Four years after agreeing to submit to federal court monitoring of its controversial stop-and-frisk program, the Philadelphia Police Department has made little progress toward curbing unwarranted stops that disproportionately target minorities, according to an analysis filed Tuesday by a group of civil rights lawyers.

The report found that 37 percent of the more than 200,000 pedestrian stops made by police in 2014 were done without reasonable suspicion of criminal activity - down from 47 percent from in 2012.

And although blacks and Hispanics make up 54 percent of the city's population, more than 80 percent of all stops involved minority targets, the report stated.

City officials, who conducted their own analysis of the same data, found a similarly high number of unwarranted stops, the report stated.

However, the officials argue that those totals do not reflect a problem with the program. Instead, they say, the analysis is skewed by police reports that do not fully explain the justifications officers had for making the stops.

That answer is not good enough, countered Mary Catherine Roper, a lawyer with the state chapter of the American Civil Liberties Union, who helped prepare Tuesday's analysis along with the Philadelphia civil rights law firm Kairys, Rudovsky, Messing & Feinberg.

If the numbers do not show significant improvement within the next year, they said, they will seek further legal action against the city.

"Four years later, the city has not done what it said it would do, which is simply to follow the law," Roper said. "What they haven't done - as far as we know - is implemented any system of discipline and punishment for officers who are making bad stops."

The group's findings were filed as part of the fifth periodic status update to U.S. District Judge Stewart Dalzell since the Nutter administration agreed in 2011 to settle a lawsuit accusing the police of using racial profiling to determine whom to stop.

A police spokesman declined to be interviewed about the findings, saying the city's law department would respond soon through filings with the court.

"The city as a signatory to the consent decree strives to cooperate fully with the plaintiff and the court in this matter," said Nutter's spokesman, Mark McDonald.

Nutter made the stop-and-frisk program a key part of his 2007 mayoral campaign, promising it would bring down gun crime, and he remains a staunch supporter of the theory behind it.

Police statistics show the number of gun robberies and assaults with firearms have fallen sharply since he took office in 2008, down 27 percent for gun robberies and nearly 23 percent for assaults. Police have also reported a 25 percent drop in the homicide rate over the same period.

At the time of the settlement, Nutter said the agreement was not to be read as a sign that the city would back off stop-and-frisk, a practice that has been upheld by the U.S. Supreme Court. He hoped that providing the ACLU with reports on thousands of stops each year would help restore the public's trust in police.

Since then, Police Commissioner Charles H. Ramsey has set up databases to track the legality of stops, instituted new training programs for officers, banned stops for loitering, and appointed a full-time inspector to review stop-and-frisk forms.

The theory behind stop-and-frisk is that intense police pressure in high-crime areas helps take guns off the streets. Police can legally frisk pedestrians if they have a reasonable suspicion of criminal activity. But experts say training is essential to ensure that officers know when they are allowed to conduct stops and searches.

For instance, among the cases reviewed for this week's report, ACLU lawyers found one instance in which an officer stopped a man for holding multiple DVDs.

"We're not arguing for the end of stop and frisk," said lawyer David Rudovsky, who helped prepare the report. "We're just saying, do it right. The key thing now is accountability."

According to Tuesday's report, Philadelphia police make more than 200,000 stops a year. Those totals have not decreased significantly since the 2011 settlement.

By comparison, in New York, a city with five times the population, officers made 700,000 stops a year. That number declined to about 50,000 last year after a hotly contested civil rights lawsuit and the election of Mayor Bill DeBlasio, who ran against the policy throughout his campaign.

Tuesday's report on Philadelphia also found that out of a randomly selected group of 2,519 pedestrian stops in the first half of 2014, 433 frisks were conducted - 43 percent without reasonable suspicion. The Police Department's analysis of those same cases put the number of improper frisks at 5 percent.

The report analyzed the racial breakdown of cases in which stopped individuals were later frisked by police and found minorities were more likely to be targeted, even after accounting for criteria like the larger minority populations in high-crime areas.

Police recovered only two guns in all of the 433 frisks analyzed - a "hit rate" of less than 1 percent.

Experts say that finding a weapon in 15 percent to 30 percent of frisks is a good indicator that officers are patting down suspects only when they have reasonable suspicion that the people they stop are carrying weapons.

But a "hit rate" in that range would not satisfy many of the policy's chief opponents, including the Rev. Mark Tyler of Mother Bethel A.M.E. Church, who helped bring thousands of demonstrators to Independence Hall last month to rally for changes including the end of stop-and-frisk.

On Tuesday, he said that even a reformed stop-and-frisk would be "a policy we can't gamble on."

"You'd find it hard to tolerate a policy of stop-and-frisk on the Main Line where middle-class white Americans were stopped," Tyler said. "When you begin to deny rights to people who are perceived to be on the bottom of the totem pole, it's only a matter of time until everyone's rights are eroded."