Erase to the Top

by Robert Pondiscio
March 28th, 2011

“On the 2009 reading test, for example, seventh-graders in one Noyes classroom averaged 12.7 wrong-to-right erasures per student on answer sheets; the average for seventh-graders in all D.C. schools on that test was less than 1. The odds are better for winning the Powerball grand prize than having that many erasures by chance, according to statisticians consulted by USA TODAY.”

A USA Today investigative piece looks at high erasure rates on standardized tests at Washington, DC’s Crosby S. Noyes Education Campus, which went from a school in need to one of DC’s ‘shining stars.’”  The report notes that three years ago, DC’s Office of the State Superintendent of Education asked test-maker McGraw-Hill to do “erasure analysis” after some schools showed big gains in in proficiency rates on April 2008 tests.  “Among 96 schools flagged for wrong-to-right erasures were eight of the 10 campuses where [DC Superintendent Michelle] Rhee handed out so-called TEAM awards ‘to recognize, reward and retain high-performing educators and support staff,’ as the district’s website says. Noyes was one of these.”

Standardized Testing is Not the Devil…

by Robert Pondiscio
March 22nd, 2011

…Test prep is the devil. 

Via Alexander Russo comes word of a “misguided war” against standardized testing, and a backlash against the backlash. “Standardized testing is rarely fun — and it could almost certainly be improved — but it’s not nearly as antithetical to real, deep learning as its detractors suggest,” writes Anna North at the blog Jezebel, who scoffs at well-off parents refusing to let their kids sit for tests. Such protests

“…. run the risk of deepening the divide between haves and have-nots that continues to plague public education — and pretty much every other aspect of society. Any attempt to scuttle standardized testing needs to acknowledge that even if the tests are problematic, the deficits they attempt to address are real — and any alternative approach needs to face these deficits, not just walk away from them.”

North slightly misdiagnoses the issue.  Personally, I have no problem with tests, per se.  But you’d have to be naive to dismiss the impact preparing for those tests have had on the children North and everyone else purports to care so deeply about.  Talk to someone who has taught in a low-performing school and you’ll almost certainly hear stories about prodigious amounts of time sacrificed on the altar of practice tests and language arts lessons in “test sophistication.”  At my South Bronx elementary school, we had a Teachers College consultant who encouraged us to ”teach tests as a genre of literature.”  But even that pales in comparison to a grad student of mine who was mandated to spend two hours per day on test prep from the first day of school.  

Testing and accountability are unlikely to disappear.  Boycott the test?  Perhaps, but if I were a parent activist, I would march into the school office the first day of school with the following bargain:  “I’m sure you agree the best test prep is great teaching and a robust curriculum, Ms. Principal.  So let’s keep our focus right there.  Don’t worry about spending my child’s time and your budget dollars on test prep materials.  Because if they show up in our kids classrooms, we can promise our kids won’t be showing up for the test.”

Our Love/Hate Relationship With “Mere Facts”

by Robert Pondiscio
March 22nd, 2011

The Daily Beast serves up that hardiest perennial of “tsk, tsk” journalism: a poll highlighting our collective lack of history and civic knowledge.   The U.S. citizenship test is comprised of 100 questions about American government, systems of government, rights and responsibilities, American history and civics, notes the Beast.  “Ten questions from the 100 are chosen randomly for the test-taker.  To pass, one must get at least six right.”  About four in ten Americans can’t clear the bar we set for would-be naturalized citizens.

Tsk, tsk. 

The essential conundrum.  We in education blithely dismiss background knowledge as trivia and “mere facts,” but we (and more importantly, the broader world) continue to judge harshly those not in possession of facts we take for granted.  Take the test yourself.  The questions are of the kind every school child used to know back when school kids used to know things.  And to be fair, some of the questions are trivia.  The ability to name the authors of the Federalist Papers, for example, is probably not as important as understanding something about the role of the papers in the ratification of the Constitution.  But the unspoken question to ask yourself is whether it would impact your opinion about a friend, neighbor or colleague if they couldn’t answer the questions.

Perhaps we should change the immigration test to a DBQ format.  Or perhaps insist on naturalization by portfolio assessment. 

(H/T Joanne Jacobs)

A Car Without Wheels

by Robert Pondiscio
March 18th, 2011

At Flypaper, Kathleen Porter-Magee responds to my post on common curriculum.  She largely concedes my point that being prescriptive on what kids should learn is neither a resource issue or a call to pick winners and losers among publishers.  But I’m still confused as to why she objects to curricular content guidelines.  Porter-Magee places her faith in standards and accountability.  But as we have discussed ad nauseum, standards are not particularly helpful in planning what to teach.  And the reading test has yet to be created that isn’t biased by its passage selection.  Thus the only fair and equitable way to meet standards and create fair assessments is with comprehensive curriculum guidance that teachers can use to guide lessons and test makers can use to create assessments.

A reliance on standards and accountability in the absence of clear curricular content guidelines is always going to be a bit of a fool’s errand.  If you insist that students should be able to effectively compare and contrast informational texts you need to specify content to some degree, since cognitive processes are largely domain specific.  It’s relatively easy to find the main idea when reading about familiar topics, but hard when the subject matter is new or unfamiliar.  If you’ve been following news accounts of Japan’s efforts to prevent a nuclear meltdown, for example, you’ve probably found it easier to do so as the days go on and you become more comfortable with terms like “millisieverts” and your background knowledge increased.  That’s not a coincidence.  Reading test passages on unfamiliar subjects are inherently unfair to those who lack background knowledge, which aids comprehension dramatically.  Refusing to describe a common body of knowledge that all children are expected to have leaves this problem unaddressed. 

In short, standards without content is a car without wheels.  The standards to which you build such a car do not matter.  You ain’t goin’ anywhere.   The skills demanded by the standards cannot be taught in the abstract, except at the most superficial level, which is why the authors of the Common Core State Standards explicitly called for a coherent curriculum. 

“In the end, most people agree that schools need strong, content-rich curriculum,” Porter-Magee writes bizarrely.  If this were true, I’d be unemployed.  And really, really happy.  The dominant approach to ELA instruction in U.S. elementary schools is the skills-and-strategy approach, which assumes that reading comprehension is an all-purpose, transferable skill that can be applied to any piece of text with equal proficiency.  It’s not, which makes another of Kathleen’s assertions equally baffling – and troubling:

“I wholeheartedly agree that a content-rich curriculum is essential. But I sincerely believe that if states get the standards and accountability pieces right, then schools will have no choice but to follow a content rich curriculum.”

It is more accurate to say that you can’t get standards and accountability right without following a content-rich curriculum.  What am I missing?  We have language arts standards that refuse to say what should be taught and learned, followed by tests to ensure it has been taught and learned.  This can only be described as Kafkaesque. Unless everyone taking a reading test has the same general store of background knowledge, the test will discriminate against those who don’t have it.  Thus the biggest losers, as always, will be those students—and their teachers—who are flying blind.  More advantaged students, who typically come to school with deeper, richer language skills and background knowledge will continue to outperform low-income and minority students. 

Faith in standards and accountability alone as levers of change implies a faith that disembodied skills can be taught and measured in the absence of a coherent curriculum.  I see no reason to believe this can be done.  I see mountains of evidence to suggest it cannot.

I’ve said it before.  If you’re placing your faith for improving outcomes on standards, accountability, or teacher quality a core curriculum is your ally, not your enemy.  Without it your shiny car will remain on the blocks and in the driveway. 

Happy motoring.

Report: Sky Not Falling

by Robert Pondiscio
March 15th, 2011

True or False:

1. The United States produces many more high-achieving students than any other OECD nation.

2. In both reading and math, the U.S. produces more high achievers than France, Germany, and the United Kingdom combined.

3. In both reading and math, in raw numbers, the United States produces more high-achieving Hispanic students than Asian students.

4. There are more high-achieving African-American students than high-achieving Finns.

All true, according to an interesting new paper  American Achievement in International Perspective by Fordham’s Mike Petrilli and Janie Scull.   The moral of the story?  Size matters and the large number of U.S. students ensures a high number of high achievers across the board.  “In raw numbers, at least, our high-achieving Hispanic and black American students outnumber the high achievers of several other countries,” Petrilli observes at Fordham’s Flypaper blog.  “At the least, this indicates that they will have a seat at the international table—on prestigious college campuses, in the board room, and in the laboratory. It’s a start.”

Pascal’s Wager on Curriculum

by Robert Pondiscio
March 14th, 2011

“Stop seeking curricular solutions to instructional problems,” urges Kathleen Porter-Magee at Fordham’s Flypaper blog.   Entering the fray over last week’s Call for Common Content, Porter-Magee’s says curriculum is essential, however, assessment and accountability matter more. 

Unfortunately her piece is a bit of a strawman-fest.  She confuses the core curriculum manifesto’s call for guidance on what students should learn with a call to pick winners and losers among published curricula, or prescribe the methods by which children should be taught.  The Call for Common Content is merely a sensible proposal to describe the common, knowledge-building content that all children must have in order to be fully literate. 

A case can be made that a content-rich approach to the critical elementary school years is the educational equivalent of Pascal’s Wager.  The French mathematician famously argued that even nonbelievers should live as if they have religious faith.  Why?  If there is a God, your potential upside is eternal life vs. damnation.  You win.  And if God doesn’t exist you’re dead anyway.  You have everything to gain and nothing to lose.

The former head of curriculum and professional development at Achievement First, Porter-Magee describes a misstep  in mandating Saxon Math at her schools.  “By focusing our energies on convincing teachers and principals to use a particular curriculum, we were, on some level, taking ownership over student achievement results and shouldering it ourselves,”  she recalls.  Note that Porter-Magee and her colleagues didn’t throw up their hands and decide to stop teaching multiplication, fractions and geometry.  They simply rallied around a different program.  The challenge for educators ought to be the best way to teach material to their students–not to decide whether to teach it at all. 

No one is talking about mandating specific programs, pedagogical approaches or delivery systems.  The call for a common curriculum is a call for a unified scope and sequence, nothing more.  It takes seriously the essential idea that what schools teach is critical and ought not, for reasons of fairness and equity, be left to chance.  Betting on coherent accumulation of knowledge is the safest wager and one with no conceivable downside.  As Dan Willingham has pointed out knowledge grows exponentially.  “Those with a rich base of factual knowledge find it easier to learn more — the rich get richer. In addition, factual knowledge enhances cognitive processes like problem solving and reasoning,” he writes.   Pascal 1, Porter-Magee 0.

She also argues that a curricular focus runs the risk of “distracting states from allocating their now very scarce resources towards policies that have the potential to much more dramatically impact student achievement.”  Wrong again. As Russ Whitehurst has pointed out, curriculum is a ”free good.”  Something is going to get taught, and there are no discounts for bad or ineffective curricula; the implementation costs are essentially fixed.  Thus a coherent, content-rich approach to curriculum costs the same as an inferior content-neutral approach.  Why bet on incoherence?  Pascal 2, Porter-Magee 0.

“States would do better to create or adopt rigorous assessments and a strong state accountability system, and then to devolve ownership over student achievement results—and that includes curricular decisions—as closely as possible to the classroom,” Porter-Magee asserts.  However, this overlooks the inconvenient truth that reading tests are de facto knowledge tests (“poor readers” outperform “good readers” when the topic of the reading test is familiar to the ostensibly poor readers) and at present are utterly disconnected from curriculum.   The correlation between accumulated knowledge and reading comprehension makes it irresponsible not to have some manner of content guidelines in place, at least at the district or state level.  Cumulative buildup of enabling knowledge literally cannot happen if curricular content decisions are left to chance and whim.   And as always, the ones who disproportionately suffer from a hands-off view of curriculum are those who can least afford gaps in their knowledge base.  

Worst of all, Porter-Magee implies that one must choose between improving teacher quality, accountability and curriculum.   Rather, they are mutually reinforcing–each is more likely to succeed supported by the others.  In fact, teacher quality advocates have the most to gain from a common curriculum.  Reading tests, as currently conceived, are poor vehicles for measuring teacher effectiveness, since they are not curriculum-based.  There is simply no correlation between what the teacher teaches in a given year and the reading passages on a typical state reading test.  A common core curriculum will make it much easier to measure teacher effectiveness. 

Pascal breaks the game open.

Unlike Pascal’s Wager, those who bet on the curriculum wager are already way ahead.  There’s no empirical proof of God’s existence, but there’s a mountain of data to support the idea that teaching content is teaching reading.  It costs exactly the same as a content-neutral approach, there is no conflict whatsoever with structural reforms, be they teacher quality, accountability, school type or management.  

It’s a very smart bet.

Ed Reformers for Illiteracy

by Robert Pondiscio
March 8th, 2011

“A common curriculum (whatever that means) is the wrong idea when we’re about ready to develop school of one–not just a 6th grade math program, but fully customized engaging learning sequences for every student,” writes Tom Vander Ark.   His post at EdReformer.com is in response to yesterday’s call signed by 250 educators, civic and business leaders for a common core curriculum. 

There are fewer ideas more seductive than the vision of customized education, where all children remain blissfully engaged solely by the ideas and subjects that interest them, and soar to ever-higher standards on tech-driven wings.  But this splendid vision ignores an inconvenient truth:  all of our most cherished goals for education are a function of the knowledge we possess and have in common with others.  To say that a common curriculum is the wrong idea is to say literacy is the wrong idea.  Let me not mince words:  If you don’t think  a common body of knowledge is important for all children, you don’t think it’s important to teach children to read with understanding, think critically, collaborate, or solve problems.  You can’t have one without the other.

You may not like it, but you cannot ignore it.  Want to build your reform agenda around technology, structural changes, or accountability but take a hands-off approach to curriculum and content?  May I suggest a name for your group?  Try  ”Ed Reformers for Illiteracy.” 

Vander Ark is obviously a smart guy.  But his vision for education is all about delivery systems. Like many would-be reformers, he tacitly endorses a false and content-neutral, skills-driven notion that how children learn is more important than what they learn. 

“Rather than a common curriculum, learning platforms to come will support not just ‘multiple pathways’ but customized playlists.  Customized learning will be facilitated by comprehensive learning platforms surrounded by application and service ecosystems. Learning platforms will replace today’s learning management systems (LMS) that run flat and sequential courseware.  Like iPhone and Android, these platforms will unleash investment and innovation.

Dazzled yet?  Before you call your broker and load up on Apple and Cisco stock understand that if we don’t attend to what we put through these brave new pipelines, playlists and service ecosystems–or say it doesn’t matter–we will make no progress.  Zip.  Zilch. Nada. 

In a speech in Virginia last month, E.D. Hirsch, Jr. invoked Jefferson’s admonition that we “follow truth wherever it may lead.”  Where it leads — inevitably, incontrovertibly — is to understand that ”a coherent and cumulative early curriculum will raise in a systematic way the knowledge and the language of our students to a much higher level, and greatly narrow the unacceptable achievement gap between blacks and whites and between other demographic groups.”   If you want to raise a child’s general level of reading skill, you must raise his or her “domain specific” knowledge.  There is no way around it.  As Hirsch put it,

“The domain specificity of skill is one of the firmest, and educationally most important, findings in modern cognitive science.   It means that if you have learned a lot about chemistry, that won’t help your critical thinking skills in history.   Cognitive scientists have become quite skeptical of concepts like “critical thinking skills” as though they were a formal acquisition that can be applied to all subject matters.   Science has pulled the rug out from under the entire edifice of the anti-fact, how-to theory of education which has dominated in our schools for many decades, and was the chief cause of the verbal decline that appeared in the sixties that gravely weakened our nation.”

As Hirsch noted, you may not like where this leads, but you can’t pretend the facts aren’t there and the path isn’t clear.  “Consider then what the principle of domain specificity means for educational policy,” he said.  “It implies specific content in the curriculum, and a cumulative building up of the most enabling knowledge and language for all students.   The human capital of our people, the skills that our students will have will be dependent on the specific knowledge they have.  We cannot afford to leave the choice of specific topics and their cumulative sequence up to chance and whim.” [Italics mine.]

Hirsch was being polite.  I will be less so.  If you are opposed teaching a common body of shared knowledge to all children, you are opposed to teaching children to read.  You are in favor of illiteracy, either by choice or indifference.  You favor damaging our most vulnerable children by denying them the most critical thing: the functional knowledge they need to succeed. 

Deal with it, don’t ignore it.  Follow the facts where they lead.  Not just where you want them to take you.

A Curriculum Manifesto

by Robert Pondiscio
March 7th, 2011

A call for voluntary common curriculum has been issued today by a surprisingly diverse group of education, business and civic leaders.  The “Call for Common Content” issued by the Albert Shanker Institute, calls for a ”coherent, sequential set of guidelines in the core academic disciplines, specifying the content knowledge and skills that all students are expected to learn, over time, in a thoughtful progression across the grades.”

Among the dozens of signatories are Kati Haycock of the Education Trust, Linda Darling-Hammond, Tom Payzant, IBM Chairman Lou Gerstner, the Fordham Institute’s Checker Finn, and Harvard’s William Julius Wilson.  Core Knowledge founder E.D. Hirsch, Jr. and President Linda Bevilacqua also signed. 

The statement supports Common Core State Standards, but makes the point long argued on this blog that while such standards are praiseworthy, they are not a curriculum–and are unlikely to amount to much in the absence of a shared curriculum.  “To be clear, by ‘curriculum’ we mean a coherent, sequential set of guidelines in the core academic disciplines, specifying the content knowledge and skills that all students are expected to learn, over time, in a thoughtful progression across the grades,” reads the statement.  “We do not mean performance standards, textbook offerings, daily lesson plans, or rigid pedagogical prescriptions.”

The manifesto addresses head-on the fear of “centralization, institutional rigidity, and narrow-minded political orthodoxy” that typically strangles any discussion of a common curriculum in its crib.  “Common curriculum guidance does not represent a straitjacket or a narrowing of learning possibilities,” it reads.  The proposed curriculum guidelines would be “purely voluntary, comprising only about 50 to 60 percent of what is to be taught”—leaving room for state, regional, and local variations.

One of the most practical arguments for a common curriculum has long been the extraordinary rates of student mobility, especially among low-SES students.  And one of the most valuable contributions of the document is its contextualization of the role of poverty in student achievement—lifting the debate from narrow and needlessly polarized arguments about whether “demographics is destiny” or “teachers can overcome all obstacles”   Economically advantaged children come to school with a head start in knowledge and language acquisition.  “It is not poverty in itself, but poverty’s accompanying life conditions that help to explain performance gaps that begin at home and extend into secondary school and beyond,” the statement notes.

“Today, the information we need to minimize these performance gaps is in our hands, waiting to be used. Thanks to advances in cognitive science, we now understand that reading comprehension — so essential to almost all academic learning — depends in large part on knowledge. In experiments, when students who are “poor” readers are asked to read about a topic they know well (such as baseball), they do much better on comprehension measures than “good” readers who know less about the subject.

“The systematic effort to establish common, knowledge-building content must therefore begin as early as possible. The younger we start, the greater the hope that we can boost achievement across all schools and classrooms, but especially among our most disadvantaged students. Further, by articulating learning progressions linked to a grade-by-grade sequence for how learning should build over time, a defined curriculum will better enable each teacher to build on what students have already been taught. Students will also benefit, as they will be much less likely to find themselves either struggling to overcome gaps in their knowledge or bored by the repetition of what they have already learned.”

The manifesto also anticipates and addresses other knee-jerk objections that typically derail discussions of a common curriculum.  Critical thinking skills, highly prized as a goal of schooling, for example, “requires a curriculum that builds knowledge upon knowledge.”

“Finally, some may fear that common curriculum guidance will neglect important cultural referents or ignore the diversity of student experiences. However, as national curriculum standards in several high-performing nations illustrate, a modern conception of curriculum in a diverse nation is explicitly mindful of how to attend to cultural connections, and how to leave room for local adaptations and resources that enable students to connect to the curriculum from their different vantage points.”

The New York Times previewed today’s release of the statement, noting that previous calls for common academic standards, curricular materials and tests for use nationwide have been “beaten back” by those who favor local control of schools. “But last year’s successful standards-writing movement was a departure, leaving the outlook for this proposal uncertain,” writes the Times’ Sam Dillon.

I’m as sanguine about a common curriculum and convinced of the need as anyone.  Still, there will continue to be those who resist calls for common anything – standards or curriculum.  What’s encouraging about the statement and the Who’s Who of heavyweights who have lent their names to it, is its recognition that the preponderance of evidence is on the side of knowledge and language acquisition as the difference maker in raising achievement.  In that regard, it is an implicit challenge to would-be ed reformers to embrace not just structural change but instructional imperatives. 

I’ll resist the worn-out phrase “game changer.”  I’ll settle for “conversation changer.”

“And the Oscar for cynicism goes to….”

by Robert Pondiscio
March 2nd, 2011

“A public school chorus singing ‘Somewhere Over the Rainbow.’ I literally — if I wasn’t going to go out to some parties I would have slit ’em right then. It was the worst. I was looking for a knife to stick in my eyes, it was so terrible.”

Andy Cohen on MSNBC complaining about the performance of the 5th grade chorus from New York City’s P.S. 22. on Sunday’s Oscar telecast.  Cohen is  Bravo’s senior vice president of original programming and development, and the creator of the The Real Housewives of Orange County, New York City, Atlanta , New Jersey, and D.C.

Knowledge Compensates for Five Years of Reading Ability

by Robert Pondiscio
March 2nd, 2011

One of the principal arguments for a coherent, content-rich curriculum is that background knowledge — knowing something about the topic you’re reading about – compensates for weak reading ability.  But how great is the effect?  A tantalizing article by reading researcher Tom Sticht on EdNews.org suggests that in adult readers background knowledge can close a gap of five grade levels of general reading ability. 

If you want to test what someone knows about a subject, Sticht writes, “you might give them a simple multiple choice test in a written format, and then ask questions about the subject matter of interest.”  However this “confounds the assessment of the person’s knowledge about the subject with their ability to read.”  In other words, it’s difficult to tell whether a person taking a reading test doesn’t know the meaning of a word, for example, can’t decode a word, or lacks the background knowledge of the subject needed for adequate comprehension.   As Sticht puts it, we risk confusing ignorance with illiteracy.  “Generally there is no attempt to separately determine a student’s knowledge in the content area separately from the person’s ability to read in the content area in an unskilled or skilled manner,” Sticht observes.  

Sticht uncoupled reading ability from subject knowledge in work that he and colleagues performed for the U.S. Navy several years ago.  They developed “a 45 hour reading development program to help sailors improve their reading ability while increasing their knowledge needed for upward mobility in their career progression…In assessing learning outcomes in this course we considered both improvements in Navy career progression knowledge and increases in reading skill,” he writes.

They designed two separate assessments–one on the Navy-specific learning taught in the course; the other a more general reading assessment.   By comparing the two “we could determine separately the extent to which personnel had increased their Navy knowledge as well as their reading skill for incrementing their long term knowledge store using an external knowledge store,” Sticht says.

In additional work for the U.S. Navy we developed separate readability formula for determining how much general reading ability as measured by a standardized, normed reading test a person needed to be able to comprehend Navy material with 70 percent accuracy. We developed formulas for those with high and low prior knowledge about the Navy. We found that with low background Navy knowledge, a person needed a general reading ability of about the eleventh grade to comprehend with 70 percent accuracy. But highly knowledgeable personnel needed only a sixth grade level of general reading to comprehend Navy-related material with 70 percent accuracy. In this case, then, high levels of background knowledge substituted for some five grade levels of general reading ability.

The Armed Services, Sticht says ”have long understood the difference between general reading ability and specialized bodies of knowledge.”  When selecting people for service,” he points out, “lower general reading ability scores may be offset by higher scores in specialized bodies of knowledge.”  Confusing ignorance with illiteracy “contributes to a serious underestimation of the intellectual abilities of America’s children, youth, and adults,” he concludes.

Sticht’s eye-opening work validates a content-rich approach to K-12 education and once again demonstrates the intellectual bankruptcy of treating reading as an all-purpose, transferable skill, which it clearly is not.   Sticht’s point that we confuse ignorance with illiteracy has two big takeaways for educators:  the achievement gap is mostly a knowledge gap; if a “poor reader” and a “good reader” take a reading test where both share the same background knowledge, the perceived differences in their reading ability will likely narrow or disappear (this is also an argument for reading tests that are based on curriculum taught, not randomly chosen subjects). Secondly, if we know that knowledge correlates with general reading ability, then the best strategy for raising achievement across the board is with a rich and rigorous curriculum from the first days of school.  Teaching content IS teaching reading

“By not talking about the knowledge needed by the workforce people think they can teach any content and thereby improve the skill needed for workingi n a broad set of occupations,” Sticht wrote to me in an email. ”Failure to focus on knowledge leads to inefficient and too often ineffective career education or job training or retraining that many out of work people need,” he concludes.