Uncertified teachers end up performing just as well in the classroom as certified teachers and alternatively trained teachers like Teaching Fellows, a study to be released today says.

The study’s results appear to challenge requirements under the federal No Child Left Behind Act that every classroom have a “highly qualified” teacher, instead suggesting that schools should put more emphasis on weeding out bad apples after the teachers have been hired.

“These are people who have no prior experience in teaching and they go into the lowest performing schools, and they do just as well,” a Columbia University Business School professor, Jonah Rockoff, who co-authored the study, said. “Where you went to college and what your GPA was doesn’t seem to tell you how good you’re going to be in the classroom.”

In the study, researchers at the Hoover Institution, a conservative think tank affiliated with Stanford University, used standardized test scores to measure the performance of New York City students taught by traditionally certified teachers, uncertified teachers, and teachers who enter the profession through alternative programs such as Teach for America and Teaching Fellows. They found that while alternatively certified and uncertified teachers do worse at first, they appear to improve at faster rates than traditionally certified teachers in their first years on the job. By the teachers’ third year on the job, students of alternatively certified and uncertified teachers are performing just as well as those of traditionally certified teachers.

That’s good news for New York City public schools, since the majority of new hires during the past five years have come through alternative certification programs. Currently, a third of all teachers in city schools received certification from an alternative program.

Alternative certification programs have grown in popularity around the country since the No Child Left Behind law introduced a requirement that all school districts have a “highly qualified” teacher — meaning a teacher with a certificate — in every classroom by July 2006. New York City had already been under pressure to hire more certified teachers after the state education commissioner, Richard Mills, sued the city in 1999 for placing uncertified teachers in the lowest performing schools. To meet the requirements, the city invented the Teaching Fellow program in 2000 to recruit teachers from other professions and speed them through the certification process. The city has hired 9,000 Teaching Fellows since then.

The study shows that uncertified teachers, who are more likely to be minorities than the other groups, end up doing just as well as the alternatively and traditionally certified teachers. Since the Teaching Fellows program was introduced, hiring of minority teachers has dropped significantly, a trend that can be attributed to the effort to remove uncertified teachers, Mr. Rockoff said. Statistics first reported by the Amsterdam News showed that in 2001, 27% of new teachers were black, while this year only 14% were black. The percentage of new Hispanic teachers has also dropped.

The study’s authors say they are not “proposing to open the floodgates into teaching” by saying certification doesn’t matter. But researchers said the study results showed that school systems, instead of focusing on whether and how teachers are certified before hiring, should worry more about getting rid of teachers who perform badly during probationary periods. Currently, Mr. Rockoff said, large urban school systems like New York with dismal teacher retention rates tend to approve tenure for all teachers who decide to stay on, rarely giving out unsatisfactory ratings to teachers who perform badly. In New York City, half of all teachers — traditionally certified or not — leave after five years.

To become certified, teachers must take a series of tests, have a bachelor’s degree that includes education coursework, or complete graduate level education coursework. Alternative certificate programs often allow teachers to do the coursework during their first year of teaching.

A professor at Stanford University, Susanna Loeb, who has conducted a study of teacher qualifications very similar to the Hoover study, said that certification status matters little in determining how a teacher will do in the classroom. She added that recommendations, interviews and grades that give information about a teacher’s past experiences and educational achievement should remain important factors in hiring decisions, though.

“I’m not ready to give up on resumes,” she said.

The Department of Education said the findings supported their argument that teachers who perform better should be paid more. The department said it would be allocating funds “to pay more to teachers who contribute more, including pay differentials that help make sure our high-needs schools get lead teachers as well as math, science, and special education teachers.”

The president of the city teachers union, the United Federation of Teachers, Randi Weingarten pointed to another finding that supports one of the union’s longstanding arguments: that experience is a defining factor in good teaching.

“The most successful teachers are the ones who have experience and have been mentored or given other supports to help them learn how to teach in their earliest days in the profession,” Ms. Weingarten said, adding that the study “also reaffirms that regardless of whether someone has been certified in the traditional way or in a new way, you can’t just plop them into the school system and see if they sink or swim.”

The other authors of the Hoover study are Thomas Kane of Harvard’s Graduate School of Education and Douglas Staiger, a professor of economics at Dartmouth.