Archives for April 2004

Testing is murder

Improving test scores will lead to school massacres, according to a Washington Post column by Margaret McKenna, president of Lesley University, on the anniversary of Columbine.

Education Gadfly annihilates the argument, citing Dave Cullen’s reporting in Slate on the FBI’s diagnosis of the Columbine killers: Eric Harris was a psychopath who hated everyone; Dylan Klebold was his depressed tool. Neither was a victim of bullying. Justin Torres writes on Gadfly:

As FBI investigators have concluded, there was no “Trench Coat Mafia,” there was no dark history of abuse by jocks and preps, there was no sick social structure at Columbine High that drove Dylan Klebold and Eric Harris to kill. In fact, while Klebold was a sad and troubled loner under the sway of a stronger personality, Harris, we now know, was a textbook psychopath, drunk with a sense of his own grandeur. He had contempt for the lower beings around him and expressed pleasure at the thought of their suffering and death.

Such troubling moral realities — the reality, indeed, of evil — are far too stark for Ms. McKenna, who mouths psychobabble platitudes about “regimentation,” “student alienation,” “anonymity,” and “supporting and developing human beings.” (Platitudes that, by the way, would have driven Eric Harris crazy with rage, and which he was extremely adept at manipulating.) And, like those who think that the jocks bullied Harris and Klebold into murder, she has her own novel theory about what’s to blame for Columbine — none other than George W. Bush and Uncle Sam!

“[S]ome of the most important lessons of Columbine,” she writes, “have been all but forgotten — left behind, so to speak, in no small measure because of another educational development of recent years: the No Child Left Behind Act. As class time becomes more regimented and tight budgets create larger class sizes, schools are becoming environments even less conducive to teachers knowing their students well. [The law] forces communities to focus more on raising test scores than on raising kids.” A fevered McKenna delivers what she apparently imagines to be the coup de grace: “The growing belief that rising test scores alone equate to successful schools is false, and it can breed a deadly complacency. The test scores at Columbine High were among the highest in Colorado.”

The logic is as tortured as the notion is repulsive. Never mind that, as McKenna notes, “statistics on school violence have shown a continuing decline since 1995” or that the reports she cites showing a rise in school bullying are routinely scoffed at by serious social scientists. Never mind that schools, high- or low-achieving, are in fact significantly safer than malls, public parks, private homes, and other places that children congregate. Never mind that, sadly enough, serious school violence, when it occurs, disproportionately does so in low-achieving schools. Never mind that there is not one shred of serious evidence linking school violence to high academic standards, and quite a bit indicating that a focused and serious curriculum can help create a disciplined learning environment.

Indeed, never mind that on every factual point, McKenna is grievously, irretrievably wrong. Let us state her central contention baldly: Those who advocate for high standards and accountability are pursuing a policy that leads, in due course, to mass murder. Such a contention is outside the bounds of civilized discourse; to use a massacre like Columbine as an excuse to score debating points about testing is despicable.

The No Child Left Behind Act, passed in 2001, gets the blame for the Columbine massacre, which occurred in 1999. It’s a powerful law indeed.

Update: Eduwonk agrees with Gadfly, calling the McKenna column “tendentious” twaddle.


How can a state prove veteran teachers are “highly qualified” to teach their subjects? California’s system, which got an F rating from the National Council for Teacher Quality, is dubious, writes Lance Izumi of the Pacific Resarch Institute. He describes the state’s High, Objective, Uniform State Standard of Evaluation (HOUSSE) which requires 100 points for a teacher to be considered qualified.

It’s possible, for example, to get up to half the points a teacher needs simply by having taught classes in the subject field. Thus, a PE teacher teaching math for five years could get half the points necessary to prove that he or she is a highly qualified math teacher.

Worse, a teacher can earn up to 90 points through so-called “leadership and service to the profession in assigned area.” Although the state lists activities that could earn teachers points in this category, such as serving as department chair, it says that, “This list is not exhaustive.” In other words, schools have wide latitude to figure out creative ways to give points to teachers deficient in subject matter.

Teachers can also earn points by having their colleagues observe their work in the classroom, always a subjective enterprise. Points can be given for vague and non-subject-matter related observations such as “Establishing and communicating learning goals for all students.”

It’s possible for a teacher to rate as highly qualified without knowing much about the subject he or she is teaching.

‘Prevailing’ lunacy

A 2001 interpretation of California law prohibits volunteers from clearing stream beds, or doing any other unpaid work on a public project, writes Daniel Weintraub in the Sacramento Bee. Everyone is supposed to get the “prevailing” union wage, whether they want it or not.

The issue first surfaced in Redding, where some college students got class credit instead of pay when they helped a local group clear a brush-choked streambed. A union rep complained, and the state responded by fining the nonprofit group involved for violations of labor laws.

Now it turns out that creek restoration projects up and down the state are on hold because many of them rely on grants that, the state says, require them to pay everyone who works on the project a wage set by state regulators. No volunteering allowed.

And this might not end with streambeds. Redding is also struggling with the labor bureaucracy in its effort to build a new city park, in part with volunteer help. A member of a Sonoma County library advisory board tells me that his group fears local Rotary Club members won’t be allowed to landscape the library’s grounds. And it’s possible that groups such as Habitat for Humanity, which use volunteer labor to build homes for the poor, could be swept up by the same regulation.

In Oakland, a new labor contract pushed up costs for a school rehab from $1.8 million to $2.2 million, write columnists Matier and Ross in the San Francisco Chronicle.


Welcome Eduwonk to the wonderful world of edublogging. The new site, sponsored by the Progressive Policy Institute‘s 21st Century Schools Project, is off to a good start. It rips NY Timesman Michael Winerip’s latest attack on No Child Left Behind.

Under NCLB different states have different accountability plans, different standards, and different rules. Winerip makes easy sport of the differences between states. But what is his solution? A national accountability system applying to all the states? A single national test or national standards? Or maybe we just shouldn’t worry about those pesky subgroups and disaggregated accountability for at-risk kids? He doesn’t say.

At Winerip’s sample school black and special education students are way behind, Eduwonk reports. Is that OK?

Why grades rise

A professor recommends Valen Johnson’s Grade Inflation: A Crisis in College Education. A Duke statistician, Johnson makes the case that students give good evaluations to instructors from whom they expect good grades. “Since these days student evaluations make or break a young faculty member’s chances for tenure, the end result of the ‘consumerism’ model of education is that many young faculty members feel they must “buy” student favor by giving away high marks, and destroying all sensible standards,” writes the professor, who wishes to remain anonymous.

What rhymes with angst?

Hatemonger’s Quarterly has a winner in its Horrible College-Student Poetry contest. Plus some horrible runner-ups. The winning author of “believing the me i’m told” is Michael E. Lopez. Could it be this guy?

One of the runner-up poems reminds me of the favorite genre of Highland Park High students in the late ’60s: The dead-fish-on-the-beach poem. The angst of lost love inevitably was symbolized by a dead fish on the beach, seen by the lonely wanderer. I was an editor of the literary magazine. We vowed never to publish a dead fish poem, and we kept that vow.

My daughter’s taking a poetry writing class this quarter. She says, “Naturally obnoxious people are 10 times more obnoxious when they’re reading their poetry.”

Books for Costa Rica

Project Appollonia is looking for English-language books for K-6 children in Costa Rica. The mailing address is:

Project Apollonia C/O
David Scott Anderson
Grupo Utopia International
1601 NW 97th Avenue
SJO 23432
Miami, FL 33172-2053

Exit exams don’t cause early exits

Graduation exams don’t increase the drop-out rate, according to a study by Jay Greene of the Manhattan Institute. The study also shows that neither reducing class size not increasing funding leads to higher graduation rates. In a New York Sun column on New York’s Regents exams, Greene writes:

First, many of the students who don’t pass exit exams would have failed to graduate anyway. For example, in Florida, home of one of the nation’s most difficult exit exams, state officials estimated that about 40% of the seniors in the class of 2003 who could not pass the state’s exit exam had also not completed the necessary coursework to receive a diploma.

Furthermore, the number of students who truly cannot pass exit exams is probably quite small. An analysis by the Fordham Foundation found that most exit exams actually require surprisingly low levels of proficiency.

In most states students are routinely given second, third, and even seventh chances to pass exit exams before they are finally denied a diploma.

Between each administration of the test students who have failed are provided with extra help specifically designed to get them past the test requirement. Given so many tries, eventually most students who are able to complete the other requirements to graduate also pass the exit exam, even if only by chance.

. . . Exit exams force schools to focus their time and resources on low-achieving students they previously ignored. This improved use of resources causes some students to earn their diplomas who otherwise would have dropped out.

The number of students helped by exit exams appears to balance the number who fail to graduate because they can’t pass, Greene writes.

First in line

Parents are putting unborn children on the waiting list for a spot at the Academy for Academic Excellence, a K-12 charter school in Apple Valley, California.

Age grouping is obsolete

These days, many classes combine students with wildly varying levels of achievement and English fluency; several may have disabilities that affect their learning. Teachers are supposed to reach everyone. If they don’t, the kids who didn’t learn will be passed on anyhow.

Social promotion is caused by an out-of-date allegiance to grouping students by age, writes Dennis Doyle in the LA Times. Better to group students by performance, he writes.

The child who is held back feels diminished and unsuccessful, but the child promoted beyond his ability is sure to be more frustrated than ever. Both sides of the social promotion debate are losers because they take for granted the antique process of age grouping.

As it is, a full chronological year separates the youngest from the oldest student in each grade, and the developmental difference is often much greater. Nothing is more frustrating to both teachers and students than trying to bridge a huge achievement gap within a single classroom.

The solution is genuinely performance-based instructional grouping, a format that schools must master in the 21st century. In performance-based schools, students would be held to high academic standards and would work to achieve them for as long — or as little time — as it took. Indeed, that is the de facto model in high school and college. A student takes Spanish 1 until it’s mastered, then moves on to Spanish 2.

One of the strengths of the Success for All reading program is that students are grouped by their reading performance, not by their age. I visited a school that created a beginning reading group made up of fourth and fifth graders who’d been promoted without learning to read.

Update: In today’s New York Times, Michael Winerip writes about an improving school that groups student by performance for reading classes.

(Principal Eileen) Castle is constantly adjusting to make her school better. She created a 90-minute morning reading period with children assigned to classes by reading ability, rather than by grade. So Vicki Pellegrino’s third-grade reading level class has second, third, fourth and fifth graders. It means a teacher can spend the entire 90 minutes working on the same material with everyone, rather than break her class into three reading ability groups and give each group just 30 minutes of her time.

It’s less common to group students by performance in math, but I know a multi-ethnic school that does that with great success. The principal lives in fear he’ll be accused of tracking students.