The conference participants agreed that psychology departments too often view education as a fringe interest not equal to other specialties, such as child development or cognitive psychology. That attitude has taken its toll on the nation's schools. Seymour B. Sarason, PhD, of Yale University suggested that psychology's lack of involvement has played a role in education's deterioration. While the Supreme Court's desegregation decision was the first to cite social scientific data, psychological knowledge was being either ignored or egregiously misinterpreted by policymakers by the time Head Start came along. Today, American education is minimally informed and rarely takes seriously what psychologists have learned about learning. But when psychology does enter the educational arena, through such documents as the Learner-Centered Psychological Principles (APA, 1995), the educational community is keenly interested.

Discussing the gap between psychological research and practice, conference participants expressed concern that there are too few translators of research into practice. Dr. Sarason reminded the participants of John Dewey's concept of 'middlemen' (Dewey, 1899) who can mediate between psychological research and practice. When Dewey first used the term in 1899, psychology's leaders could be comfortably seated around a boardroom table, and at least two of them - Lightner Whitmer and John Dewey himself - could be seen as such linkers. Today, said Sarason, the number of psychologists who serve such functions is extremely small. Arguing that the future of the country will be determined by what happens in its cities, he concluded by challenging the field to live up to its self-proclaimed mission of improving human welfare.

Participants then began the process of answering the question of what is worth scaling up. Samuel C. Stringfield, PhD, of the Center for Social Organization of Schools at Johns Hopkins University, discussed the differences among ideas, inventions, and innovations. According to Peter M. Senge's book The Fifth Discipline: The Art and Practice of the Learning Organization (Senge, 1990), ideas are what come first. Inventions are ideas that have been proven to work in the lab. Innovations are inventions that can be reliably implemented in diverse circumstances at reasonable cost. Educational reform is stuck somewhere between the idea and invention stages, and to talk about educational innovation is to claim something that hasn't been achieved yet.

Participants debated whether or not American education as a whole is more effective now than it has ever been and whether or not powerful economic and political factors are fueling the move for even better schools. Among the factors cited for schools' failure to improve were political pressures on curricula and a lack of plausible outcome measures. Dr. Stringfield suggested that educational reform may not yet be at the innovation stage, although he noted that several conclusions could be drawn from efforts that have been successfully undertaken:

• Whole school change works better than programs targeted to subgroups within schools,

• Starting earlier in children's lives is a better investment than starting late, and

• Adapting national designs is generally more cost-effective than 'home-brewing' your own design.

Participants agreed that the reform effort needs more research on promising programs to see if they really do work and if so, under what circumstances. Consumer guides summarizing the research literature and evaluation results on various efforts would also be helpful.

Abstract research on promising practices isn't enough, participants agreed. Robert D. Felner, PhD, of the University of Rhode Island, cited two concrete models of successful educational reform: land grant universities and the automotive industry's retraining of American workers between 1975 and 1985. These examples suggest that scaling up does not mean 'tinkering around the edges.'

Several criteria already exist for deciding what's worth scaling up. Dr. Felner cited three:

• The program should be effective in real-life settings, meaning it's no longer acceptable to say things like, 'It's a great program, but it doesn't work in schools.' Researchers need to define a problem and then measure whether or not it was solved.

• The program should be viable and sustainable. As programs sweep in and out of schools, teachers left holding the bag become disillusioned.

• The program should be ecologically congruent with the setting. Instead of popping programs into schools, innovators should respect the resource limitations, values, and other conditions in schools.

Many successful programs already exist. Dr. Felner noted that the Carnegie Corporation's Turning Points project shares several key elements with other exemplary programs:

• Empowering teachers and administrators,

• Re-engaging families in their children's educations,

• Connecting schools and communities,

• Improving academic performance by fostering students' health and fitness,

• Teaching a core curriculum,

• Staffing schools with experts, and

• Creating small communities for learning.

Another issue participants discussed is the way in which educators and psychologists view children's potential. Rhona S. Weinstein, PhD, of the University of California, Berkeley, used a child's statement-'If I were the teacher, everyone would win and nobody would lose'-as the jumping off point for a discussion about the need for a paradigm shift in American education. Our measurement assumptions currently relegate half of all school children to the status of 'below average.' Participants discussed how this view limits children's growth and opportunities.

Dr. Weinstein and other participants presented newer perspectives on intellectual ability. Because society cannot afford to waste a single drop of talent, educators must replace their idea of intelligence as fixed and measurable by test scores alone and instead see it as malleable and multidimensional. To enable this shift, structures in the school need to change, too. Schools could be restructured to get rid of stratified classrooms that cut children out of interesting opportunities in the hope of remediating them. Participants acknowledged psychologists' role in the testing movement and said that psychologists also need to work to change public perceptions. Most people look at schools' test scores without thinking about differences in resources or children's prior opportunities.

While there was some consensus about key elements in school change, there was less agreement about the conditions that need to exist in order for district-wide reform to occur, said James Connell, PhD, of the Institute for Research and Reform in Education. Nonetheless, he brought up several conditions for the group to discuss:

• A collective sense of urgency among adults in the school system, driven by both data and fear;

• A sense of possibility and recognition that while the problem is serious, the resources, leadership, and models necessary to solve it are available;

• A feeling of equity in terms of both pain and gain; and

• A feeling that change is inevitable.

According to Dr. Connell, a plausible, testable, doable theory of change must precede evaluation. And the change process must include the following elements: building awareness, designing an evaluation, implementing data collection, analyzing and interpreting the data, and adjusting the theory. Identifying critical stakeholders and building consensus among them are also key.

Evaluation is a key component of the scaling up process, and the key to designing a useful evaluation is accountability. Abraham Wandersman, PhD, of the University of South Carolina stimulated discussion by offering eight accountability questions:

• Why do you need a new program?

• Why do you want to use this particular type of program?

• How will this new program fit in with the programs you already offer?

• How will you carry out the program?

• How well did you carry out the program?

• How well did the program work?

• What can you do to improve the program the next time around?

• If the program is worth keeping, what can you do to institutionalize it?

Answering these questions is a way to bridge the gap between research and practice. An evaluator can act as a linker who mediates among the needs of the program's funder, the practitioner or service provider, and the children and community.

Participants discussed various models of evaluation, including an 'empowerment evaluation' that represents a more collaborative model than is traditionally used. Participants agreed that few graduate students are trained in program evaluation and that psychology needs more theory and research on evaluation as a process in itself.

Once the quality of programs and processes is known, scaling up becomes essentially a problem of implementing promising and/or exemplary programs as well as helping consumers make wise choices of programs for their setting. William Modzeleski of the Safe and Drug-Free Schools Programs at the U.S. Department of Education provided a context for a discussion of the issue. Noting that the federal government distributes funding for the Safe and Drug-Free Schools Programs with few restrictions and that bureaucracies have built up around programs, Mr. Modzeleski described the results:

• Programs that suffer from poor design,

• Programs that don't work yet receive funding anyway,

• Programs that don't reflect sound science or include evaluation components,

• Programs that don't have a clear purpose,

• Programs that are not implemented the way they're supposed to be,

• Programs that are not targeting the schools with the greatest need,

• Programs that are implemented by teachers with no training in the subject,

• Programs that aren't linked to other school or community efforts, and

• Programs that are too narrowly focused on one small aspect of human behavior.

Pointing to the 10 programs listed in the National Institute on Drug Abuse's Preventing Drug Use Among Children and Adolescents: A Research-Based Guide as exemplary, Mr. Modzeleski noted several reasons there aren't more programs like them. Obstacles to change noted by the participants included inadequate knowledge about education research and the change process at all levels of schools and government, the absence of training and technical assistance, entrenched bureaucracies that resist change, and a lack of models showing what works. The Drug Abuse Resistance Education (DARE) program is popular, said Mr. Modzeleski, because law enforcement personnel offer to do the work of program implementation for harried teachers who lack training in the subject.

Closer and more collaborative relationships between researchers and practitioners at the very beginning of the program development process would make it easier to disseminate knowledge effectively. The reform effort requires adequate resources, additional research and evaluation, and better targeting so that dollars go where they are needed most. At the level of individual schools, reformers need to know what the problem is before they launch a program, set measurable goals and conduct periodic evaluations, and base programs on sound science and research.

Knowledge dissemination is often built on professional development designed to enhance competence in new skills. Deanna Burney, PhD, of Harvard University and the Learning Development Research Center at the University of Pittsburgh brought the perspective of someone who has been an educator and administrator in urban schools for 24 years.

The characteristics of good professional development include concrete application of ideas, opportunities for critique and reflection, support and collaboration, and feedback from skilled practitioners. What is not so well known is how to organize professional development so that it reaches into classrooms and improves student achievement.

Citing efforts in New York City's District Two as an example of effective professional development, Dr. Burney described several of its characteristics:

• System wide improvement that sees professional development as a central function rather than a discrete effort;

• A goal of getting knowledge into the hands of every teacher;

• A culture of shared values about instructional improvement;

• A shared focus on an issue, such as literacy;

• The commitment of time and resources to professional development;

• A focus on instructing children rather than dealing with disruptions;

• The use of a variety of activities to keep teachers engaged in the process of change;

• An emphasis on collaboration, using teachers to coach each other and encouraging honest conversations with principals;

• The recognition that change can't happen instantaneously and that teachers can't change their practices all at once;

• An unapologetic exercise of control over the selection of teachers and principals;

• A clear, consistent message about accountability; and

• The use of outside expertise.

Teachers want more than money. They want to be treated like professionals. District Two's program - in which everyone is a learner in a nested learning community - provides a model of possibility.

In the spirited discussion that followed, participants applauded Dr. Burney's case study approach and agreed that more research needs to be done to understand how to improve teachers' performance. They agreed that before reformers can scale programs up, they need to scale down and look at the interaction between teachers and children in classrooms. And they agreed that changing structures, such as school management, does not necessarily change classroom practices. Adult learning and interpersonal skills lie at the heart of many school reform efforts, which often flounder because of human factors.

The conference also addressed critical issues related to diversity in urban school reform. A. Wade Boykin, PhD, of Howard University's Center for Research on the Education of Students Placed at Risk, reminded participants of public education's historical roots. Mass public education came into being in response to widespread concerns about mainstreaming southern European immigrants. Schooling is never done on culture-neutral terrain, and schools aren't really about reading, writing, and arithmetic. The schools' 'hidden curriculum' inculcates students with the 'Anglo' values of materialism, individualism, and competition. According to Dr. Boykin, school reform efforts can't move forward without acknowledging the hidden role that culture plays in schools and addressing the issue of diversity headlong.

A related phenomenon is the 'deficit thinking mindset' that characterizes public education. Researchers are often preoccupied with what children don't know and can't do, Dr. Boykin said, noting that schools serve sorting and pacifying functions. They should focus instead on the positive aspects of children's lives and delve more deeply into why they don't know or can't do certain things. Perhaps a child is simply unwilling to do something. What is needed, said Dr. Boykin, is a context basis for performance evaluation.

Reformers also need to look at issues related to cognition, said Dr. Boykin, noting that assumptions about the nature of learners have gone unchallenged for decades. Learning is a socially constructed process, and moving beyond rote learning to higher-order skills may challenge existing ideologies. Reformers need to think about what education is preparing children for, because shifts in the American economy mean that students must be better prepared than ever before. Dr. Boykin argued that all children - not just poor, urban children - are at risk.

In the discussion that followed, participants discussed different types of reform. Dr. Boykin and others championed a style called 'co-construction' in which teachers become part of the change process. Anthony W. Jackson, PhD, of the Carnegie Corporation emphasized the importance of holding teachers and others accountable. In one Texas school district, for instance, data were disaggregated by race and economic class. Principals were evaluated based on how well those children did. 'Suddenly,' said Dr. Jackson, 'people knew how to teach African-American kids.'