Includes updates and/or revisions.
Charter schools, teacher-credentialing requirements, and programs like Teach For America can all make a positive difference for student achievement in high school, according to a trio of new federally funded studies.
The studies, presented here during the March 24-28 annual meeting of the American 91制片厂视频al Research Association, all come out of the , or CALDER, a federal research center based at the Urban Institute in Washington.
The center was launched two years ago to mine the longitudinal databases on student achievement that are piling up as states move to new data-collection systems that use individual 鈥渋dentifier鈥� numbers. Such numbers make it possible to track students鈥攁nd teachers鈥� anonymously over time as they move from classroom to classroom or district to district.
In the first of the three studies, Tim R. Sass, an economist at Florida State University in Tallahassee, drew on data from Chicago and across Florida to explore whether students fare better in charter or regular public high schools. Are they more likely to graduate, for instance, or attend college within a year of graduation? Do they score higher on state-mandated tests?
To find out, Mr. Sass tracked successive waves of 9th graders, starting in 1997, as they moved through high school. In an effort to keep the playing field level, though, Mr. Sass limited his sample to students coming from charter middle schools.
His analysis showed that students who attended charter high schools were 11 to 14 percentage points more likely to graduate with a standard diploma, and 10 to 13 percentage points more likely to go on to college, than were their counterparts in regular high schools.
Noting that some students had no nearby charter high schools,Mr. Sass also refined his analysis to account for distances between the charter middle schools and different types of high schools, but he said the magnitude of the results remained basically the same.
鈥淲hile I鈥檓 not sure I鈥檇 take these results to the bank,鈥滿r. Sass said, 鈥渢hey鈥檙e certainly suggestive that school choice might be part of an overall package to improve high schools.鈥�
In the second of the three studies, researcher Helen F. Ladd and her colleagues used statewide data from North Carolina to analyze whether the type of credential a high school teacher holds has an impact on student performance.
鈥淢ost studies on teacher credentialing so far have been done at the elementary level,鈥� said Ms. Ladd, a professor of public policy studies and economics at Duke University in Durham, N.C.
Degrees and Experience
To see if credentials make a difference in high school, Ms. Ladd drew on four years of scores from state-mandated end-of-course tests taken by 9th and 10th graders in five subjects: English, algebra, biology, geometry, and social studies.
The researchers found that students fared better academically when their teachers had regular teaching licenses as opposed to alternative or emergency credentials, or when teachers were certified in the subjects they taught or a related subject鈥攑articularly in mathematics and science.
And teachers certified by the National Board for Professional Teaching Standards, a private Arlington, Va.-based group that awards credentials to teachers who complete a process to document their performance, tended to have higher-scoring students than did teachers without that certification.
Student performance was also slightly better, on average, when teachers had a master鈥檚 degree. Likewise, more-experienced teachers appeared to produce better results than novice teachers, according to the Duke study.
The positive effects were small but cumulative. The study calculates, for instance, that the effect of having a teacher with strong, rather than weak, credentials is similar in size to the effect of being taught in a classroom that has five fewer pupils than average.
鈥淲e conclude that teacher credentials matter in a systematic way for student achievement at the high school level, and that the magnitudes are large enough to be policy relevant,鈥� the authors write in their paper.
The Duke study also found, however, that such teachers are unevenly distributed in North Carolina schools. Students who are African-American or who come from poor families are less likely, on average, to have teachers who are experienced, who have regular teaching licenses, or who hold national- board certification than white or better-off students. 鈥淚f the teachers assigned to black students had the same credentials on average as those assigned to white students,鈥� the study concludes, 鈥渢he achievement difference between black and white students would be reduced by about one-third.鈥�
For a study on the Teach For America program, researcher Jane Hannaway and her co-authors also rely on data from North Carolina鈥檚 end-of-course tests. Begun in 1990, TFA recruits graduates from top colleges and universities to teach two years in disadvantaged, hard-to-staff schools.
Ms. Hannaway鈥檚 study, the first to look at the program鈥檚 effects in high school, finds that students perform better in classes with TFA teachers than they do with the non- TFA teachers who would otherwise be teaching them. The size of the effect, though small, is two to three times larger than the effect of having an experienced teacher compared with one who鈥檚 been on the job less than three years, according to the study.
鈥淎t least in high school, our findings suggest there may be a greater payoff to focusing on teacher selection rather than teacher retention,鈥� said Ms. Hannaway, CALDER鈥檚 principal investigator and the director of the Urban Institute鈥檚 education policy studies center.
What鈥檚 still unclear about all the studies, said another CALDER researcher, Daniel D. Goldhaber, a research associate professor at the University of Washington in Seattle, is whether they will generalize nationally or to other states.
