Follow up letter to Harvard President about Fryer's large scale experiments on inner-city youth

Dear President Faust;

Thank you for your email from Sept. 12. I was relieved to hear that though Harvard may be hosting Roland Fryer’s new institute, you do not endorse the large-scale experiments being carried out on inner-city students to pay them for good test scores and the like.

However, I was disturbed to discover from a recent Washington Post article that Harvard’s name is printed on the checks that DC students receive from participating in this experiment, which appears to be a blatant attempt to exploit the good name of Harvard, and implies a literal endorsement by your institution:

“Others [students] sat quietly and studied the pale green checks with "Harvard University" in boldface across the top. Sixth-grader Kevin Sparrow-Bey, who took in $20, said he was annoyed by the assumption that he and his classmates have to be paid to take school seriously.” (Washington Post, “Delighted -- or Deflated -- by Dollars”, October 18, 2008).

Moreover, today’s Washington Post discusses the research that points to the possibility of a long-term decline in student morale and motivation resulting from such incentives – especially when the money runs out. As DePaul University Professor Ronald Chennault, is quoted as saying,

“…there are ethical issues involved, most of which are experimental and dependent on private funding and local political support. "The potential for harm is, what happens after the incentive no longer exists?" Chennault asked. "Not everything is worth trying." (Incentives Can Make Or Break Students,” November 2, 2008).

I wonder what Harvard’s attitude would be if this were instead a large-scale medical experiment on inner-city students, which earlier research had indicated posed a long term risk to their physical health – and whether you would want your institution’s name associated with it.

Finally, your letter did not respond to the question posed in my earlier email about whether it was appropriate for Professor Fryer to be in charge of evaluating the results of his own experiments – a practice contrary to accepted academic practice, which requires independent evaluation.

Please let me know if you think that 1- Harvard’s name on the checks awarded inner-city students does not imply Harvard’s endorsement of this experiment; 2- whether these experiments, if they are to occur at all, should not be small scale in nature -- rather than applied to half of all middle school students, as they in DC public schools, especially as the research points to a substantial risk of long-term harm; and 3- whether you think its appropriate that Prof. Fryer should evaluate the results of his own experiments.

Yours sincerely,

Leonie Haimson
Executive Director, Class Size Matters

Our Children--Only Pawns in Their Experimental Game

"We will have the willingness to try new things and be wrong — the type of humbleness to say, I have no idea whether this will work, but I’m going to try." --Dr. Roland Fryer; 9/24/08

Perhaps discouraged by the refusal of NYC children to respond to financial incentives by actually performing better as opposed to just taking more tests, Dr. Fryer is returning to Boston to head something called the "Educational Innovation Laboratory" (see the splashy EdLabs website).

Dr. Fryer laments that billions are spent researching drugs and developing airplanes, while little is spent “to scientifically test educational theories.” Thus his friend, Eli Broad, (see picture above) and the Broad Foundation are helping him with the first $6 million of a $44 million, 3-year, “research and development initiative” that will have EdLabs “partner” with NYC’s Department of Education, the Chicago Public Schools, and the District of Columbia Public Schools.

What does this “partnership” mean? EdLabs will “connect” top academics from various fields with its own “R&D teams that will be embedded in these three school districts.” (emphasis added). There, the EdLabs folks will “foster innovation and objective measurement of the effectiveness of urban K-12 school district programs and practices” and “quantify the expected "student return from an investment" (sic.) to help leaders direct their limited resources into high-return programs and initiatives.”

In other words, the cheerleader-in-chief for market-oriented education strategies will evaluate the results of programs devised by ideologically aligned education officials, his own teams or even himself (such as the preposterous scheme to reward student performance with cell phones, which apparently has collapsed.). This passes as “rigorous research.”

When a drug company funds research to study the safety and efficacy of its own product, we have no difficulty understanding that conflict of interest is a problem and means the results are suspect. Imagine what credibility a drug study would have if the research team actually included drug company personnel! And would anyone even entertain the suggestion that the head of Philip Morris USA’s Youth Smoking Prevention Program should be included in any study of teenage smoking?

The incestuous relationships in this new initiative would not be tolerated in a scientific study involving drugs or other products. That the proposal is made with a straight face by people who are smart enough to know better shows the utter contempt they have for our children. This is fundamentally a business enterprise, not a serious attempt to evaluate educational strategies by standards that are applied to scientific research. Calling it a “lab” and putting it at Harvard doesn’t cleanse it of this taint.

And here’s the kicker for all us parents and taxpayers. The Broad Foundation is committing a mere $6 million in “jumpstart” funds--where do you suppose the other $38 million will come from? Need I say it? EdLabs’ sources of support include “the three participating school districts.”

-- Paola de Kock

the response to our letter from the President of Harvard

In response to my letter to Harvard and other foundations about their support for the controversial large-scale experiments in NYC and DC, being carried out by Roland Fryer, to pay students for test scores, good behavior and the like, I received the following email today:

From: Drew Faust [mailto:president@harvard.edu]
Sent: Friday, September 12, 2008 10:24 AM
To: Leonie Haimson
Subject: RE: re paying for high test scores vs. reducing class size

Dear Ms. Haimson,

Thank you very much for your email and for taking the time to share your concerns with me. I appreciate your candor. I must tell you, however, that academic freedom on university campuses, which serves us all well, includes the freedom to express controversial views, with which others may disagree. The views held by Professor Fryer are his own and do not necessarily reflect the views of Harvard University.

With my best wishes,

Drew Faust

What do you think, folks? Does this response invoking academic freedom get Harvard off the hook? And what about allowing Fryer to evaluate his own experiment and not requiring an independent assessment of the results -- which is contrary to accepted practices and was not mentioned in her reply?

An open letter to Harvard's President about its support for large-scale experiments on urban public school students

See this Washington Post article about the large scale experiment that will pay 3,000 middle school students up to $100 per month for good attendance, behavior and grades; this experiment is being partially funded by Harvard Univ. and is directed by Prof. Roland Fryer, who is carrying out a similar experiment in the NYC public schools.

Cc: The National Science, Kaplan, Smith Richardson and Broad foundations.

Dear President Gilpin Faust:

I applaud your efforts to reduce class size, which according to the AP, led to Harvard regaining the top spot in the recent US News and World report. According to US News, 75% of Harvard’s undergraduate classes now have fewer than 20 students

At the same time, I want to protest Harvard’s participation, and that of the other foundations copied on this email, in financing the large-scale experiments in Washington DC, New York City and elsewhere that will pay cash rewards to high-needs public school students for high test scores. An article in yesterday’s Washington Post reported that approximately half the cost of a new $2.7 million experiment in DC schools is being covered by Harvard’s American Inequity Lab.

Roland Fryer, the Harvard professor and author of this experiment, as well as the experiment in NYC schools which gives up to $500 to middle school students who have high test scores and provides them with free minutes on their cellphones, claims that “Surveys of students and parents show support for the concept.”

To the contrary, our survey of over 1,000 NYC parents showed that over 70% strongly opposed paying students for good scores. Another survey done by EdWeek showed that an overwhelming majority (81%) of respondents were against schools offering cash rewards to students.

These views would matter less if the research indicated that such programs were likely to be successful. A similar NYC program that paid $2 million to reward students for high AP scores led to fewer students actually passing the exam. Numerous studies show that in the long run, cash rewards undermine the intrinsic satisfaction that otherwise results from positive behavior. This particular scheme is also likely to lead to increased economic disparities and resentment between those students who would do well in any case, and others who simply need more academic help and support.

Moreover, there are far more effective strategies to enhance student engagement and learning, particularly among low-achievers and in high-need schools. Many studies show that providing smaller classes narrows the achievement gap and creates more student engagement and focused learning in the middle and upper grades. See this recent study by Thomas Dee of Swarthmore and Martin West of Harvard, showing that smaller classes in 8^th grade are associated with significantly higher levels of student engagement and eventual earnings, with the expected benefit from reducing class size in urban schools nearly twice the estimated cost.

Prof. Peter Blatchford also recently released a detailed observational report, showing that when secondary students are place in smaller classes, much greater time is spent “on task” and focused on learning, with special benefits for low-achievers and twice as much negative behavior per student exhibited in large classes than in small.

Clearly, Harvard believes in the importance of smaller classes for its own students, having devoted considerable resources to reducing class size, and limiting the size of freshman seminars to 12 students or less.

Yet despite the abundant evidence, urban and minority students tend to be placed classes much larger than this. Indeed, more than 70% of middle school students in NYC are in classes of 26 students or more, and about 40% of eighth graders crammed into classes of thirty or more.

I urge Harvard, the National Science Foundation, and the other foundations that are supporting these large-scale experiments to instead shift their considerable resources towards research on the multiple benefits of smaller classes, and towards the effort to provide the same sort of individualized attention to public schools students that are currently enjoyed by students at our more elite private institutions.

I also hope that you make sure that the results of any experiments you help finance are fully evaluated by a completely independent investigator, not by the author of the experiment himself, and examine the long run as well as the short run effects.

Leonie Haimson
Executive Director
Class Size Matters

Please send your own letters to Harvard’s President and the other participating funders at the following addresses: president@harvard.edu; msantona@nsf.gov; info@kaplanedfoundation.org; jhollings@srf.org; dk@broadfoundation.org

Wrap-up of commentary on the teacher incentive pay proposal

Today, lots of incisive commentary about DOE’s plan to pay teachers at schools with low-performing schools that improve on test scores. In the Daily News, Diane Ravitch questions the wisdom of this proposal, writing:

But will merit pay fix our schools? The quick answer is no. There is no evidence that students learn more because their teachers get differential pay tied to their students' test scores.

The assumption behind merit pay is that teachers are not working hard enough. Pay them more if their students get higher scores, say corporate-style school reformers, and they'll work harder. Politicians in Washington and many state capitols in both parties are attracted to the idea.

Diane goes on to argue that this particular proposal is not really merit pay, since schoolwide improvement will be rewarded, rather than the achievement of individual teachers, and then points out:

Most teachers understand that the tests now in use are imperfect measures of children's learning, which may be influenced by all sorts of external events in their lives - such as illness, disruption in their home life, emotional distress, financial worries and overcrowded classes. In effect, in its purest form, merit pay rewards individual teachers based on a single test score without regard to issues that are entirely beyond their control.

But if within every school, conditions for teachers may vary, they differ far more between schools, especially as regards class size and overcrowding.

One example: In District 1 schools on the Lower East Side, the average class size in grades K-3 is 17.6; in District 6 in Washington Heights, there is an average of 22.3 students per class. The difference is just as glaring in 4-8th grades, where class sizes in D1 average 20.4 compared to 25.8 in D6.

Even within District 6, class sizes range widely – and in some individual schools classes rise to 30 or more. Both districts have large numbers of poor and ELL students, but as a result of these far more difficult conditions, District 6 has loads of schools on the failing lists – when District 1 has very few.

So how can any reward system be fair that does not take into account the easier time teachers on the Lower East Side have reaching struggling students in classes of 18 or less – especially as compared to those in Upper Manhattan, working with the same high-needs children but teaching classes of 26, 28 or more students at a time?

Some proponents of this reward system may argue that the measures to determine which schools will receive cash rewards will be determined not only by test scores, but by survey results and attendance as well, but these factors are surely as negatively affected by overcrowding; moreover, like the school grades, in which these other factors account for only 15% of a school’s grade, I predict this will be nothing more than a fig leaf.

See also Eduwonkette, who addresses Klein’s argument that pay for performance is a routine practice in other professions and the business world. Instead, in most other fields, performance there are “holistic” evaluations – not based on numerical outputs alone (which in this case would be test scores.) This is true even in medicine:

Physicians I talked to in preparing this post laughed at me when I asked if their performance bonuses were based on patient outcomes. The most common response was that those outcomes were largely out of their control, so their hospitals rewarded them based on their inputs - i.e. hours, procedures, and revenues.

Of course, test scores are even more out of the control of teachers, unless they devote all their time to test prep or resort to cheating – neither of which would be considered desirable. Eduwonkette discusses other reasons that this proposal fails to pass the smell test:

...they do not control for differences in structures over which the school has little control - i.e. class size and large intradistrict differences in per-pupil expenditures - which are relevant for plans that plan to compare performance across schools, not only within a given school.

The administration argues at great length that their reforms have worked to improve student achievement – as in, for example, the small schools initiative. Let’s take this claim as true, for argument’s sake – though I would argue that the possible gains achieved are due more to the smaller classes at these schools rather than their smaller enrollments.

But whatever the reason, if one believes that small schools have inherent value, as Tweed does, how can one then also claim that a reward system that withholds bonuses to teachers who work at large schools could be evenhanded?

Last but not least, see Barry Schwartz' excellent oped in the NY Times. Schwartz is a psychology professor at Swarthmore who wrote an earlier, equally convincing oped in July about Roland Fryer’s experiment to pay kids for good test scores.

This new oped takes off from Joe Torre’s decision to leave the Yankees, and his rejection of the hefty bonus Steinbrenner offered if the team won the World Series next year. Clearly, Torre was insulted by the suggestion that he wasn’t already working as hard as he could to go the distance. Schwartz goes on to write:

If teachers are thwarted by their working conditions, then we need to fix the conditions, and not try to paper over them with bonuses. There are settings in which bonuses may make sense — if the work offers no opportunity to find satisfaction, for instance, or if it really is all about the money. And yes, there should be public acknowledgment of extraordinary performance. But that acknowledgment needn’t be financial, and it certainly shouldn’t be contractual.

The more society embraces the idea that nobody will do anything right unless it pays, the more true it will become that nobody does anything right unless it pays. And this is no way to run a ballclub, a school system, or a country.

Joel Klein on the Colbert Report

Colbert was very funny on Wednesday night's show-- the Chancellor not-so-much. See Colbert commenting on the administration's controversial program to pay kids for high test scores:

"Now, I love this idea of paying the students because, what it does is, it brings free-market forces to knowledge. If you score in seventh grade and make $500, you can invest that money in nerds to write term papers for you when you get to eighth grade.”

And: “As long as you’re going to be paying kids and making it seem like a job, why not just bring back child labor? Because, I’ve got to tell you, you give those 10-year-olds a couple of shifts in my textile factory, you know, a couple of fingers missing, they will hit the books hard and achieve.”

Klein made the following dubious claim: “There have been programs where they pay kids for abstinence and it’s worked.”

In fact, none of the abstinence programs have been proved effective.

Klein's comments about poor children

See this NY Times column about the controversy over the administration's plan to pay poor kids for higher test scores, covered previously in this blog. Only now the Chancellor apparently has a new rationale for this experiment:

“... Schools Chancellor Joel I. Klein esponds to skeptics by arguing that no one has figured out how to get more poorer children engaged in learning. Trumpeting the long-term benefits of education, the better jobs and lives well lived has not worked. Cash just might.”

No one has figured out how to get poor children engaged in learning? Perhaps he might try improving classroom conditions.

The truth is that many experts have indeed figured out how to achieve this. Increased access to preK and smaller classes are two, proven programs that research has repeatedly shown results in more engagement and learning, especially for poor and minority children . Unfortunately, DOE has shown little interest in providing either option; in particular, our classes remain the largest in the state and among the largest in the nation.

The truth is that if some students are disinvested in the learning process it is because the system has not invested in them; their lack of caring – to the extent that it exists -- results directly from the fact that the people who run our schools do not care sufficiently about them.

Paying for test scores: Anti-social, bone-headed perversity?

Check out the scathing critique in the Huffington Post by Diane Ravitch, contributor to this blog, of the Mayor's proposal to pay students for getting library cards and good test scores, as well as their parents each time they show up for parent-teacher conferences. An excerpt:

It demeans the poor parents who do meet their children's teachers; who do have library cards; who do care desperately about their children's schooling. And it insults the kids who are trying their best but having trouble because New York City has the most overcrowded classrooms in the state of New York.... The pay-for-behavior plan is anti-democratic, anti-civic, anti-intellectual, and anti-social.

There have also been negative columns in the New York Post from the Manhattan Institute's Nicole Gelinas here (called "Mayor Mike's Poverty Perversity") and from Andrea Peyser here, who writes that it is the " the most insulting, bone-headed plan ever cooked up. "

We now have an unusual consensus of the Huffington and NY Posts, which rarely agree on anything, that this proposal is morally repugnant. Too bad our Mayor doesn't appear to have the same scruples.

Paying for test scores

Harvard Professor Roland Fryer who intends to experiment on our students by giving them cash awards if they score high enough on their interim assessments is going to become "chief equality officer" at Tweed, according to the NY Times.

So much for these supposedly "no-stakes" exams.

The results of this experiment will be monitored by Fryer.

So much for independent evaluations.

The Mayor apparently loves the idea of paying kids cold hard cash for performance, just as much as he approves of the "pay for play" arrangement in his controversial Randall's island deal, now in court, in which exclusive rights to most of the fields in a public park would be granted to private school students for the next twenty years.

But what if you're a conscientious student and you are trying hard, but still can't make a perfect score? What about kids with disabilities, or ELL students?

Too bad for you.

Question: can anyone tell me why this experiment is likely to lead to more equality, rather than even more disparity between high and low achieving families and kids?

This and other similar experiments (in pdf) involving cash incentives to kids for high scores are being funded by the Rockefeller Foundation, Robin Hood, the Open Society Institute (George Soros' foundation), and the insurance company AIG, as well as the Mayor himself.

Students who pass all five Regents would be paid $3,000. If the program is seen as "working" and is expanded, it could cost the city "hundreds of millions" a year, according to Deputy Mayor Linda Gibbs in the NY Post.

DOE still mum about cash for kids who ace their "no-stakes" assessments

In his Daily News column on Thursday, Juan Gonzalez broke the story about the proposal to give cash awards to students in the empowerment schools who ace the new interim assessments. (Here is an earlier blog entry about these $30 million assessments to be given four to five times a year.)

More than a week ago, we posted a letter to principals from Roland Fryer, the 30 year old Harvard professor leading the project.

Students will be paid five to ten dollars for taking these tests -- and 25 to 50 dollars if they score a perfect score. Every empowerment school that buys into the project will also be awarded a $5,000 cash "gift."

This proposal severely undermines the official DOE claim that these are “no-stakes tests” See this headline on the Tweed press release from June 1 for example: "Customizable, No-Stakes Assessments to Give Teachers, Principals, and Parents Timely Feedback on Student Progress"

Or this, from the DOE website: "These assessments are solely tools for teaching and learning. There are no stakes attached to the results for schools, principals, teachers or students."

The new project was inspired by the Mayor’s poverty initiative. Bloomberg established a commission on eradicating poverty, and then ignored most of their recommendations and instead announced he would implement experiments to pay parents and students cash if they attend school regularly and get good results.

Of course, if the kids who score well are not poor, they will presumably get receive these awards anyway – so this proposal might even enhance economic disparities among students at the school level.

In the NY Times today, Ernest Logan, the president of the principals’ union, said that “We are troubled by additional pressure being placed on children to achieve perfection,” he said. “What really matters in education is continued student progress, not perfect test scores.”

Apparently, the project has not been formally approved by DOE, despite the fact that the Fryer has already held sessions with principals and signed up a number of schools. Eric Nadelstern, head of the empowerment zone, endorses the project, according to the NY Times:

“He has my enthusiastic support,” Mr. Nadelstern said in an email to principals.“I encourage you to give the program serious consideration.”

Bloomberg also likes the proposal: "If we aren't looking at everything, shame on us,” he said at a news conference yesterday, “I hope there are people thinking about how we would implement that and every other idea." (Unfortunately, they still aren’t thinking seriously about how to reduce class size – despite a new state law that requires them to do so.)

According to NY1, late Friday DOE issued a statement: "Performance-based incentives are one element in a range of strategies we are considering. We are still at a preliminary stage in considering how to implement such a program."

Does this reluctance to endorse the project signal a split within the administration?

Check the negative commentary from New York magazine, in a piece entitled "Mike Bloomberg wants to bribe your kids," the blog Ed in the Apple, and a column criticizing a similar program of cash incentives in Florida here:

“Gov. Charlie Crist was correct when, as the state's education commissioner, he said that the prospect of cash rewards for students who score well on FCAT made him “a little queasy.”

We spend loads of time and effort telling kids what's right about learning, that education is something inherently good and valuable. By using cash as an incentive for students to do well on state tests, schools act hypocritically, and they may even be undermining their core missions.”