By Jay Mathews
Washington Post Staff Writer
Sunday, December 19, 2010; 7:09 PM
There was something strange in The Washington Post a week ago. A chart on page A16, using data provided by the D.C. public school system, showed that in late summer and fall 2009, Spingarn High School had by far the lowest number of assaults, thefts, threats and other crimes. There were just six incidents in four months compared with an average of 31 in the other eight high schools assessed.
At that time, teachers at this allegedly safest of all regular D.C. high schools were reporting a rash of crimes and classroom intrusions. The situation became so intolerable that by January they had persuaded D.C. Schools Chancellor Michelle A. Rhee to replace the Spingarn principal.
How could the incidents being reported by security guards under school district rules be so different from what people at the school were experiencing? Why did Rhee ignore the data in changing the school's leadership and her successor, Kaya Henderson, use data from the same source last week to replace the principal at Dunbar High?
I asked D.C. school officials those questions several times last week. They declined to answer. There is no question that Dunbar, which recorded a system-high 46 incidents this summer and fall, had much disruption, including the arrest of six students on rape charges (which were later dropped). Its high total reflects what teachers and parents were telling school headquarters.
But the likely phony count from Spingarn is too typical of security accounting in the District and many other school systems. Educators agree that not much learning can take place when students don't feel safe. Many urban schools suffer from absenteeism, tardiness and disorder, which rob their students of a chance to concentrate.
School safety figures have been considered notoriously unreliable for decades. "Generally speaking, the tendency is to downplay incidents," said Mel Riddile, associate director for high school services at the National Association of Secondary School Principals. "Some principals pressure school resource officers to downgrade some incidents."
More than eight in 10 school-based police officers in one survey said school crime was underreported. The federal rules requiring states to report all "persistently dangerous" schools were so inadequate that in 2004 only three states - Pennsylvania, New Jersey and South Dakota - admitted to having any.
School crime reports can be distorted for many reasons. D.C. teacher Anthony Priest, who kept a diary of disruption at Spingarn, said: "I had one special-ed student that assaulted me several times only to be placed back in my room the next day. . . . The determination was that this behavior was due to his illness, which was bipolar, thus he couldn't be suspended for it."
A Maryland high school teacher said administrators often think, "If the kid is special ed, it is exceedingly difficult to expel, so why even charge?" That leads to deceptively low incident counts and false confidence in the school's climate. The numbers are also distorted by state rules designed to make sure there are not too many schools listed as persistently dangerous. One experienced Northern Virginia school official said Virginia often records only incidents serious enough to lead to at least a suspension.
D.C. officials say security guards generally report security incidents. They are often young and malleable, willing to do whatever school officials tell them. Teachers can also make reports, but the paperwork is time-consuming and the likelihood of serious action small. Some experts said the District once had a more accurate system, but it was dismantled to make sure the city did not look too bad on the federal persistently-dangerous schools list.
The solution, most of the experts I spoke to said, was regular surveys of school climate that depend heavily on interviewing teachers. "What teachers say is a much better indicator of school climate than incident reports," Riddile said.
Neither the District nor any other local school system does regular surveys of teachers on school safety. They might start trying that, because all efforts to improve achievement will fail if the school climate won't support a feeling that this is a place to learn.