education of about $250,000 per year, starting in 1979. From these funds, teachers who sought to refine classroom innovations could receive "developer" grants of $300. Teachers who wished to try out the innovations of others could apply for "replicator" grants of $260. The IMPACT II staff (part of the Division of Curriculum and Instruction) created a variety of activities to build systemwide teacher to teacher networks. The major goal was to establish contacts between developers and potential replicators. About 500 grants have been made since 1979, evenly split between developers and replicators.

The IMPACT II staff organized a series of proposal competitions, designed to alternate waves of potential developers with waves of potential replicators. The staff invested heavily in attractive news letters and catalogues describing the project, in receptions and ceremonies honoring participants, and in personal help to individual developers or replicators. But, except for the rudiments of fiscal accountability, the staff avoided heavy handed management techniques. Grant recipients did not have to  promise that their students' reading scores would zoom upward. There were no contract reviews or evaluation visits. Teachers were presumed to be capable of such managerial activities as dealing with vendors, purchasing services, and scheduling activities. Moreover, there was no predetermined orthodoxy. The ideas that teachers proposed for development under IMPACT II funding had to deal with classroom instruction, but that was the only restriction. Some 250 replicators have copied about a hundred different ideas since IMPACT II began, but which ideas they chose to use (and in what areas) has been left strictly up to them.

The Exxon Foundation and the board of education decided to approach teacher improvement through personal, low key interchanges among teachers and maximum teacher autonomy rather than through contracts and milestones and tightly controlled products. Moreover, the foundation and the board gambled that small, nonrenewable grants would motivate teachers to change their professional practices.

The Impact of IMPACT II

To evaluate the program, we asked 136 grant recipients to complete a questionnaire inquiring into their teaching practices before and after receiving their grants. Forty four teachers who had applied for but did not receive IMPACT II grants also completed this questionnaire. These teachers functioned as a control group for our study. We also completed an extensive case study of a program in another school district, which was similar to IMPACT II in all respects except for not including the Teacher to teacher network. Taken together, these data sources increase the credibility of our findings.

The New York teachers we studied were a representative group. There were more women than men, half were between the ages of 30 and 39, virtually all had considerable formal training, and 75% had been teaching for 10 years or more. Most had tenure and held membership in the teacher union; they could choose what to do in the privacy of their classrooms, when to do it, and how. Nonetheless, 80% said that they were enthusiastic about teaching.

Before their involvement in IMPACT II, most grant recipients had relied primarily on large group instruction; only about 25% of them had regularly used small group or individualized techniques in their classrooms. Prior to IMPACT II, these teachers had also been remarkably isolated. One stated a common lament: "It's kids, kids, kids for six hours and 20 minutes a day." More than half said that they had never visited a classroom outside of their own school buildings prior to their IMPACT II experience.

Of course, the ultimate test of any innovation is whether it changes classroom practice. Did IMPACT II change teachers' instructional strategies? To answer that question, we asked IMPACT II participants and the control group of unfunded applicants whether their use of five instructional strategies small group instruction, individualized instruction, interdisciplinary instruction, independent study projects, and self instruction had increased during the interval covered by IMPACT II. We chose these five strategies because the literature suggests that each helps children to learn more effectively.

Figure 1 shows our findings. Far more IMPACT II participants than unfunded applicants reported that they used each of the instructional strategies more frequently after their involvement in the project. For example, two thirds of the developers and three fourths of the replicators reported that they engaged children more often in small group instruction after participation in IMPACT II than before. Slightly more than one third of the unfunded applicants reported a similar change in their teaching techniques during this interval.

Most classroom teachers are regularly exposed to new teaching ideas and instructional strategies. Principals and central office personnel suggest some innovations; others arise out of in-service training sessions and articles in professional journals. But, for the most part, these exposures to new techniques do not alter teachers' classroom strategies. IMPACT II did so, however. Why?

We believe that the two stage cycle of teacher involvement may be the explanation. The first wave of IMPACT II grants went to teachers who had ideas that they wished to refine. These  developers then conducted workshops, visited classrooms, and provided other kinds of implementation assistance to a second group, the replicators. whose grants supported classroom adoption of mature ideas. When we compared the two groups, we found that the replicators, were younger, were more often female, were more likely to be teaching in the lower grades, and had fewer years of teaching experience. Ninety two percent of the developers but only 64% of the replicators described themselves as "master teachers."  Clearly, the two kinds of IMPACT II grants tapped different strata of  teachers. IMPACT II used one group of teachers, the developers, to reach another group of less accomplished teachers, the replicators.

The program challenged developers to come up with their own ideas, to manage the implementation process, and to take responsibility for convincing other teachers to adopt their ideas. It gave replicators less autonomy and fewer opportunities for active participation. We would thus expect IMPACT II to affect the teaching behaviors of replicators less than the teaching behaviors of developers. But, as Figure I indicates, the opposite was generally the case.

More replicators than developers by margins ranging from 6% to 16% reported that they engaged students more frequently in small group instruction, individualized instruction, independent study projects, and self instruction after participating in IMPACT II than before. On the average, 60% of the replicators reported increased use of these four teaching strategies, compared with 54% of the developers and 31% of the unfunded applicants. The replicators were also more likely than developers or control group members to have tried to convince other teachers to use these instructional strategies. Thus the replicators, although less actively involved in IMPACT II than the developers, had their teaching practices more profoundly affected by the program.

IMPACT II participants also reported attitudinal changes. Twice as many developers and replicators as unfunded applicants said that their attitudes toward teaching had improved during the grant period. Here again, more replicators. (69%) than developers (65%) reported gains in self-esteem.  Moreover, 83% of all IMPACT II participants reported gains in self esteem. 

What explains the remarkable success of this program? IM PACT II grew from the bottom instead of being imposed from the top.

The United Federation of Teachers backed its endorsement of the program with publicity in the union newspaper, statements of support from union officials, and constant cooperation from the building level up. This union support smoothed the way for the individual teachers, who owned and directed the changes in instructional strategies. ²

Carpenters aren't expected to change their work simply because the boss will make more money; merchants don't use new techniques solely because customers might benefit. Progress in the building trades and in private enterprise comes because a change leaves the workers themselves better off. Innovations without benefits do not happen.

IMPACT II offered participants several benefits. One was money, but the $200 to $300 that each received for materials was quickly spent and not a very significant sum, given the average annual teacher salary of $20,000 in New York City in 1979.IMPACT II could not offer participants the prospect of transfer to better schools, quick promotion, or even job security. Instead, grant recipients received an intangible benefit, professional self fulfillment. When asked which facet of the program was most important to them, half of the replicators. and one third of the developers chose "networking," i.e., meeting other teachers. Opportunities to be trained and to train others, visiting other schools and being visited, publishing one's ideas, collegiality, and recognition accounted for most of the teacher self interest harnessed by the program.

Developers of programs aimed at improving classroom instruction often view school faculties as enemy camps that must be infiltrated. But teachers generally win in the contest between their own schools and other people's projects, and changes from such programs prove temporary. To avoid this consequence, IMPACT II focused on individual teachers, not whole schools. The program encouraged almost, but not quite, required individual teachers to recruit other teachers, who regarded the recruiters as credible, reliable sources of help because they, too, spent "six hours and 20 minutes a day with kids, kids, kids." IMPACT II facilitated recruitment by providing substitutes to cover classes for developers (although less than 50% of the total budget went for this purpose).

For teachers who desired prestige and visibility, IMPACT II provided recognition. For those who felt the need to fight isolation, there was the teacher network. For those whose only barrier to better instruction was a little money for extra materials, IMPACT II supplied small grants.

The program even protected the fragile reality of an outstanding teacher in a less than outstanding school. Like prophets, the best teachers are seldom honored where they work. The majority of replicators came to developers from outside the developers' own schools. The IMPACT II program facilitated such contacts through social events and other voluntary activities.

Staff development projects often fail to use what we know about effective education. Learners do best when the tasks are clear and feasible, when the learners are expected to achieve, and when they arc rewarded for those achievements.

IMPACT II has high expectations of its participants. Most of the recipients we studied used their IMPACT II development grants to work on curricular projects of their own choosing, not on projects mandated by administrators. Half wrote new materials. At every step, the program challenged teachers to define and control their own work. The involvement of their building administrators was minimal. The fidelity with which teachers completed the work they had proposed and the positive outcomes of their efforts reflect the participants' professionalism.

The IMPACT II network is a sticky web. Last year's grant recipients are still being asked to help with workshops. No one is ever dropped from the mailing list; unsuccessful applicants continue to receive catalogues, announcements, and invitations to program functions. Meanwhile, IMPACT II teachers are praised, recognized, and otherwise reinforced for their exemplary work.

The landscape of school improvement is littered with the wreckage of projects that never achieved their goals. IMPACT II did not set out to transform the most listless, hostile, or racist schools in the city (although someone should). Instead, it worked with already good teachers who had the potential to become even better. Remember that the replicators were younger, less experienced, and less confident than the developers. But they came out of the IMPACT II program more likely to have changed their basic teaching strategies and more enthusiastic about their new instructional practices than were the developers. The IMPACT II staff patiently ground out better classroom practice, teacher by teacher.

Freud once remarked that "much is gained by having transformed hysterical misery into common unhappiness." IMPACT II does better than that, but in the same modestly remedial spirit. No one asked IMPACT II participants to solve the problem of failing scores on reading tests; in fact, the staff declined to use test scores as evidence of program accomplishments.

Forty cents of each project dollar went to teachers as grants; much of the balance was invested in making the teachers feel good about themselves. Through effective public relations invitations to prestigious places, a glossy newsletter, frequent social interactions IMPACT II participants were drawn into a select reference group, far more likely than the average social circle in a teachers' lounge to support risk taking.

The New York City Board of Education recently demonstrated its commitment to instructional improvement by voting to assume approximately half of the cost of IMPACT II during the 1982 83 school year. Meanwhile, the IMPACT II staff recognizes the fact that each teacher must reinvent each good idea. Central board memos and Bureau of School Lunch dieticians notwithstanding, the melancholy reality is that each teacher must discover "The Problem of Potato Chips" before he or she will implement nutrition education. Thus, at both the individual and the project level, IMPACT II is profoundly (and appropriately) incremental, multifaceted, modest, and persistent.

The demonstrated effectiveness of IMPACT II should provide encouragement to other educators who would rather improve urban schooling than abandon it. The yearly investment for IMPACT II amounts to about 27 cents per child enrolled in the New York City public schools. The program has awarded its small grants directly to 500 teachers but within the larger context of a systematically supported teacher to teacher network.

The program outcomes speak for themselves:

• more of the grant recipients than of the unfunded applicants reported significant changes in their instructional strategies after participation in the program;
• 75% of all IMPACT II participants have tried to disseminate their work;
• more replicators report greater amounts of instructional improvement than do developers;
• several hundred teachers have accepted grants and have carried out their proposals--with some support, but with no monitoring or supervision.

When the public is clamoring for accountability, focusing on process takes courage. The successes of IMPACT II come from trusting teachers, supporting their judgment, honoring their commitment, and then carefully crafting and nurturing a network to facilitate the changes they undertake.

The importance of the network became sharply apparent when we compared IMPACT II outcomes with those of an otherwise similar program that lacked the feature of network support. In that similar program, the money did help teachers and children during the year it was spent and in the classrooms for which it had been targeted. But even when two teachers had ideas for similar projects, there was no exchange. Teachers who received mini-grants never had an opportunity to persuade others to adopt their instructional practices. When the money ran out, the classroom practices it had paid for also ended. The improvements were only temporary.

By contrast, IMPACT II establishes teacher networks, brings together developers and replicators, makes possible released time for classroom visitations, and provides a supportive structure that buoys teachers who are helping others to improve their professional practice. IMPACT II took on the difficult problem of teacher improvement in a large urban school system. It succeeded because teachers who are accorded trust and respect are willing to try new strategies. A network of teachers provides an effective vehicle for peer teaching and learning. For those who would improve public schools by improving classroom teaching, the lesson of IMPACT II echoes something that the Chinese philosopher Mencius said more than 2,000 years ago:

When people are subdued by force they do not submit in heart. They submit because their strength is not adequate to resist. But when they are subdued by virtue, they are pleased in their inner hearts, and they submit sincerely.

1. Leslie Goldman, a doctoral student in the Department of Educational Administration at Teachers College, Columbia University, assisted me in the evaluation of the program.

2. The following discussion uses a theoretical framework that I have developed more fully in Making Change Happen? (New York: Teachers College Press, 1979)