Teachers certified by the National Board for Professional Teaching Standards are for the most part no more effective in producing student academic progress than teachers without the special status, a long-awaited study concludes.
The research, which draws on one of the largest data sets used so far to examine the credential, was completed well over a year ago. But the board did not provide any public information about the less-than-flattering portrait until earlier this month, when an 鈥渙verview鈥� was posted on the organization鈥檚 Web site. And it put out the summary only after being pressed by a prominent education blogger. National-board officials say they do not intend to release the full study.
Conducted by William L. Sanders, the statistician who pioneered the concept of 鈥渧alue added鈥� analysis of teacher effectiveness, the study found that there was little difference in students鈥� achievement levels for teachers who earned the prestigious NBPTS credential, those who tried but failed to earn it, and those who never tried to get the certification.
鈥淭he amount of variability among teachers with the same NBPTS certification status is considerably greater than the differences between teachers of different status,鈥� says the summary. The study examined more than 35,000 student records and more than 800 teachers in the Charlotte-Mecklenburg and Wake County districts in North Carolina, the state with the most board-certified teachers.
Mr. Sanders, who manages the value-added assessment and research center at the private SAS Institute in Cary, N.C., said one way to think about the implications of the study would be to envision two teachers with identical experience and education applying for the same job鈥攐ne holding national-board certification and one not. To choose the board-certified teacher over the teacher without the credential would be 鈥渙nly trivially better than a coin flip,鈥� the researcher said.
Mr. Sanders drew conclusions that appear significantly different from those reached by Dan Goldhaber, who also looked at student-test results from North Carolina, though several times the number used by Mr. Sanders and from different years. In a 2004 press release put out by the board to announce his findings, Mr. Goldhaber, a researcher at the University of Washington in Seattle and the Urban Institute in the District of Columbia, said his study appeared to confirm that the board 鈥渋s in fact identifying those teachers who are more effectively producing student-learning gains.鈥�
Similar Yet Different
Nonetheless, both researchers last week referred to similarities in their results. But while Mr. Goldhaber stressed the reality of small effects on student-learning gains in some grades related to national certification, Mr. Sanders said he thought too much had been made of those.
A number of studies have examined the worth of the National Board for Professional Teaching Standards credential.
October 2000: Nationally certified teachers from Delaware, the District of Columbia, North Carolina, Ohio, and Virginia performed better than their colleagues without the credential on day-to-day dimensions of teacher expertise. (65 Teachers)
May 2002: The gains of student of board-certified teachers were no greater on average than those made by students of other teachers in Chattanooga, Tenn.
(16 Teachers)
September 2004: In three standardized tests given to 3rd through 6th graders in 14 Arizona districts, nationally certified teachers were linked to an overall average one-month gain in their students鈥� performance in contrast to others in the same districts. (35 Teachers)
December 2004: Ninth and 10th gradersin the Miami-Dade County school district whose mathematics teachers were certified by the national board scored slightly higher than other students on a Florida math exam. (100,000 student records)
SOURCE: Education Week
鈥淭he one thing I feel people have read over the top of is the magnitude of the effects in the Goldhaber report,鈥� he said. The link between several years of classroom experience and student academic progress is many times stronger than any link with national certification, Mr. Sanders pointed out.
Mr. Sanders鈥� findings came to light last week after Andrew J. Rotherham鈥攁 co-founder and the director of the Washington-based Education Sector, a nonprofit think tank鈥攗sed a posting on his to note that the board had apparently been 鈥渟itting on鈥� the results because they were not favorable.
The summary of Mr. Sanders鈥� research appeared soon after that, although board officials denied the posting was prompted by Mr. Rotherham鈥檚 entry on his blog.
The Arlington, Va.-based NBPTS offers its stamp of approval to teachers who successfully negotiate a lengthy and rigorous evaluation. So far, it has certified some 47,500 teachers at a cost estimated at more than $600 million in private, state, and federal money. On top of that, states and districts are spending tens of millions a year on bonus pay for nationally certified teachers as a way to promote better teaching and more learning.
The board has been under increasing pressure to show that the money is wisely spent.
In fact, board officials approached Mr. Sanders to conduct his study even before they launched a broader effort in 2002 to examine the worth of the credential. The board is not under a legal obligation to make the studies it has commissioned publicly available, but it has mostly done so.
鈥楤rutally Fair鈥�
Mr. Sanders鈥� research is represented by a summary that he did not see before it was posted on the board鈥檚 Web site. 鈥淚t鈥檚 not an adequate summary, and in some ways, it is terribly misleading,鈥� he charged in an interview last week.
The overview is largely critical of the study, citing methodological problems. For instance, it said the study failed to account adequately for a change in the state test and lacked a sufficient number of teachers.
Mr. Sanders rebutted the charges and defended his work, noting he had added two analytical models at the request of board officials. 鈥淚 spent weeks and weeks and weeks and weeks on this analysis trying to be just as brutally fair and professionally objective as I could,鈥� he maintained.
Mary E. Dilworth, the board鈥檚 vice president for higher education initiatives and research, said she 鈥渨ouldn鈥檛 look at [Mr. Sanders鈥橾 results as damaging in any way.鈥�
She noted that the board is under no legal obligation to make the paper available since it commissioned the work, and that such decisions are made on a case-by-case basis.
鈥淲e hope to use this report as well as others for a better certification system,鈥� she said. Some half-dozen additional studies of the credential鈥檚 worth are due in the next several months, she pointed out.
Despite the findings of his study, Mr. Sanders said he believes in the concept of the national board and has been urging officials there to modify the certification process so that it would better reflect the research on student academic gains.
How much doubt the study will cast on the value of national certification among policymakers remains to be seen.
Lynn M. Cornett, the senior vice president for education policies at the Southern Regional Education Board, a research and advocacy group based in Atlanta that works to improve schools across the South, said she could not comment on the study because she had not seen it.
Still, she said, 鈥淚 think it adds to the continuing discussion of how to use 鈥� incentives in states,鈥� a discussion that is likely to broaden out beyond the bonuses now paid to nationally certified teachers in almost all the SREB states. She noted, for instance, that Georgia now requires newly minted nationally certified teachers to teach in low-performing schools to get bonuses and also pays mentor teachers extra.
Mr. Rotherham, a former White House education aide under President Clinton who was named in 2005 to the Virginia state board of education, said in an interview that the board鈥檚 failure to be more open about the research was likely to hurt it in the long run.
鈥淭hey have needlessly aroused suspicion about what they鈥檝e done and needlessly handed their critics ammunition,鈥� he said. 鈥淚t鈥檚 all so 鈥� political.鈥�
