South Carolina places "high-risk" 4-year-olds in preschool programs designed to build self-esteem and skills that will help them in school.

Missouri sends "parent educators" on regular visits to the homes of thousands of families with small children, hoping to encourage healthy child development.

Minnesota has set aside $27 million to provide child care for teen-aged parents who want to return to school.

Programs like these, aimed at helping students "at risk" of failing in school, cover the spectrum -- proactive and reactive, carrot and stick, preschool, in school and post-dropout. The programs share a common goal, but they come in a remarkable variety, reflecting both the extent of the problem and the lack of consensus over strategy.

"There may be as many different mechanisms as there are states," said Maryland School Superintendent David Hornbeck. "There are lots of programs that demonstrably work. The issue is not whether we can be successful, but whether we will make the kind of commitment it takes to be successful."

In the process of making that commitment, educators and policymakers have become engaged in a debate over tactics. The question is standard -- are new programs unnecessary reinventions? Or are the traditional models -- Head Start, tracking by ability, federal programs that pull disadvantaged children from the regular classroom for remedial tutoring -- inappropriate for the new educational target, "child at risk"?

"We have to be careful when we debate the issue we don't throw out the baby with the bath water," said Cynthia Brown, who oversees programs on equity in education for the Council of Chief State School Officers, a group of state superintendents. Her organization is exploring new approaches, but she also argues that some of the old methods must be preserved.

Use of traditional programs has inspired fear that "child at risk" may become a new label for troublesome youngsters who could be shoved aside in programs carrying a stigma and self-fulfilling prophecy of failure.

"Tracking and placement has done more harm than good," said Robert Palaich, a policy analyst for the Education Commission of the States. "The identify, label, pull-out model may be a model of 20 years ago."

Among officials in Dade County, Fla., the answer is clear -- the old programs are not enough.

"We've tried an awful lot of approaches . . . and we really didn't see the results we wanted," said Gerald O. Dreyfuss, an assistant superintendent for the Dade County Schools.

As a result, 32 schools, many in Miami's inner city, are involved in a pilot program testing whether student performance can be affected by giving parents and teachers more control over budget, organization and curriculum, decisions traditionally sent down by central-office administrators.

In some cases, the schools have decided to open on Saturdays. Others have instituted peer evaluations among teachers or redesigned the schedule, doing away with the seven-period day.

Approaches as drastic as restructuring schools are rare; it is much more common for schools to beef up drug, dropout and pregnancy prevention programs. While few school officials would argue against such efforts, many believe they reach the child too late.

"That's a very short-term way of looking at the problem," said Henry M. Levin, professor of education and economics at Stanford University. Levin has designed a program aimed at helping disadvantaged children by setting clear goals, improving nutrition and health and avoiding repetitious teaching.

"The present approach is just downright silly. None of these programs is very successful," Levin said of the secondary-school programs for troubled students. "If we put resources into young and healthy people, we'll be far, far more successful in solving those problems."