If It’s Crappy Enough for Bill…

Vanity Fair this month has a long look at how Microsoft has managed to screw up a lot of things about it’s business over the past decade or more. Having been deeply involved with personal computing since the time the company began1, I found the whole thing fascinating. Your mileage may vary.

However, there is one section of the article that I think should interest anyone involved with public education.

In it the author discusses what she sees as a major cause of Microsoft’s lack of innovation and subsequent decline, a personnel evaluation system used within the company called “stack ranking”.

Every current and former Microsoft employee I interviewed—every one—cited stack ranking as the most destructive process inside of Microsoft, something that drove out untold numbers of employees. The system—also referred to as “the performance model,” “the bell curve,” or just “the employee review”—has, with certain variations over the years, worked like this: every unit was forced to declare a certain percentage of employees as top performers, then good performers, then average, then below average, then poor.

“If you were on a team of 10 people, you walked in the first day knowing that, no matter how good everyone was, two people were going to get a great review, seven were going to get mediocre reviews, and one was going to get a terrible review,” said a former software developer. “It leads to employees focusing on competing with each other rather than competing with other companies.”

Now the CEO who signed off on using this “destructive process” (that would be Bill Gates) is considered a “visionary” leader in the education reform movement. And he and his billion dollar foundation are advocating for several other similarly adversarial assessment programs for teachers such as merit pay and “value added” rankings.

Assessment programs which continue to envision classrooms as discreet spaces sealed off from the rest of the world, and teachers as independent contractors whose work is the only influence on student achievement (aka test scores).

I’m certainly not the first person to make the connection between Microsoft’s stack ranking and Gates pushing the idea that teacher assessment should be a more competitive process. But that point needs to be repeated as often as possible.

The bottom line that Gates and others miss entirely in their efforts to “reform” education is that schools are not businesses and those corporate practices cannot, and should not, be applied to the process of teaching and learning.

Especially an evaluation system that has been a major contributing factor to screwing up what was at one time the most valuable company in the US.

1 Although, in all those many years, I’ve never actually bought a Microsoft product or anything containing one. I do have Windows and Office on my MacBook Pro but those licenses belong to my school system.

Not Much Incentive in This Plan

Speaking of merit pay, our governor is inviting 57 districts in the state (including ours) that “may have difficulty attracting, retaining and rewarding experienced, fully licensed teachers” to participate in something called the Virginia Performance-Pay Incentives initiative.

The invitation comes with a $3 million pot of money, which, if my calculator is working correctly, would work out to about $17,750 for each of the 169 schools listed in the announcement. For one year, with no assurances of any continuing funding.

Strings? Why, of course, the money comes with strings.

Whether such programs succeed hinges largely on the criteria used to evaluate teachers. McDonnell plans to require that districts accepting the merit-pay funding also adopt a newly overhauled teacher evaluation system, driven largely by student performance on the state’s standards of learning tests, often called SOLs.

Fortunately, our superintendent is quoted in the article as essentially telling the governor to keep his small change, recalling the far more expensive experiment in merit pay we tried twenty years ago.  He and some of his compatriots in the area also wonder how the schools were chosen since more than a few in our system have no trouble attracting good candidates for teaching positions. (Me too. It’s an odd collection.)

I wonder if McDonnell or anyone advising him has read any of the recent research showing that pay for performance plans don’t improve learning, even when measured by artificial standards like our SOL tests, and can be detrimental to schools.

Probably not. When it comes to education, Bob is far too busy to do more than repeat the talking points from his morning memo.

Still Not Finding Merit in These Pay Plans

Last fall, the results of the “first scientifically rigorous review of merit pay in the United States” were released and the researchers found the financial incentives “produced no discernible difference in academic performance” (aka test scores).

Now a new, larger study, conducted by a Harvard economist who is responsible for designing some of these schemes, “examines the effects of pay-for-performance in the New York City public schools”.

And what did he find*?

Providing incentives to teachers based on school’s performance on metrics involving student achievement, improvement, and the learning environment did not increase student achievement in any statistically meaningful way. If anything, student achievement declined. [my emphasis]

The impact of teacher incentives on student attendance, behavioral incidences, and alternative achievement outcomes such as predictive state assessments, course grades, Regents exam scores, and high school graduation rates are all negligible. Furthermore, we find no evidence that teacher incentives affect teacher behavior, measured by retention in district or in school, number of personal absences, and teacher responses to the learning environment survey, which partly determined whether a school received the performance bonus.

When it comes to research, especially dealing with human behavior, the results of any one study should not be taken as definitive proof one way or another on the issue being studied.

Two showing the exact same results, however, should at least cause thoughtful people to question their beliefs and assumptions.

Now we just need to find some thoughtful people in leadership positions at the DOE and in Congress. States like Florida could use a few as well.

*Link to pdf of the study results.


It Sounds Good, So Spare Me The Details

On the Bridging Differences blog, Diane Ravitch addresses the idea of merit pay plans in education and makes this simple point on why they consistently fail.

Note that they assume that most people—in this case, teachers—are lazy and need a promise of dollars to be incentivized to get higher scores for their students. It never seems to occur to them that many people are doing their best (think people who play sports, always striving to do their best without any expectation of payment) and continue to do so because of intrinsic rewards or because of an innate desire to serve others.

Over the past eight years, No Child Left Behind has clearly demonstrated that the carrot-and-stick-approach (NCLB is 90% stick) did nothing to improve American education.

Now the carrot side is popular, despite studies showing “performance incentives” don’t work (outside of education as well), and not a few trial runs that crapped out.

Merit pay, like charter schools and standardized testing, is popular with many politicians and other education “experts” because it’s simple, relatively cheap, and easy to implement.

It starts with a premise that is easy to sound bite and seems logical.  Or it seems logical if you don’t think about the concept for more than a minute or so.

In other words, these are education reform proposals tailor made for our shoot from the gut, short attention span, simplistic approach to governing this country.

Not Much Merit In These Pay Plans

In his Class Struggle column this week, Jay Mathews spotlights a study which concludes that districts don’t necessarily need to pay more in order to find and keep good teachers.

They just need to do a better job of selling the idea that teacher pay isn’t all that bad. Especially if you can get two teachers to marry. Or something like that.

A marketing campaign to show students that teachers made more than they thought they made “would induce a 7 percent increase in the number of top-third students entering teaching each year (or an equivalent nationally of 4,000 additional top third students above an estimated baseline of roughly 55,000 who enter today,)” the report [from McKinsey and Co., the giant management consulting firm] said.

Paid training increased the number going into teaching by 11 percent. A 20 percent performance bonus to the top-performing 10 percent of teachers would produce the same 11 percent gain in top-third students.

But providing training costs money, something that is usually the first thing to go when politicians start cutting school budgets.

And that idea of performance bonuses? It lost some credibility this week after the release of findings from the “first scientifically rigorous review of merit pay in the United States” showing that paying big money incentives “produced no discernible difference in academic performance”.

Which, of course, did nothing to slow the Secretary of Education from pushing the concept as a major part of the Race to the Top competition and, also this week, sending $442 million to a bunch of RTTT lottery-winning school districts so they can set up merit pay plans.

So, whatever happened to the idea of only paying for “researched-based” concepts that have been demonstrated to be effective in improving student learning?

You know, the concept that was one of the cornerstones of No Child Left Behind.

And which has been consistently ignored by politicians and education “experts” since long before the law’s inception, going back to W Bush’s “Texas miracle”, which also turned out to be based on gut feelings and no proof.