Teacher certification.

Kane, Rockoff, and Staiger's findings ("Photo Finish: Certification Doesn't Guarantee a Winner," research, Winter 2007), supported by other recent research (including my own), are persuasive in suggesting that certification requirements do little to create or identify effective teachers. The findings might have been strengthened if measures of teacher ability had been included in the analysis, because the alternatively certified teachers in the study, from Teach For America (TFA) and the NYC Teaching Fellows program, represent high-achieving students from the nation's elite universities. Thus one might interpret the finding of little difference between regularly certified teachers and their more pedigreed peers to show that teacher certification requirements compensate for greater intellect or better overall undergraduate academic preparation.

The wide variation in quality within groups should come as no surprise, though documenting it with sound research is helpful. What is unexpected, and of concern to those who believe recruiting better-quality undergraduates will improve teacher quality, is that by year three the TFA alumni aren't doing substantially better than their regularly certified counterparts.

This thoughtful research raises important policy questions. Can we do without certification requirements? This study doesn't answer that question. What we really want to know is how to create effective teachers. If, as the authors suggest, classroom experience and not certification is linked to effective teaching, we might consider moving teacher preparation programs toward a model of more K-12 classroom experience, and away from coursework.

LAURA M. DESIMONE

Assistant Professor of Public Policy and Education

Vanderbilt University 

Thank you for featuring research on the fundamental role of teachers in the achievement of New York City's students in "Photo Finish." The Board of Regents and the New York State Education Department are committed to improving the quality of teaching for the benefit of all our students.

[ILLUSTRATION OMITTED]

We agree that multiple pathways to certification can be effective at recruiting teachers who can improve student achievement. The conclusion, however, that "certification matters little" is not supported by the analysis presented. By using teachers' certification status when hired, rather than when providing instruction, the authors cannot assess the impact of certification on student achievement. Because of state requirements, all teachers hired without certification would have been on a pathway toward certification and many did become certified during the study period. In addition, because certification is required, it is virtually impossible to explore what would happen in its absence. For example, ill-prepared individuals who do not attempt to teach because certification is required, but who could become teachers if certification were not required, are not represented in the group labeled uncertified in the study.

The Board of Regents has always required teachers without certification to be making satisfactory progress toward certification in order to remain employed. Teachers receive "transitional" certificates that permit them to teach for up to three years provided they receive school-based mentoring and make satisfactory progress toward full certification. Similar requirements applied to individuals with temporary licenses prior to the elimination of that pathway. The preparation and induction of these individuals likely would be quite different if certification were not required.

JOHANNA DUNCAN-POITIER

Deputy Commissioner

New York State Education Department