Tuesday, May 20, 2014

Go Home, Gramps!

In the ongoing battle to get older, more experienced, and (most importantly) more expensive teachers out of schools, the only surprise is that the latest push didn't come sooner.

Education Next, the magazine by and for the discerning corporate conservative educrats at the Fordham Institute, Harvard Kennedy School, and the Hoover Institution, brings us an article about incentive programs for early retirement.

Early Retirement Payoff is the headline. The payoff is in the subhead-- "Incentive programs for veteran teachers may boost student achievement." Oh, may. "May" is such a word, such a miniature poem that promises everything but requires nothing in the way of proof or substance. Shaking my monitor may make my computer run faster. Riding a bicycle may make my hair grow back. Investing in beet farms may lead to riches.

The article offers two chunks of research, but first, authors Maria D. Fitzpatrick and Michael F. Lovenheim would like to provide some background. You may have heard that public budgets have grown tighter (kudos on the correct use of political passive voice-- an effect has been created and we will stay mum about the cause-- those budgets grew tighter in the same mysterious way that my pants have grown tighter). Teachers get old and then they retire, which leads to hiring cheaper teachers. But sometimes teachers get really old. In 2010, one third of teachers were over fifty (wish I had a link for that, but no). Of course, we could also say that experienced teachers are replaced with inexperienced ones. So can we find a link between teacher early retirement and student achievement?

After all, an early-retiring teacher might be burned out. Or he might be bailing because he sees a roomful of loser coming at him and he wants to dodge that bullet. Our intrepid researchers are curious, but fortuitously they found a big bucket full of data. From twenty years ago.

No- no- wait! Don't walk away. I am sure that there are really useful things we can learn by studying what happened in Illinois schools in the early 90s. Seriously. Okay, no, I don't believe it either. But I do believe by looking at whatever meal these folks cooked up from the leftovers that had been in the freezer for twenty years-- whatever that meal is, it will at least tell us what the Fordham, Hoover, Harvard Kennedy folks (hereafter known as FHHKs) are hungry for. So let's dive in, shall we?

The Illinois Early Retirement Incentive

There are plenty of numbers and even a graph, but the bottom line here is pretty simple.

1) When offered financial incentive to retire early, more teachers will retire. Hope we didn't spend too much money workin' that one out. Pro tip: The sun? It's rising in the East tomorrow.

2) The Illinois Teacher Retirement System needed a good accountant in the early 90s. The ERI saved school districts money and cost the state retirement system more, for a net loss. Oops.

The Effect on Student Test Scores

Now it becomes really challenging to figure out how these numbers are being cooked. I'm wishing that I had the Jersey Jazzman or Mercedes Schneider looking at this because it is a convoluted mess.

But let's start with the conclusion (which I am thinking may be how the researchers managed this as well). The conclusion is that getting teachers to retire early not only doesn't hurt test scores, but actually causes them to rise.

Here are some things we either don't know or will not say:

We don't know exactly who retired, how many years of experience they had, and whether or not they retired early. We assume that teachers retired at age 50 or older and that this means they had at least 15 years of experience. We also assume that because retirement spiked during these years, some significant number of the retirees were early retirees (this would be why we bothered to make obvious point #1 above.)

Illinois schools took standardized tests in 3rd, 6th and 8th grade, so we used those data. We never do say exactly what test we're talking about, including what, exactly, the tests were supposed to be measuring.

We don't know (and we admit as much) what other changes might have accompanied the staff turnovers in schools that lost many retirees. Was curriculum changed, class size altered, rooms moved, ice cream ordered in every Wednesday? We don't know.

We also don't know (and the author's don't admit this) what position the teachers retired from. Did we lose people in the tested grades, the grades immediately before or after? We don't know.

We don't know why the effect occurred. Our best guess is that teachers who are burned out and crappy are most likely to grab the opportunity to get out of Dodge (I'm paraphrasing).

What we do seem to know is that twenty years ago, as average years of experience went down, test scores went up. And we frame our conclusion like this:

But the implication of our results is clear: offering expiring incentives for late-career teachers to retire does not harm student achievement on average.

This is not how you frame an argument in favor of an action; this is how you head off anticipated objections to it. This is somebody saying, "We need to entice older teachers to get out sooner. People will claim that losing older teachers is bad for business. Let's assert that it isn't." It is also possible (this is just my suspicion here, but we seem to be open to WAGs here) that you lowball a finding that you can't explain and don't expect anyone to believe.

Their Policy Implications

Early retirement doesn't help the state if retired teachers are more expensive than employed ones. Make sure you fix that before you start getting people to retire. But if you can do that, it's a win all around.  FHHK would really like that.

My Policy Implications

In the 90s, PA also had an early retirement bill passed by the state (in a rare example of legislative charm, it was named after its proposer, Sen. Robert Mellow, and so teachers who were retired early under that bill were said to have "Mellowed out"). Lots of teachers at my school retired, a large cohort who had entered in the hard-to-find-teachers 60s. Those guys were a completely different breed of teacher. I can't think of much to learn about my colleagues of today from my colleagues then.

What we have is research based on a very narrow sample of data from twenty years ago in which we have to tease out, deduce, and suppose the very details that we're treating as critical factors here. It is only slightly more rigorous research than the research which declares that public education has been going to hell in a handcart ever since the Supremes took prayer out of school.

Nevertheless, pay attention, because I suspect this isn't the last we've heard of this study. I fully expect the FHHKers to trot it out again.

1 comment:

  1. The Illinois Public Employee Retirement System (along with many other states') is in dire straits, so looking to incentives for early retirement for any purpose is a losing proposition from the get-go.