Does Teacher Certification Matter?

A quarter of Michigan’s charter school teachers are not government certified. At the schools run by Charter School Administration Services, nearly two-thirds are uncertified.1

These statistics were the subject of universal hand-wringing during a State Board of Education meeting this month. The Board’s President, Kathleen Straus, called them “pretty scary.”2 Charter schools spokesman Dan Quisenberry subsequently explained how charter schools’ urban settings lead to lower teacher certification rates, but no one directly challenged the idea that teacher certification is a critical factor in student learning.

Stay Engaged

Receive our weekly emails!

In Michigan, the only requirement for initial certification is that the candidate graduate from a government-approved teacher training program. To continue working for more than six years, teachers must eventually complete some additional training classes.

Are teachers college graduates really better at raising student achievement than teachers who lack that pedagogical pedigree? Much has been written on this subject, most of it grossly flawed. After sifting through the research in 2001, Kate Walsh of the Abell Foundation identified seven well-designed studies of the effects of teacher certification on student achievement. All of them concluded that “New teachers who are certified do not produce greater student gains than new teachers who are not certified.”3

Needless to say, Walsh’s review of the research sparked controversy. Linda Darling-Hammond, a leading advocate of certification with whose work Walsh had found considerable fault, wrote a harsh critique of Walsh’s findings. That critique was soon persuasively rebutted, however, by a joint response from Walsh and education economist Michael Podgursky. A preponderance of evidence now suggests that college of education degrees really don’t lead to better teaching in most cases.

So what does? Some researchers have found that experience is a significant factor, but others have found that it is not. Two recent studies suggest that a few years of experience helps, but that teacher effectiveness eventually starts to decline — a phenomenon ascribed to “burnout.”4 Other key factors associated with high-quality teaching are verbal ability and having a college degree in the subject being taught.

These findings confirm the common-sense conclusion that more knowledgeable, eloquent people make better teachers. Unfortunately, instead of encouraging public school principals to hire the best communicators with the most subject-area knowledge, we forbid them from even considering anyone who lacks a teaching degree — even though teaching degrees don’t usually pay off in the classroom.

Of course, it’s possible that Michigan’s teacher certification program is somehow exceptional and that teacher certification is therefore worthwhile here. But if so, charter schools in Michigan would tend to produce smaller student gains than other public schools, since charter schools employ fewer certified teachers.

In reality, however, students’ MEAP scores go up more from one year to the next in Michigan’s charter schools than they do in the state’s traditional public schools. And schools run by Charter School Administration Services compare favorably overall to those of the nearby Detroit and Flint school districts, despite the reports that two-thirds of CSAS teachers are uncertified.

Consider the following table, which shows the MEAP passing rates for the all the CSAS-managed schools for which data were available, along with those of the nearby Flint and Detroit public school districts.

Winter 2004 MEAP Passing Rates


Grade 4

Grade 5

Grade 7

Grade 8


Detroit Public Schools






Flint Public Schools






Academy of Flint






Academy of Lathrup Village





Academy of Oak Park






Academy of Southfield






Academy of Westland





Academy of Detroit-West




Academy of Waterford



Source: Computed from raw data on the Department of Education Web site5

The average 2004 MEAP passing rate at these CSAS charter schools is 36.7 percent, somewhat higher than the 34.6 percent in Detroit public schools, and far higher than the 29.3 percent in Flint public schools. Since the research consensus does not support the idea that charter schools are “skimming” good students from other schools, these results are hard to ignore.6

It would seem, then, that teacher certification confers no greater benefit in this state than in any other. What does make a difference is a combination of school autonomy and parental choice. When school principals are free to hire anyone they want, but are forced by competitive pressures to hire only the best-qualified people they can find, they usually make wise selections.

Instead, we have built a public school system that could not hire Bill Gates to teach computer science, Sandra Day O’Connor to teach American government, or Lance Armstrong to teach physical education. These folks might not be looking for work just now, but many other knowledgeable and eloquent people are — people who will never have the chance to sit behind a teacher’s desk simply because they lack a dubious government credential.

That’s “pretty scary” – just like the state Board of Education’s blind spot on this issue.


Andrew J. Coulson is senior fellow in education policy for the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland, Mich. Permission to reprint in whole or in part is hereby granted, provided that the author and the Center are properly cited.


1Michigan Information and Research Service (, “Teacher Non-Certification Rates High In PSAs”, August 11, 2004.

2Michigan Information and Research Service (, “Teacher Non-Certification Rates High In PSAs”, August 11, 2004.

3Kate Walsh, Abell Report,

4Hanushek, E., Kain, J., Rivkin, S., 1998 (revised, 2000), Teachers, schools and academic achievement. NBER Working Paper No.6691.,%20schools,%20and%20acheivement.pdf
Kain, J. and Singleton, K., 1996, Equality of educational opportunity revisited. New England Economic Review, May-June: 87-111.

5For each grade shown, these figures represent the average percentage of students passing the MEAP test (i.e., scoring at Level 1 or 2) across all subject areas for which scores were available.

6The consensus among researchers is that charter schools, on the whole, do not “skim” the most academically able students off of their surrounding school districts. In Michigan, there is evidence that at least some charter schools do the opposite, attracting students who had been performing poorly in the public schools or who had already dropped out. [Note: To most easily view these documents, which are in Adobe’s .PDF format, right click on the links and choose “Save As” from the menu.]