For some seniors, an appealing way to “save the world” right out of college has been through Teach for America, an organization that trains and places young people, many of whom are recent college graduates, to teach low-income students.

But a controversial study released today by the School of Education suggests that uncertified teachers — including those recruited through Teach for America — may be doing more harm than good. Some educators, however, question the methods used in the study, and Teach for America representatives say that other studies have shown contradictory results.

The study linked the achievement data of more than 130,000 fourth and fifth graders to the certification status of more than 4,000 teachers in Houston, Texas. The students took six different reading and mathematics tests over a six year period, and those taught by certified teachers consistently out-performed the students of uncertified teachers.

“A lot of questions have been raised as to whether bright college students can be effective when thrown into a classroom setting and asked to teach,” said Linda Darling-Hammond, the School of Education professor who conducted the study with three of her colleagues. The bottom line of their findings is that “certified teachers always out performed uncertified teachers,” she added.

While the typical Teach for America participant has a strong background in their field, is eager to help the kids advance and receives training the summer prior to the start of his or her first school-year, Darling-Hammond said this is not enough.

“They didn’t know what kids are supposed to learn in third grade,” she said. “They don’t know how to put a curriculum together. They didn’t know how to assess the student’s achievement.”

This often frustrates the young instructors and can make them resort to stricter teaching strategies, simply because they do not know what the alternatives are for creating a curriculum that is exciting and engaging, Darling-Hammond said.

“It is common for these teachers to create a setting in which the kids are under very, very tight control,” she added.

Special education students and non-native English speakers had the lowest academic growth rates when taught by under-qualified teachers.

Assessing the Costs

Besides having a negative effect on the academic achievement of their students, Darling-Hammond said Teach for America recruits tend to leave only two to three years after they are hired. The cost of replacing them ranges from $8,000 to $48,000.

“This raises the question of whether the money spent on replacing teachers quickly wouldn’t be better spent recruiting better teachers,” Darling-Hammond said.

Largely due to a shortage of funds, many California school districts avoid hiring large numbers of qualified applicants because they must be paid significantly more than their younger, less qualified counterparts.

“What they seem not to have factored in is the turnover cost,” Darling-Hammond added.

These problems vary depending on the policies of individual school districts, and some have made significant strides toward putting a certified teacher in every classroom, she said.

One recent success story took place in San Diego, where school districts have hired approximately 1,000 certified teachers and raised salaries to keep them there. The result, Darling-Hammond said, has been higher test scores overall.

Likewise, in New York City, half of all teachers lacked credentials. But when the city raised salaries by 20 percent, the amount of certified teachers rose to 90 percent within a year — demonstrating that the turnaround can happen very quickly, Darling-Hammond said.

But according to Abigail Smith, vice president of research and public policy at Teach for America, the teachers recruited right out of college receive proper training before they face their students.

Besides reading curriculum text books, these teachers must complete a five-week “intensive” training program and must shadow a veteran teacher to get a feel for the classroom setting, Smith explained.

In districts such as Houston, all Teach for America recruits must enroll in a certification program while they are actively teaching. The program requires them to take weekend or evening classes for their first two years on the job.

However, Darling-Hammond said that these teachers tend leave almost as soon as they become certified, a trend which exacerbates the problem of turnover costs.

Flawed Methodology?

Susanna Loeb, assistant professor at the School of Education, questioned the methods Darling-Hammond used in interpreting the data.

“While the research raises some important questions, I would want to see some more detail,” Loeb said. “There aren’t too many analyses in the paper so far, and often you need to look at the study in multiple ways to make the results more robust.”

Smith, of Teach for America, also attacked the rigor of Darling-Hammond’s methods, and said that Darling-Hammond has an “inexplicable, 12-year vendetta against Teach for America.”

Although Smith said Teach for America teachers are looking to get better at working with students who are learning English, she said that certification alone does not guarantee a good teacher.

“Certification status per se is not my concern,” Smith said. “My concern is whether teachers are having a positive impact on their kids. There are lots of factors that go into that.”

As an alternative to Darling-Hammond’s research, both Smith and Loeb suggested an alternative study put out last summer by the Mathematica Policy Research, Inc., an independent policy research firm.

The study — titled “The Effects of Teach For America on Students: Findings from a National Evaluation” — compared Teach for America instructors with other inexperienced and often uncertified teachers, and concluded that “Teach for America teachers had larger impacts on both math and reading achievement when compared with novice control teachers.”

Loeb added that Darling-Hammond’s comparison between Teach for America recruits and older, certified and more experienced teachers may not have been fair.

Darling-Hammond is slated to present her study today at the annual meeting of the American Educational Research Association in Montreal, where Loeb predicts it will face criticism from experts in the field.

“We are eager to see the peer review process for this study,” Smith said. “If other researchers give positive reviews in terms of the methodology of this study, then we are extremely open to it.”