Yet, when the right kind of professional development is offered to teachers for a sufficient time, the connection between teachers' development and higher student achievement is clear.

Quick-fix professional development has been losing favor in current school reform efforts. However, there has been little evidence to determine what does work. That may be changing.

Not only is research developing principles for a transformation of professional development, but in a series of presentations David Cohen of the University of Michigan is laying out evidence useful to policymakers and staff developers. Both at the annual meeting of the American Educational Research Association and at a special forum on Capitol Hill, Cohen described a study of 1,000 elementary teachers in California who participated in professional development to improve their teaching of math.

His analysis of their experience and its effect on student achievement is significant because it confirms that there are ways to make professional development pay off.

Helping elementary teachers with their math skills is particularly difficult, Cohen says, because "they don't know much about math and they learned it like most of us--as arithmetic that is mechanical and without big ideas or concepts behind it." The professional development system available to remedy the lack of teacher knowledge and skills "is a non-system," he says, even though it seems very large and it spends 2-4 percent of a district's budget.

Even though seven of 10 teachers in his study had attended workshops on math in the previous year (1993-94) and two-thirds of those had focused opportunities to learn (the staff development related to instructional reforms), none dealt with the content and context of math. Only two opportunities--Madeline Burns and work with the state-developed replacement units--helped teachers dig deeper into context. Furthermore, only 9 percent of the teachers in the study spent more than one day in any of the workshops.

"Almost all of the teachers knew the leading ideas for reforms in math, and most had added some practices like manipulatives or cooperative learning," Cohen says, "but they added them in small doses in a context that didn't change their drill and practice routines."

The teachers who had more than two days of development in the quality programs reported much more change and less drill and skill stuff, he says. And in schools where there were several of these teachers, "kids had higher scores" on the state test used at the time.

It is possible, Cohen concludes, "to have a coherent relationship among student learning, teacher learning and curriculum change."

Even though such studies show a pay off for in-depth, sustained professional development, it may take a long time to get there. Another research study presented at the forum, sponsored by the Consortium for Policy Research in Education, found little investment by states in teacher professional development. Researcher Thomas Corcoran could find few states that even kept data on professional development at the state level and none monitored what school districts were doing.

One-fourth of the states, according to Corcoran, require districts to set aside time for professional development, and 36 require it to be part of school improvement plans, "but there is no review, feedback, or analysis of the plans."

Only a few states link professional development to student learning standards, and these do so generally because of prodding from the National Science Foundation Systemic State Initiative grants.

"It is not a pretty picture," Corcoran concluded. "Professional development is fragmented, not linked to student standards and tends to be about process and generic teaching skills rather than content."

Deborah Ball, also a presenter at the forum, emphasized the enormous change that must take place in professional development because the teaching force is so "underprepared." Teachers reviewing the results of the Third International Math and Science Study, for example, concluded that the principal aims of math reforms are to use manipulatives, give students problems grounded in real life and organize group learning. Few, she said, "believed the reforms were meant to change the substance of what was taught." Telling or showing teachers about how to change the substance is insufficient, she said. "Teachers need ongoing opportunities to learn interactively with students."