Pupils of Licensed Teachers Found to Score Higher in Math
Students appear to learn more mathematics when their teachers are certified, according to three studies out this month.
The studies, presented during the April 12-16 meeting here of the American Educational Research Association, may fan the growing nationwide debate over how to determine whether a teacher is qualified.
Under the federal No Child Left Behind Act, schools have until the end of the 2005-06 school year to ensure that their classrooms are staffed by "highly qualified" teachers. The law defines highly qualified teachers as those who have a bachelor’s degree, hold state certification, and have demonstrated competency in the subjects they teach. But some states, eyeing teacher shortages, are lowering their certification requirements to meet the federal mandate, which allows states wide leeway in deciding what’s needed for certification.
Even before the federal law was passed, some experts argued that state certification requirements kept otherwise-qualified teaching candidates out of the profession.
But leaders of schools of education and others maintain that teachers cannot succeed without the pedagogical know-how that state licensing exams and other certification requirements reflect.
The new studies tested that proposition with groups of students ranging from a national sample of kindergartners to Texas middle schoolers. After accounting for differences in students’ socioeconomic backgrounds, teachers’ experience, and students’ previous achievement, all three studies concluded that students of certified teachers seemed to fare better—and, in some cases, gain more—on standardized tests than students whose teachers had only emergency or temporary licenses.
Presented at different sessions here, the reports were among hundreds of studies highlighted at the meeting, which drew more than 12,000 education researchers.
Gains in Kindergarten
"Certification matters, and it matters in elementary school," said Vanderbilt University’s Kristie J. Rowley, the author of one study. Focusing on kindergarten pupils and their teachers, her study draws on data from the Early Childhood Longitudinal Study, a federal initiative that has been tracking students since 1998 and testing them in the fall and spring.
By the spring of the kindergarten year, Ms. Rowley found, children whose teachers had temporary or emergency licenses scored an average of 1.26 test-score points lower than peers who were taught by certified teachers.
The only other factor that seemed to be associated with math-score gains, the study found, was whether teachers had taken more preservice courses in elementary education or in early- childhood education. Pupils’ scores improved more in classes with teachers who had focused more on the former.
The number of years that teachers had taught, or the number of math courses that teachers had taken in college, did not make much difference in terms of children’s math-score gains in kindergarten, according to that study.
Having had some preservice training in math, however, was deemed more important in another study, conducted by the American Institutes for Research, a Washington-based research group. The AIR researchers focused on 8th graders, analyzing their scores from the 2000 administration of the National Assessment of Educational Progress math test.
Because the congressionally mandated exams are designed to take a snapshot of academic achievement at one point in time, researchers could not completely control for differences in students’ academic achievement the way the other two studies did. They could determine, though, whether students had been tracked into high-, middle-, or low-ability math classes.
As was done in the kindergarten study, the AIR researchers tested several teacher characteristics to see which ones linked to better student achievement on the tests. They found that only two seemed to matter: whether pupils’ teachers were certified, and whether they had majored or minored in math in college.
The differences translated into an advantage of 9 percentile points for students with certified teachers and 5 percentile points for 8th graders whose teachers had specialized in math, according to Elizabeth Greenberg, the study’s lead author.
The researchers also found that teachers with those characteristics were unevenly distributed among classrooms nationwide. Poor students and students in low-ability classes were less likely to have teachers with licenses or math specializations.
Results in Texas
"We’re taking the students most at risk of educational failure, and they seem to be those least likely to have teachers with the characteristics of qualified teachers," said Ms. Greenberg.
Unlike the other two studies, the Texas study began in response to efforts in that state to loosen teacher-licensing requirements so that people who had not graduated from education schools could earn temporary certificates.
Researchers from the Southwest Educational Development Laboratory, a federal regional education laboratory, and the University of Texas at Austin analyzed math-score improvements from 1998 to 1999 on the state-mandated Texas Assessment of Academic Skills tests.
Among the 578,000 middle school students the authors studied, the same pattern held: Scores improved more when their teachers had regular certification. It mattered much less whether their teachers were novices or veterans.
Such findings run counter to other research in the field suggesting that students do just as well in classes with emergency-certified teachers as they do with teachers who take regular routes to certification.
Dan D. Goldhaber, whose own 2000 study found no achievement differences between students in those two groups, said national studies are difficult to do because state certification requirements vary widely.
"People tend to read certification as one thing," he said, "when in fact it’s 50 different things."
Vol. 23, Issue 33, Page 10Published in Print: April 28, 2004, as Pupils of Licensed Teachers Found to Score Higher in Math