A recent government study on the pretrial activities of criminal defendants throws some cold statistics on the politically popular notion that preventive detention of criminals will cut crime in the streets.

The four-year study, commissioned by the Justice Department's National Institute of Justice, actually recommends releasing more defendants pending trial, saying that the evidence "strongly suggests" that such a policy would not lead to an increase in crime.

"No single solution"--whether preventive detention or speedier trials or others--"is likely by itself to reduce pretrial arrest rates dramatically," the study concluded.

The study, which surveyed some 3,500 defendants in eight jurisdictions--including Washington and Baltimore--shows that 87 percent of those released show up for their court appearances, and that 84 percent of those surveyed were not arrested a second time while out on bond.

D.C. Pretrial Services Agency officials say that locally, about 10 percent of those released are rearrested, but most of the arrests are for writing bad checks or for prostitution, not for felonies.

In addition, more than 90 percent of those released locally show up for court. Those who don't show up are often found to have been in jail in another jurisdiction, in the hospital or sometimes simply at home, having forgotten to appear. Only about one percent actually flee, D.C. officials said.

The study points out that in the District, more than half of those released and later rearrested were picked up more than eight weeks after their original release. Although the study recommends speedier trials as one way to reduce the rearrest rate, veteran court-watchers here say it would be practically impossible to reduce the current waiting period of more than one year from arrest to trial for felonies to only eight weeks.

Another proposed step, according to the study, would be to inform judges who had released defendants that the defendant had been rearrested, thus giving the original judge a chance to revoke the earlier release.

That, however, was the practice in the District until 1979. It was discontinued, sources said, because judges by and large failed to take action, even when they were notified.