This week brought the twelfth edition of the yearly Quality Counts report, which evaluates public education in the states and nation. Each of those political entities is graded in six areas, and those six grades are then combined to yield for every state (and the country and D.C.) an overall grade. The best thing about QC Version 12.0 is its focus on teachers and a selection of indicators by which it judges whether policymakers are committed to putting the best educators in the classroom.
In the past, QC has concentrated on "the state's role as a gatekeeper" for teaching by looking at licensure requirements and such. This year, the report shifts its lens to "the state's role in attracting, developing, deploying, and keeping the very best education workforce possible." And QC's editors rightly think part of that role is advancing smart pay-for-performance initiatives (see here). The report contains much valuable data about the teaching profession; for example, 43 states formally evaluate their teachers, but only 23 formally train the evaluators. In only 12 states is teacher evaluation tied to student achievement.
But QC has sundry drawbacks. For starters, its creators unaccountably (and, we think, knowingly, willfully, and politically) ignored the path-breaking work in this field by the National Council on Teacher Quality, including that organization's superb state teacher policy "yearbook," which you can find here. But there's more. An essay by the report's research director, Christopher Swanson, purports to compare teacher salaries to those in "comparable" jobs. But it fails to account for the fact that teachers work about 9 months a year while architects, editors, etc. typically work a full 12 months. That's both unfair and misleading. Asked about that disparity, Swanson answered (1) that some teachers may think they work year-round and (2) that many other jobs--such as consulting positions--have substantial vacation time. Sorry, Chris, it just doesn't wash. (This QC graph, which borrows from the left-leaning Economic Policy Institute, notes that teachers make 88 cents for every dollar earned by those in "16 comparable occupations." But because teachers work only 9 months out of the year, or 75 percent of what those in the other 16 occupations work, 88 cents on the dollar ain't half bad!)
Swanson's essay also contains a graph showing that teacher salaries are "constrained" and that a mere 2 percent of educators make $80,000 or more. But teacher salaries are so similar because of the lock-step salary schedules negotiated by their unions. An all-star teacher will probably never make over $80,000, but neither will a lousy teacher; instead, both will make roughly the same amount. Educators' wages are in this way "constrained" by their own union representatives and collectively bargained contracts. As the graph shows, the salaries of other professions have a more gradual slope. That's because there is no, say, National United Computer Programmers union that sets salaries for all the nation's tech nerds, leaving programmers free to negotiate for salaries based on their individual experience and skills.
This is a big reason why comparing teacher wages and wage progressions to those of other occupations (even supposedly "comparable" ones) is tough to do right. Not only do teachers work 180-190 days in a typical year, but they are also public-sector workers who have their salaries negotiated by affiliates of giant national unions. And teachers, unlike the folks working in the 16 "comparable" professions, are not accountable for their on-their-job performance. Plus, QC doesn't seem to factor in benefits, which are enormously generous for educators. For all those reasons, a straight salary-to-salary comparison is therefore largely misleading.
Another problem with QC: the Chance-for-Success Index is back, the "demography is destiny" approach that the Gadfly lambasted about a year ago. This edition of QC doesn't give the Index as much exposure, but the Chance-for-Success rating still comprises one of the six components factored into a state's final grade. It shouldn't. Not only does the Index unwisely pre-determine a child's odds of success by citing factors such as his parents' income and whether his parents speak English fluently (and then, inexplicably, rank states by such factors), but it also makes no allowance for the fact that Americans are highly mobile. If a well-educated person lives her entire life in Arizona, earns her college degree there, and then moves to New York City to work on Wall Street, QC counts her success toward New York's overall education ranking!
The Chance-for-Success Index cannot account for such a person, nor for one born in Baton Rouge who went to college in Atlanta, got his first job in Boston, and now lives in Seattle. The Gadfly noted last year that "well over 30 percent of college-educated Americans received their k-12 educations in more than one state," and QC ignores that fact, too. The Chance-for-Success results are easy to predict. The top five: Massachusetts, New Jersey, New Hampshire, Connecticut, and Vermont (average median income $58,000). The bottom five: West Virginia, Nevada, New Mexico, Louisiana, Mississippi (average median income $41,000). It might be easier for the QC editors if, in place of their Chance-for-Success rankings, they simply substituted census bureau income data. The logic behind this so-called research is just shoddy, through and through.
A final problem with QC #12: Where's the choice? Educational choice, especially in the form of charter schools and state charter-school laws, needs some weight in QC's evaluation and gets none at all.
Still, you'll want to take a look. The grades and ranks are actually the least informative or helpful part (although, if you don't like the grades, you can make your own). Much better are the articles, especially those about the present condition of states' teacher preparation and human-capital development. You can find it all here.