Innovation Ohio ? a newly birthed ?non partisan progressive think tank? (incidentally run a by former top aide to Governor Strickland, with research conducted by a former Democratic lawmaker who helped push through then-governor Strickland's ill-conceived school funding plan) ? just released its first research document, timed with heated protests over the state's collective bargaining reform bill (SB 5). But ?Ohio Teachers and Collective Bargaining: An Analysis? is not so much an ?analysis? as it is an amalgam of statistics from the national Bureau of Labor Statistics, wrought with faulty assumptions about the purpose of collective bargaining reform and an assortment of cardinal sins when it comes to basic statistics.
Where to begin? It's only eight pages long, but manages to mislead on a number of fronts.
For starters, the brief draws on two years' of data (2008 and 2009) showing the average annual wages for kindergarten, elementary, middle school, and high school teachers in Ohio. Comparing average wages for teachers between the two years, all four categories saw a drop from 2008 to 2009, with an average drop of 3.8 percent.
The average annual wage dropped from one year to the next. The brief builds on this trend ? an observable wage change from year to year?and calls it ?pay cuts? in the next paragraph, arguing ?only Utah and Michigan's teachers have seen larger pay cuts,? that the state was ?only one of six ? whose teachers saw salary cuts to all four categories of teachers,? and that ?teachers saw $150 million less in wages last year.?
Innovation Ohio uses this drop in wages as evidence of the extraordinary ?financial sacrifices ? among the largest in the country? that the state's teachers have faced. But a drop in the statewide mean wage (mean = the mathematical average of a set of numbers) does not mean teachers saw their salaries slashed. The vast majority of teacher contracts protect teachers from pay cuts, and even during the Great Recession many teachers unions managed to eke out pay increases during negotiations.
An alternative explanation might be that the makeup of Ohio's teacher pool changed from 2008 to 2009 and could be the cause of a lower average. Not quite as sexy as ?financial sacrifice,? but when conducting such research it's important to be honest about reasonable hypotheses.
Here's one: Ohio's teacher workforce, like that of most states, is aging rapidly. Teachers are retiring and younger ones are replacing them, so the overall face of the profession is getting younger. Teacher data listed on the Ohio Department of Education's website shows a unique drop between 2007-08 and 2008-09 in terms of Ohio teachers' average years of experience. The drop from 14.14 years to 13.98 years of experience means that in a given district, teachers were younger on average (and therefore cost less, since teacher salaries are based on degrees and years of experience) in 2009 than they were in 2008. ?
Of course, there are dozens of other reasons the statewide average might have been lower; the point is that a change in an average could mean lots of things other than that teachers saw pay cuts. (But if anecdotes are sufficient evidence for Innovation Ohio, then read here, here, here, here, and here for proof that most teachers unions have held the line and refused even wage freezes despite the recession? for dozens more, just google search ?Ohio? and ?teachers union rejects pay freeze?.)
Secondly, the report mistakes correlation for causation. The brief is riddled with this vice. It moves from a basic correlation or trend ? drops or increases in average wages across states, correlated with whether states happen to require collective bargaining or outlaw it ? to sweeping claims that collective bargaining (or its absence) is the primary cause of the observed trends. Using several frightening if-then statements, the brief attempts to make the case ? based on simple correlations found in two years worth of data- that if Ohio eliminates collective bargaining for teachers it will ?lead to salary increases ? that are more volatile,? cause ?uncertainty? for taxpayers, and will ?certainly? not have a significant positive impact on student success.?
But the absence of collective bargaining in a number of states didn't cause low achievement or spending volatility any more than it caused the flooding in northern Ohio this week. At the very least, if Innovation Ohio wishes to make such profound claims, it should hire the services of a statistician who can perform regression analysis or any number of techniques that can viably distinguish correlation from causation.
Third, the report uses Education Week's ?student achievement? rankings to measure whether states in collective bargaining or Right-to-Work states are correlated with student success. Why not use actual student data instead of a somewhat random ranking system (which by the way, changes thematically from year to year) that measures, among other things, non K-12 variables like ?adult outcomes.? State-by-state NAEP data are readily available, as are state achievement trends.
Fourth, Innovation Ohio uses a single Race to the Top reviewer's flattering comment about the state's application as evidence that current systems of teacher accountability and evaluation are sufficient. If the viability of Ohio's teacher evaluation current system rests on an anecdotal comment by a single grant application reviewer, Innovation Ohio's threshold of evidence is criminally low. One only has to read a few more pages of reviewers' comments to see phrases such as ?A weakness in the discussion of the design of the teacher evaluation system is a lack of specificity on how important the measure of student growth is expected to be.?
In sum, it's understandable that Innovation Ohio ? led by former buddies of Gov. Strickland ? wish to defend collective bargaining. But it fails to offer up compelling arguments (and there are such arguments out there) about retaining collective bargaining and instead presents some of the most flimsy evidence we've seen yet.
According to the brief, weakening bargaining will cause spending volatility; will lead to decreased student achievement; and is unnecessary because teachers (on average) have already seen wages fall. The first of these two claims are based on flawed statistical reasoning and questionable data. The third is incredibly ironic given that the people behind this report worked for an administration that never even acknowledged ? let alone came up with solutions for ? the profound fiscal crisis Ohio faces.
The authors of the brief also assume that the primary purpose of weakening collective bargaining is to save money. They defend existing evaluation systems and antiquated personnel practices that harm poor students the most, arguing that ?it can even be argued that Ohio has done it better and more comprehensively than other states.? Really? Tell that to Louisiana, Colorado, Illinois, Florida, and others.
The teacher-related provisions in SB 5 are not merely about saving money. The bill seeks to undo the destructive practice of seniority-based layoffs, require that pay raises for teachers be based on effectiveness, and essentially gets rid of lifelong tenure (as many reform-minded states have already done). It also removes from collective bargaining issues such as class size and personnel placement.
These are common sense changes that many states have already put into law, changes that seek to put the most effective (rather than veteran) teachers in classrooms that need them the most. By failing to put forth reasonable evidence against the teacher provisions in SB 5 the authors make obvious that this new ?think tank? is anything but nonpartisan.
- Jamie Davies O'Leary