January 27, 2009

Gates Foundation still has a lot to learn about education

The Gates Foundation has squandered a tiny bit of Bill Gate's personal fortune on some very silly ideas on education.

Based on Bill Gate's 2009 Annual Letter it's apparent that trend is going to continue in the near future.

First, Bill sets a silly goal:

Our goal as a nation should be to ensure that 80 percent of our students graduate from high school fully ready to attend college by 2025.


The country's drop-out rate is presently higher than 20%. This means that at least every student who now completes high school must be capable of doing college level work.

Now consider this: Pennsylvania just found out that a third of its students who enrolled in a state school or community college were not prepared for college level work. Those are the kids who thought they knew enough to go to college. How about the ones who knew better to apply in the first place? About half the students know better than to even try.

Bill, just upped the ante big time on NCLB's lofty goals. It's one thing to claim that you'll get 100% of students to a loosely defined standard (i.e., "proficiency") based on the standard made up by each state, measured with a testing instrument devised by each state (the fox guarding the hen house), and with cushy safe-harbor provisions. But, it's another to set a high and easily measurable standard that can easily be verified (Hello remediation rates).

And one more thing, Bill. At least have the good sense to give us a cut point that will now result in a large achievement gap between the races/classes. An 80% cut point will leave us with a 23% achievement gap absent we finding the educational magic that allows the low group to gain relative to the high group. Good luck with that. The cut point has to be at least at 95% to get the achievement gap down to a politically acceptable level.

Next, Bill asks an easily answered rhetorical question:

Unlike scientists developing a vaccine, it is hard to test with scientific certainty what works in schools. If one school’s students do better than another school’s, how do you determine the exact cause?


There are basically two effective ways to do this:

Type 1 approach. The most obvious way would be to ask and check. Because successful applications have been created by design, ask the designer of the successful program what the variables are that the design controls and why. The answers imply tidy, controlled experiments that involve systematic investigation of the designer’s assertions.

Type 2 approach. A related approach would involve fully implementing a successful instructional program and then systematically altering the details of it (one at a time, while trying to maintain the others as they were in the original program). The changes would be correlated with changes in student performance. If no difference results from a change, the dimension that was changed does not function as a variable (at least within the range of variation observed). A manipulation that results in improved student performance identifies a variable that was not well designed by the original program. A change that results in inferior student performance identifies a variable that was designed better by the original program than it was in the modified program. This approach would need clear descriptions of what constituted improved performance. Efficiency is an important variable. If the change resulted in improved performance but required three times the instruction of the original, the rubric for judging efficiency would have to compute the ratio of improved performance over the time to arrive at a reasonable overall judgment of the net “desirability” of the change.


of course, this isn't the way we currently evaluate research. But that's a whole 'nother problem.

Lastly, Bill tells us how his foundation is all ready to jump to the next poorly researched education fad: teacher effectiveness.

It is amazing how big a difference a great teacher makes versus an ineffective one. Research shows that there is only half as much variation in student achievement between schools as there is among classrooms in the same school. If you want your child to get the best education possible, it is actually more important to get him assigned to a great teacher than to a great school.

Whenever I talk to teachers, it is clear that they want to be great, but they need better tools so they can measure their progress and keep improving. So our new strategy focuses on learning why some teachers are so much more effective than others and how best practices can be spread throughout the education system so that the average quality goes up.


Bill's rocking a dead baby with this new teacher effectiveness initiative of his. Five minutes on Google Live Search would have told him that.

here's some free advice, Bill. Next time some education expert claims they know what they are talking about, you need to follow Ronald Reagan's advice -- trust, but verify.

This brings us full circle back to the need for ascertaining what works in education research:

Type 1 approach. The most obvious way would be to ask and check. Because successful applications have been created by design, ask the designer of the successful program what the variables are that the design controls and why. The answers imply tidy, controlled experiments that involve systematic investigation of the designer’s assertions.


Never trust what anyone who in education tells you.

12 comments:

Brett said...

Gates' fundamental flaw (at least one of them) is common to many businesspeople I've seen involved in education. In their business roles, they succeed by setting high-level goals and delegating responsibility for achieving them to competent people. They take the same approach to education, looking for competent people (which they define by years in the field, or letters behind the name) and letting them run. It's pretty easy to predict the results.

But Gates is a particularly egregious case. The man who declared high schools "fundamentally obsolete" for some reason thinks it makes sense to hire former superintendents to lead his reform efforts. Nothing says dramatic change like an insider pick...

KDeRosa said...

Does any other industry have as many charlatans masquerading as experts than education?

Brett said...

"Does any other industry have as many charlatans masquerading as experts than education?"

I've always suspected that some psychics can't really speak with the dead. But that's about it I think.

And actually, education may be a harder gig than being a psychic. At least psychics don't have to deal with someone like Alfie Kohn, criticizing them for demanding that the dead answer their questions. Why can't the dead and the living just coexist without expecting so much from one another, and why can't we leave the dead to pursue their higher-order haunting skills?

Anonymous said...

Well, right now economists and financial wizards are "above proficiency" on the charlatan scale. But arguing about who has more charlatans won't get us anywhere.

Gates would do well to take your advice, Ken. And Brett explains why that won't happen.

The corporate sector has a clear "dependent variable"/"bottom line"--monetary profit.

So too in the medical/health sector. It's transparent when person is sick, when a person dies, and so on. The BMGF has done impressive things in health, but has flopped miserably in education.

Why? Because transparent effects analogous to those in business and medicine have to be invented. And psychologists have been chasing "latent traits" for a hundred years instead of trying to pin matters down.

I give Gates points for taking on Education. To date he's been a very slow learner, would be "below basic proficiency" on a statistical scale. We can only hope he doesn't have a "specific learning disability."

The two methodologies you suggest, Ken are legitimate, but they take a long time and great care, because the independent variables identifying and manipulating the independent variables is complicated and very difficult to effect in a natural situation.

Instruction involves (1)a student(s), (2) an instructor(s) and (3) an instructional product/protocol(s)

Prevailing thought treats the instructor as of paramount importance (Gates swallows this) Any failures are in student, family and societal "defects" And instructional product/protocols are of no importance--they are left to teacher/school "choice." That just won't get the job done.

When an instructional aspiration is operationally defined--I use "read any text composed of words within the individual's spoken vocabulary" as the poster-child example, then it's very easy sort out whether the determinants lie with 1,2,or 3.

The answer is 3, and virtually no on has picked that answer. (DI and present company excepted).

If we told Gates that success in "using computers" was a function of OEM sales and customer "defects." A "qualified user" can make any computer work, we'd get laughed out of the room.

Gates is walking proof that generalized "higher order thinking" "problem solving skills" and "creativity" are figments of the imagination. It's in the background information.

Again Gates would readily concede this in business and medicine. But minds turn to mush when it comes to education. Belief systems are very difficult to change. And when the individual is insensitive to empirical feedback, it's really an uphill job.

Gates didn't visit a random sample of "small high schools" or "charter schools." Had he done so, he'd have very different views.

He didn't even get the views of the kids and teachers in the "small high schools." Many "took the money and ran" Nothing changed and many things that did change were for the worse.


Go figure.

Jason said...

I always find the Gates Foundation amusing because they have something that gives them instant credibility with school districts: liberal amounts of grant money. Schools buy in to whatever nonsense they are peddling (in my case, it was "small schools," which was to look shockingly like middle school teams in high school) in order to get the cash money.

dweir said...

"The profession of teaching will improve (1) in proportion as its members direct their daily work by the scientific spirit and methods, that is by honest open-minded consideration of facts, by freedom from superstitions, fancies or unverified guesses, and (2) in proportion as leaders in education direct their choices of methods by the results of scientific investigation rather than by general opinion" -- (E.L. Thorndike, 1927, The Principles of Teaching.)

Mr. Gates says teachers need tools to measure "their progress." Is this laziness of language or lack of understanding that what we want to measure is not teacher progress, but student achievement.

Mr. Gates has a history of not creating anything new but simply repackaging existing technology with aggressive tactics to obtain market dominance. I'm not surprised he tries to present his initiative as novel.

Mr. McNamar said...

Speaking of researched ideas, I begin teaching Corrective Reading's Decoding program tomorrow. Implementation has been a bit shoddy, but I'll do my best to read the script and snap my fingers well.

Dick Schutz said...

Q:"Is this laziness of language or lack of understanding that what we want to measure is not teacher progress, but student achievement."

A: Both. But with more weight on individual ignorance.

The quote from E.L. Thorndike was right for its day but it's now anachronistic. The Old Man and his colleagues personally did both research AND developed products--which were very widely used.

Over time "science" in EdLand came to be viewed as "research" and the D in R&D has been virtually throttled (DI and present company excepted). Ed Researchers seek to put "research into practice" and just as teachers blame kids for instructional failures, researchers blame teachers for "lack of fidelity."

Go figure.

rightwingprof said...

What's ironic about Bill jumping on the "everybody needs to go to college" wagon is that he's a college dropout, and has probably never looked back.

jh said...

rightwingprof

yes, he dropped out

but to be fair, he had enough college prep credentials to get into and attend Harvard

Tracy W said...

And there's a big difference between being able to do something, but choosing not to do it, and being unable to do something.

Ken said...

This is a terrific thread which, unlike the local papers on such issues, is actually educational dialogue on how to ACTUALLY IMPROVE the system. For that, I want to personally thank all contributors. You have given me the rare opportunity to stop and think about the issue at hand - student learning. I believe Mr. Gates has positive intent here, but, an admitted cliche, we are not dealing with widgets sir. Therefore, study of teacher progress in and of itself is misinformed/misdirected/only a part of the story. We need to have Mr. Gates join us for a data study, an instructional improvement plan based on that data, the development of formative assessments to check that plan, and the construction of a professional development plan based on all of the above. He could then see that this 2-3 year cycle focuses on one thing and one thing only - improving student learning (and the associated opportunities for practice).

THanks again folks