African American students in large school systems are arrested far more often on campus than their white peers, new federal data show.

The data, from an Education Department civil rights survey to be released Tuesday, provide the government’s most extensive examination yet of how public schools across the country bring police into the handling of student offenses.

The new figures also show continuing racial disparities in out-of-school suspensions and expulsions, which are far more common in schools than arrests and referrals to law enforcement.

“The sad fact is that minority students across America face much harsher discipline than non-minorities — even within the same school,” Education Secretary Arne Duncan said. Duncan cautioned that the government is “not alleging overt discrimination in some or all of these cases.” But he said educators and community leaders should join forces to address inequities.

The department’s Office for Civil Rights collected data from 72,000 schools across the country for the 2009-10 school year.

Overall, the data showed that 96,000 students were arrested and 242,000 were “referred” to law enforcement by school leaders, meaning the students were not necessarily arrested or cited.

In a more focused analysis of school systems with more than 50,000 students enrolled, the data showed that African American students represented 24 percent of enrollment but 35 percent of arrests. White students accounted for 31 percent of enrollment and 21 percent of arrests. For Hispanic students, there was less of a disparity in arrests. They accounted for 34 percent of enrollment and 37 percent of arrests.

Such data about student contact with police had not been collected before on such a large scale.

Police action on campus has become a growing concern as law-enforcement presence has increased markedly in the past two decades, especially in the aftermath of the massacre at Columbine High School in 1999, and a zero-tolerance culture has taken hold in many schools.

“It’s an issue that’s gotten more and more attention, and it’s good that we finally have numbers,” said Russell Skiba, an Indiana University professor who has done research on equality in education. “Where there are clear disparities, that should be of concern to us as a nation.”

On arrests, no data were collected about the types of offenses involved. Critics say school arrests are too often made for adolescent misjudgments — such as insubordination, disrespect, class disruption and fighting — that, in previous generations, were handled with calls to parents and visits to principals’ offices.

“There are concerns that a lot of kids are winding up in custody or juvenile detention that probably shouldn’t be there,” said Daniel J. Losen, director of the Center for Civil Rights Remedies at the University of California at Los Angeles, author of a recent report on discipline for the National Education Policy Center at the University of Colorado.

Duncan himself has told stories about how, when he was schools chief in Chicago, he was stunned to learn that “the vast majority” of arrests of young people could be traced back to schools.

The stakes for students are high, said Matt Cregor of the NAACP Legal Defense and Educational Fund.

“The harms of suspension pale in comparison to the harms of arrest,” he said. “A first-time arrest doubles the chances a student will drop out. A first-time court appearance quadruples them.”

Cregor noted the example of a judge from Clayton County, Ga., Steve Teske, who led a community effort to curtail the surge in cases sent from schools to juvenile court.

Beyond police contact, the data show persisting disparities in out-of-school suspension. African Americans were more than 3½ times as likely to be suspended or expelled as white students, the data showed.

Black males stood out, with 20 percent being suspended from school during the 2009-10 school year. By comparison, 7 percent of white males, 9 percent of Hispanic males and 3 percent of Asian American males were removed from school for disciplinary offenses.

Students with disabilities were more than twice as likely to be suspended as students without disabilities.

Racial disparities in suspensions have been tracked by researchers for years. Experts say there are no studies to show that differences in behavior cause the gap between blacks and whites. Exactly why the gap exists is unclear.

Poverty is an important factor that affects rates of school suspension, but when researchers account for these and other factors, disparities by race still exist.

Many researchers say that unconscious bias is likely to be a factor, as is unequal access to highly effective teachers who do better at managing behavior and engaging students. The culture and leadership of a school are also important. But more research is needed, many agree.