We need to get as serious about improving the schools as we have been about safety in the streets.

Over the past two decades, crime dropped significantly overall in New York City, and experts and officials give much of the credit to CompStat, the NYPD's method of tracking, reporting and reacting to crime, and thus preventing it. The system was introduced in NYC in the 1990s to combat a crime epidemic. The CompStat approach includes weekly reports of criminal activity in every precinct and weekly grillings of commanders about how they respond. Since the program was created, homicides have gone from around 2,000 a year from 1988-1993 to 328 in 2014.

Mayor Bill de Blasio, faced with several schools whose failure to educate students might itself be considered a crime, says his administration would create a version of CompStat for the city's 94 worst schools. "We're going to hold every one of the principals to the same kind of standards that our precinct commanders are held to via CompStat," he said. "They'll get the same kind of forceful questioning, and they'll also get the support to succeed."

It sounds promising in principle -- and for far more than just the worst schools. Where the atmosphere is violent or attendance is poor or too many students fail classes, there's not a moment to lose. Eventually, it should probably happen in every school.

But de Blasio and Schools Chancellor Carmen Fariña have been targeting these 94 worst schools schools for improvement since November, promising extra instruction for students, training for teachers and support for administrators. But principals at many of those schools say no help has arrived. The mayor and chancellor also have rejected the idea that the staff of such schools should be immediately replaced.

A data-driven approach to solving school problems is wise, but getting resources to the schools is necessary, too. So is an openness to real changes in staffing, philosophy and methods. Otherwise, CompStat in the schools is just a neat-sounding gimmick.