The Displaced Plainsman did a fine job last month of pointing out how businesses don't work the way Referred Law 16 would force schools to work. Permit me to pile on....
Governor Dennis Daugaard's big education reform plan (House Bill 1234, now Referred Law 16 on the November ballot) classifies all teachers according to a state-mandated four-tier evaluation system, then requires each school district to identify and reward its top 20% of teachers.
A Facebook friend blog-bounces me to this Vanity Fair preview of Kurt Eichenwald's report on results of a strikingly similar management initiative at Microsoft:
Eichenwald's conversations reveal that a management system known as "stack ranking"—a program that forces every unit to declare a certain percentage of employees as top performers, good performers, average, and poor—effectively crippled Microsoft's ability to innovate. "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," Eichenwald writes. "If you were on a team of 10 people, you walked in the first day knowing that, no matter how good everyone was, 2 people were going to get a great review, 7 were going to get mediocre reviews, and 1 was going to get a terrible review," says a former software developer. "It leads to employees focusing on competing with each other rather than competing with other companies" ["Microsoft's Downfall: Inside the Executive E-mails and Cannibalistic Culture That Felled a Tech Giant," Vanity Fair, 2012.07.03].
Eichenwald reports that Microsoft developers had a touch-screen e-reader in the works in 1998. Top management (you know, Bill Gates, the corporate genius who thinks he knows how to run schools) rejected the e-reader, because it didn't look like Windows and use the keyboard the way they thought all computing devices should. Apple now makes more on iPhones than Microsoft does on everything.
Supporters of Referred Law 16 say they just want to make schools work the way businesses do. But the management techniques of Referred Law 16 killed innovation and morale at Microsoft. They'll kill innovation and morale in our public schools, too.
Learn the lesson of business: Nix 16!
I'll add that businesses also motivate without merit pay: http://thedisplacedplainsman.blogspot.com/2012/06/this-just-in-testing-merit-pay-and-hb.html
Yes! Love that example! Thank you, LK!
I read the Attorney General Jackley's plain sense wording of 1.2.3.4. It says that Mr. H's school board can opt out of the stacked racking and use another process or even choose not to participate. They can also opt out of eliminating the union tenure-for-life.
Moots the whining, doesn't it?
Ha! You don't have much experience with school boards, do you, Grudz? I don't know a single school board that would opt out of eliminating union tenure-for-life.
South Dakota higher education had the same experience as Microsoft when it was trying to find a way to distribute merit bonuses. Department co-ordinators and deans complained about what an arbitrary, time-consuming, and unproductive task it was to rate faculty. The scheme was advanced to rank department faculty by thirds, with an upper third, a mediocre third, and a bottom third. The message sent to the bottom third was that their efforts were not considered worthy and in the eyes of the administration, they were consigned to and designated as inferior class. The middle third took the message that they were designated as mediocre and would have to live with their status. The upper third found that their cited achievements were put under comparative scrutiny, and department heads, deans, and faculty union officers were besieged with accusations with examples of inferior work that faculty found as they examined their colleagues' performance in teaching and scholarship. Comity and collegiality became nonexistent displaced by suspicion and distrust.
Some filed grievances on the basis that they had been rated on false and faulty information. These grievances led to a study that included examiners from outside the system, one selected by the administration and one by the faculty. The resulting report ended that ranking system. It found that the department heads placed faculty within the ranking system and then selected and contrived information to justify their rankings. All examiners agreed that the process was one of the crudest kind of rationalization, to say nothing of the obvious instances of favoritism and cronyism.
The report stated that the entire system of awarding merit was based upon an assumption of inequality, and that it assumed that humans existed on a scale of worth. It pointed out that there was no provision or attempt to consider the relative strengths in various aspects of work. It stated that most of the people placed in the lowest rank were diligent, effective scholars and teachers for whom the low ranking was undeserved and insupportable.
Merit pay did achieve one thing that the administrations liked. It decimated the faculty union. As part of the contract that authorized evaluations and merit pay, the union lost any credibility it had with faculty. When I first joined, more that 75 percent of the faculty were dues-paying members. When I left less than 10 percent were.
I dropped out when I realized that South Dakota needed to affiliate with a parent group that was much more knowledgeable and conversant with the issues and processes of education, particularly higher education. As a union officer, I found that the AAUP was much more supportive and informative on faculty issues, whereas the NEA and AFT modeled their strategies on industrial unions.
Over the years, merit schemes both inside and outside of education have been tried and failed, as happened with Microsoft. Industrial merit awards come through piece work and simply counting the work product. Intellectual work does not lend itself to bean-counting.
The strongest critic of the South Dakota merit scheme was the person selected by the administration to help study and settle the grievance issues on merit pay. He was a dean from a medium-sized public university. He said the problem with merit pay is that is based upon the premise that humans are created unequal in terms of worth and should be managed accordingly. That premise refutes and undercuts the very premise on which public education is based. The idea of merit pay comes largely from the business community, which sees workers not as citizens, but as labor units which need to be managed according to where they fit into schemes for encouraging productivity and creating wealth. The dominant ideologies of business are in direct conflict and refutation of the premises of our democracy. Liberty, equality, and justice are noisome obstacles to the business world. That is why when most workers enter their workplaces, they leave the trappings of democracy behind and revert to the feudal status of peasants and serfs. Wisconsin has successfully enforced that status on its public workers by taking away their collective bargaining rights.
The ideological conflicts between viewing humankind as deserving of the considerations of equal and as beasts of labor who tend to hold impertinent notions of their own value and worth are apparent in our current political campaign. Few people are educated enough to recognize the philosophies that are in play. That's because we have been operating education on the business model for some time now.
Schools aren't businesses and can't be run like one.
I hate to call a fellow American a liar on the Fourth of July, but Grudz, you're lying again. You know there is no tenure-for-life or guaranteed job for life in South Dakota. You know that even without Referred Law 16, state law says any teacher can be fired at any time for just cause. You know that any teacher who changes jobs starts back at zero and must re-earn continuing-contract.
As for the possibility of opting out of merit pay, the default position that the state chose is to require them to participate in the stack ranking that failed so miserably at Microsoft. If Dennis Daugaard had his druthers, if no school board took any action, he'd have every school district use a system that would make the schools worse. To avoid this failed system, schools make take the extra step of composing a new plan on top of their current educational efforts and submitting it for the approval of a panel mostly beholden to the Governor. To avoid even that unnecessary hassle (remember, your public school isn't broken!), schools must opt out completely and sacrifice the meager funding that represents less than a third of what Governor Daugaard took away from them last year.
It's pretty much like the Affordable Care Act individual mandate: you aren't required to participate, but if you don't we penalize you, and you end up with less money than your neighbors who do what the government tells you to do. You're perfectly fine with that logic, right, Grudz?
Super Sweet, right on! I hope you'll keep managing according to that philosophy.
David, huge indictment. Why do we not learn these lessons? Oh yeah, because our leaders don't want to learn that lesson, because they want to do damage to the education system as a nexus of political opposition. What years did that failed experiment and the decimation of the union take place? Did that merit pay evaluation make any news online? Can we get documents from that "strongest critic" of the South Dakota system?
Mr H, it's our Attorney General's words. I only read them.
I think an effective teacher evaluation programs has to start with each individual teacher, not from some artificial top-down approach. It should be collegial and constructive. The goal should not be to pay or penalize people, but to improve teaching.
It's really a matter of finding the right model, whether it comes from business or elsewhere. The "stack ranking" model is just not appropriate or effective.
I would suggest adopting something like what was done in the 1990s for environmental regulation: businesses were encouraged to obtain a third-party evaluation to assess how they could improve their compliance with environmental law. The idea was the company would not be punished if violations were uncovered during the third-party assessment, provided the business corrected the problems. If the company was being run in compliance with law and regulation, the third-party evaluation provided suggestions for improvement.
What I would suggest is doing a pilot project, where teachers in three or four districts to go through an evaluation process structured to help improve teaching, not to use top-down criteria to punish. Make the whole thing voluntary, but provide some sort of credential at the end. Then districts and local unions can decide how they use that credential to provide a bonus.
Don't we grade our students based on achievement? I seem to remember teachers "grading on the curve". Doesn't that sound similar to merit pay proposed by HB1234?
No, it doesn't sound similar in any way, whatsoever, Michael. Actually, it sounds so dissimilar, I really don't even know what you're talking about. How is merit pay similar to grading on a curve at all?
Also, I'll just add that teachers aren't required to grade on a curve, anyway. It's just that some teachers do it because they think it makes up for their teaching mistakes (it does, in most ways), but it isn't based on achievement, and it isn't school policy.
Mr. Black and Mr. Pay both make great points. If we focus not on just paying people more (what teachers want) and instead turn our glare at doing better teaching, that's a good thing. Then, we grade our teachers on a curve and pay the ones that get As more than the ones that get Cs and then put the Fs on some sort of probation to make them better or send them to a class.
PS: Mr. Sibby, what say you on this God Particle? Does this mean God has fractured?
Cory,
The time frame for South Dakota's attempts a merit pay is between 1984 and 1991. (It's been a long time, and I had to consult the centennial history of NSU to jog my memory.) The initial attempt at merit distribution caused a flood of grievances as did the one five or six years later which divided the faculty into stacks of thirds. The document that summarized the review of the process was suppressed. It cited specific passages from evaluation documents, and although it was for the purpose of demonstrating their faults, faculty did not want them to be public for fear of how they could be construed in the future. The merit distribution did not amount to much so the faculty did not consider it worth fighting over. However, a number of concessions were made: the rank-stacking was abandoned, specific requirements were made in regard to teaching, research, and service and each faculty's work in each component had to be summarized and reviewed by faculty for accuracy, and very specific criteria were applied to how student opinions of instruction were to be used. The evaluation task placed upon department heads and deans was so onerous that they complained, almost unanimously, that it interfered with their own teaching duties and administering their units. Of course, the administration did not want the critical review document circulated either. Those of us involved had to read it in the conference room and take notes for our reference. Some of us thought a summary of the review should be written and put on file, but by this time the faculty were so demoralized that they wanted the whole business to end. Many of those faculty took early retirement or moved on. And many dropped out of the union because they did not think it adequately served their interests. NSU experienced a marked decline in reputation at this time, including a loss of NCATE accreditation, the firing of some deans, and a faculty that isolated itself from the institution as much as possible.
I felt that the failures of of the attempts to apply merit should be recorded and documented so that it would be on file if the concept was ever raised again. And here we are.
As for giving merit on achievement being akin to grading on the curve, I cannot see the parallel. But a quick read of any manual on testing and measurement should dispel any notions that the two are similar. Students are evaluated on very specific parameters of performance and work, which is documented. The ranking for merit, as proposed in HB 1234, is discretionary and contains all the faulty assumptions that almost brought down the institutions when it was first applied in higher education in the mid-1980s. The history of NSU contains a fairly comprehensive account of its first application.
David, I hope we can find some more advocates from higher education who can speak to the failure of that merit-pay/stack-ranking program, identify the parallels between that failure and Referred Law 16, and help us protect K-12 from a similar debacle.
Grudz, you know and I know that AG Jackley sat down with the Governor and crafted that document to suitable wording to be ready to launch ASAP (note: Jackley at least does his work more efficiently than Gant). The facts I lay out are still facts. Now pay attention to reality and not your imagination: we are showing you with numerous examples from the business world that merit pay doesn't work, that it stifles innovation and cooperation, that it will hurt our schools. Come to reality and stop risking our schools with policies that have done clear damage to other institutions. I'm not even asking for more pay on this issue; I'm asking for you to let me do my job!
I worked for the State of South Dakota in Pierre from 1982 to 1988, most of the first Janklow era. We had an evaluation process instituted during that time as well. It was partly merit-based, partly based on longevity, and eventually partly based on catching public workers up to private sector pay for equivalent positions.
My supervisor thought the whole merit pay idea was nuts, and he always gave everyone high scores regardless of the stacking requirement.
Our lab had annual certification requirements anyway, where we had to prove our capability to do lab tests and correctly read the results of standard samples. So, really, our merit scores were based on our certification scores, and since everyone did great, we all got high merit scores.
I do want you to keep doing your job, Mr. H, as I know you do it very very well. I would also like to see you earn more for doing it so well.
Cory, I have the book "Drive" by Daniel Pink if you are interested in reading it.
Michael, I think you should call Governor Daugaard and ask him if he'd like to borrow your copy of Pink's work. Maybe then the Governor would accept that his plan won't work. Imagine if we could get him to admit that, go on TV, and urge everyone to vote the plan down so we can start fresh.
Donald makes the point that good managers know merit pay is nuts. I could live with a regime like the annual certification, where everyone who works hard and well enough can meet the standard. Actually, we already have that. Every five years, we must get re-certified, and every year, we must get rehired. And any one of us, even the oldest veteran teachers, can be let go for poor performance.
Grudz, I appreciate the vote of confidence. Please understand that Referred Law 16 is not the mechanism that will allow me to earn more for the job I'm doing right now. Referred Law 16 will require me to add a bunch of other rigamarole to my job to have only a gambler's chance of getting more pay. If you believe teachers deserve more pay for the work they are doing right now, you call Dennis Daugaard and tell him to put back the $52 million he took away from them last year.
If you would like to read the book Cory, I have it available. I did not offer it to our governor. I offered it to you.
Thank you, Michael. I think I've got the main thesis I need... plus a big stack of other books already on deck. That's I recommended you offer the Governor a read, because he clearly does not get Pink's main thesis and is risking our schools on that misunderstanding. Having the Governor read that book would produce more return on investment. Call him, or maybe make the offer to Chief of Staff Dusty Johnson or spokesman and policy advisor Tony Venhuizen.
http://harvardmagazine.com/2012/01/chetty-and-friedman-release-education-study
'“I think the main message of our study is that standardized-test score impacts can be a useful input into evaluating teachers, but by no means are we saying that test scores are the end-all and be-all of how teachers should be evaluated,†Chetty said on the PBS NewsHour. “We think that they’re one aspect of what should factor into the formula. One would also want to use things like principal evaluations or maybe even student evaluations or other measures of teacher quality. But I think there’s some useful data here that could be very useful in improving teacher quality.â€
'Detractors claim the system is more effective at singling out both the highest- and lowest-performing teachers, but reveals little about the majority of teachers who fall in between—and may unfairly penalize strong teachers whose students do not test well for reasons standardized exams can’t detect.'
Let me mention first that $1,000 per year per standard deviation isn't much. That means most students are going to make within $2,000 of each other, no matter what.
Also, the study fails to take into account how students are divided up. The study does mention that the students are not randomly assigned to teachers. Generally, classes are divided up into how successful the students are, with some teachers getting the very successful, driven, intelligent students, and other teachers getting the students who can't manage to figure out the Pythagorean theorem after 5 years of being taught it.
The standardized tests don't test for that. They don't account for that. The smart students will always improve on their standardized tests, unless their teacher is just horrendously awful. Most of the less intelligent students won't improve at all, or will get worse, regardless of how good the teacher is, because 95% of them stop caring entirely when they realize they can't do it.
Only someone entirely naive about how schools work would assume that good test scores = good teachers, or even that improving test scores = good teachers and nonimproving test scores = bad teachers. It's simply wrong.
What is it you are measuring with a standardized test administered to students? If you are using them to try to derive some useful information about teaching, a standardized test isn't going to give you any information about teaching itself. They are probably more helpful to determine whether the curriculum being used by a school district is covering the subject matter in a way that the test maker thinks is appropriate. But here you have to come to some judgement about whether the test maker's assumptions are appropriate.
Carter has a point here. The way classes are structured matters, but he's wrong about "smart students always improving" on standardized test. Smart students have to be challenged with accelerated material. Otherwise they get bored and tune out, and their scores decline. In this case, it's not really the teachers fault that he or she has to go too slow over material.
I find the line "the test maker thinks is appropriate" as excellently summing up how standardized tests fail. They're not written by teachers, or professors, or any education experts. They're written by businesses, with input from the administration of the school(s) in question.
I can't speak for other subjects, but as I've progressed in math, it's become clearer and clearer that the people who write the standardized tests for math for the school I went to had no idea at all about what math was important, and what math wasn't. They seemingly just choose what they think seems good. There's a lot of stuff I wasn't taught that I should have been, and a lot of other stuff that I was taught that should have taken backseat to more important things.
So, what the test writer thinks is appropriate isn't always appropriate, or even all that useful. So tell me, should we really be evaluating our teachers based on how well their students do on material that may or may not be useful? Should we really evaluate teachers based on how well they prepare students for particular tests?
[Hey, "Sung," you also haven't complied with the comment policy. Quit whining, read the comment policy, and check your e-mail.]
Sung, the study showed a $1,000 per year difference within 1 SD. That means that 68% of students ended up within $1,000 from the mean income (which was something like $25,000/year or something). That isn't a whole lot of money. You can take one of two things from it.
1) The bad teachers' students ended up doing essentially just as well as the good teachers' students. Or...
2) The tests were not a very good predictor of achievement.
Also, the only test they even discussed the contents of was a kindergarten test. Not middle school or high school. So that's a pretty easy thing to figure out. If your 4-year-old can do multiplication, he's probably going to go on to discover how to make faster-than-light travel possible. If he can't figure out how many fingers he has on his left hand because he can't figure out how to count past two, he's probably going to end up sweeping floors.
Show me a study where they carefully examine test scores every year for students from kindergarten through 12th grade, and then tell me how the test scores and later salary were connected. One or two tests doesn't do it for me.