From a New York Times article about a school visited by NYC Schools Chancellor Joel Klein.

Mr. Klein...emphasized in his remarks that the school did more than the basics, and had robust arts, music and computer programs. In one fourth-grade class, he encouraged students to tell him what they liked about the school. And he seemed surprised when Andrew Xu, the first boy to raise his hand, replied, “They help us get ready for the state ELA test.” Andrew, 9, was referring to the test for English Language Arts to be taken next week by all students in grades three through eight. “They teach us what methods to use and how to write.”

There's one 9-year-old who's absorbed the lesson.

Posted by John at 11:50 AM | Permalink | Comments (0)

Yesterday's PISA results showed 15-year-old students in the U.S. lagging behind the top industrialized countries on PISA, a test that emphasizes the practical application* of science knowledge. 

The two extremes in responding are to say (1) the results prove there's a "crisis" in science education or public education or (2) the negative findings don't matter at all. 

I resist the crisis talk because it's really the same crisis identified 24 years ago by A Nation at Risk.  When do the effects of this ongoing education crisis hit?  The motivation for the crisis talk is understandable -- Gov. Romer used the c-word at yesterday's event -- but even today's supposedly miseducated kids can understand the story of the Boy Who Cried Wolf.  Let's figure out a way to respond to these results and improve science education without playing the crisis card.

I also want to push back just as hard against the notion that PISA results should be dismissed.  Some have argued that the tests are unfair to students in the U.S.  Reasons cited include our students' lack of motivation to excel on PISA, the nature of the test questions themselves (different from the way science is taught here), and our nation's diversity.  All these may have some validity, but dismissing the results seems disingenuous at best.

Here's a possible response* that avoids Scylla and Charybdis:

Learn from our own successes.  Identify the areas where U.S. students do well and where we do poorly.  One area where we do well is Earth Science and Space.  Figure out why we are relatively successful in this area and extend some of the same practices to areas where we do poorly.

Learn from others' successes.  Examine and learn from some of the practices of the most successful countries.  Here are two takeways from yesterday's briefing:

Successful countries set a high bar for entry into the teaching profession, and

They do an excellent job of collecting and reporting data on student achievement (though they don't necessarily attach consequences or interventions or sanctions to the data).

Use PISA results as one reality check for measuring the quality of science education in the U.S. However, don't stake everything on PISA.  Consider TIMSS, NAEP and state tests.  Course-taking is important.  And don't forget students' grades in science courses (remember, those subjective grades are often viewed as a better predictor of success in college than are SAT scores).  We should also consider the number of scientists, engineers and mathematicians graduating from our colleges and other appropriate measures.

Don't overlook out-of-school factors. Even the loudest voices for school reform acknowledge that out-of-school factors affect student achievement.  It would be ridiculous to believe otherwise.  The U.S. ranks near the bottom on international rankings of how we treat our children. What do high-achieving countries do outside the classroom that we could emulate?

Invest in science labs and improve hands-on science instruction so students can learn to apply their science knowledge, especially in urban schools. By the way, for those who pooh-pooh class size reduction in the upper grades, think for a minute about about whether class size matters for science labs in middle school and high school. 

Recognize that nation-to-nation comparisons are imperfect.  This goes well beyond the PISA's idiosyncrasies and the role of out-of-school factors or example, PISA seems to a lot of stake in comparing expenditures by translating everything into dollars.  I've spoken with an IMF economist who argues that a better method for some international comparisons is to express them as a percentage of GDP. 

I haven't run these ideas by the smart people at the AFT, so I'm sure I've missed some key ideas, misstated some things, and left holes in my reasoning you could drive a truck through.  But soon the PISA results will have faded from memory and all the public will be left with is a vague impression that students in the U.S. are somehow lagging behind those in other countries.  We could run around like our hair's on fire bemoaning the awful state of our schools, hide our heads in the sand, or use the results to help create rational plan for improving education.  I'm just making my case for option 3.

*Sample PISA questions can be found here.

Posted by John at 11:29 AM | Permalink | Comments (0)

Here's a post in anticipation of tomorrow's release of PISA science results, which, no doubt, will show America's 15-year-olds far from the top.  Some organizations (perhaps even the AFT in a statement responding to the results) will point out that students' science achievement today is critical to the nation's economic success in the future.

That may well be true, but let's put things in perspective by considering whether American adults know anything about science.  In a recent (and well-written) report titled, "Chance Favors the Prepared Mind," Gary Phillips of AIR noted a few problems with adult Americans' scientific knowledge. 

According to the National Science Foundation, the average U.S. citizen understands very little science. For example:

• Two-thirds do not understand DNA, “margin of error,” the scientific process, and do not believe in evolution.

• Half do not know how long it takes the earth to go around the sun, and a quarter does not even know that the earth goes around the sun.

• Half think humans coexisted with dinosaurs and believe antibiotics kill viruses.

On the other hand, according to the NSF, the general public believes in a lot of pseudoscience.

• Eighty-eight percent believe in alternative medicine.

• Half believe in extrasensory perception and faith healing.

• Forty percent believe in haunted houses and demonic possession.

• A third believes in lucky numbers, ghosts, telepathy, clairvoyance, astrology, and that UFOs are aliens from space.

• A quarter believes in witches and that we can communicate with the dead.

The average citizen is also not very literate in mathematics. According to the National Center for Education Statistics:

• Seventy-eight percent cannot explain how to compute the interest paid on a loan.

• Seventy-one percent cannot calculate miles per gallon on a trip.

• Fifty-eight percent cannot calculate a 10% tip for a lunch bill.

Phillips isn't making excuses for today's students' poor performance on international tests, and neither am I.  But if you think mediocre results like those that will be formally announced tomorrow are a recent phenomenon, I've got some lucky lottery numbers to give you for the next PowerBall.  No, seriously, I got them from an alien who swears he heard it from a witch.  And if you win $100 million, my commission will be 10%, which is $40 million.

(Image from Perfesser via Flickr.)

Posted by John at 05:05 PM | Permalink | Comments (0)

While there is increasing evidence that NCLB has led to more "teaching to the test" and a narrowing of the curriculum, the law's supporters remain dubious.  To document what is happening in the classroom, the Texas Federation of Teachers (TFT) is encouraging teachers to keep track of the amount of testing throughout the school year, using this handy calendar.  It's a smart campaign that engages members and will provide strong evidence to policymakers about the dangers of overemphasizing standardized testing. To learn more about the TFT's "Reclaim Your Classroom" campaign, click here.

Posted by Michele at 10:06 AM | Permalink | Comments (0)

From yesterday's Rose Garden remarks:

"Last month, we learned that 4th graders earned the highest math and reading scores in the history of our Nation's Report Card -- and that's good news. I'm able to report that because we actually measure now in the schools."

Two points:  First, scores are increasing at a slower rate since NCLB passed.  Second, uh, NAEP, or "our Nation's Report Card," has been "measuring...in the schools" since the late 1960s.

Posted by John at 04:48 PM | Permalink | Comments (1)

I am late out of the box on this one but, tardy or not, it's worth checking out.  Jackie Bennett at Edwize points out the inconsistencies in this recent editorial from her hometown paper about NCLB and whether it promotes science instruction.

Posted by Michele at 06:03 PM | Permalink | Comments (0)

I’ve been informed by Edwize that I linked to an article that didn’t tell the whole story of English language learners and test scores in NYC.  Another Daily News article informs us that test scores were purposely exempted during the mayoral race of 2005 and the school system looked very good as a result.  Now that the scores have to be included, the mayor and chancellor are blaming the federal rules for the drop in scores. 

This gaming of the system is exactly the kind of thing that that NCLB was trying to avoid.  Alas, implementation and human nature mucked everything up and, as I said in the original post, we still can’t tell if NCLB is working. But, if this allegation is true, shame on the mayor and chancellor for using children to win political races.

Posted by Beth at 08:30 AM | Permalink | Comments (1)

Catching up on last week's news...the U.S. Department of Education says the rise in NAEP history and civics* scores "proves NCLB is working."  Scores are rising, so it makes sense to say something's working.  But is it NCLB?  If NCLB were working, then scores would go up faster after its passage, right?  Let's check the stats from NCES.

• Fourth-grade history rose 3 points between 1994 and 2001, and 2 points between 2001 and 2006.
• Eighth-grade history rose 3 points between 1994 and 2001, and 3 points between 2001 and 2006.
• Twelfth-grade history rose 1 point between 1994 and 2001, and 3 points between 2001 and 2006.

Using the same logic and looking all the way back to 1994, President Clinton's Improving America's Schools Act was working, too -- working better than NCLB for fourth-graders, not as well for 12th-graders, but working. 

I'm not trying to argue for President Clinton's law, just trying to point out the folly of judging everything that happens in education in terms of NCLB.  Other factors affect students and their families: poverty, health care, teacher labor market forces, immigration patterns, tax cuts for the rich that cripple state budgets, etc.  It's ludicrous to single out one of the factors without some kind of statistical analysis.  Ludicrous -- but that's what Sec. Spellings did to conclude that "NCLB is working."

*The civics scores go back only to 1998, so there's no way to tell if NCLB "is working" better than what was happening before or, for that matter, how much of the rise occurred before NCLB was signed in the pre-NCLB era of 1998-2002 and how much occurred in the NCLB-era of 2002-2006.

Posted by John at 03:18 PM | Permalink | Comments (0)

Over the weekend, NPR looked at a school in Oprah's hometown of Baltimore, the city's Northwestern High School, which is focusing on the so-called bubble kid.  Those are the students just short of passing the state's high school assessment, or HSA. 

NPR's Larry Abramson reports, "Two to three times a week, certain students are pulled out of their regular classes and put into these special test preparation sessions.

The principal, Tajah Gross, says the students are chosen because they're believed to be just a hair's breadth from passing the HSA.

A state department of education official objects to the school's test preparation regimen, saying, "We would prefer that school systems consider the entire school year as preparation for the test."

Professor Daniel Koretz of Harvard says the school's strategy may distort the test: "If it's done intensively, it inflates them, produces gains in test scores that are much larger than the real increases in student learning."

Education Trust's Amy Wilkins says, "What people are, sort of, doing is rather than saying we're going to have teachers that can teach all kids, they're pulling the classes apart now and saying, you get this, you get that. I'm not sure that's really the way to do this.

Text here, audio here.

Posted by John at 11:02 AM | Permalink | Comments (0)

I expect to see more articles like this one as science assessments come online as required by NCLB in 2007-08.

The narrowing of the curriculum because of NCLB’s focus on math and language arts is not news, but the fact that it is done at the expenses of science is shockingly shortsighted for a couple of reasons:

• It is not like we didn’t know about the 2007-08 deadline for science assessments back in 2001 when the law was passed.  Any decision to cut science instruction in say, 4th grade was obviously going to have implications for science assessment results in middle and high school.
• There is a big push now for high schools to focus on 21st century skills/ international competitiveness/college readiness.  Though science is one of those subjects that requires less sequential building upon content (you can learn about the solar system and the structure of plants without knowing anything about volcanoes, for example), you can’t just start high school biology without an understanding of the basic structure of plants or what a cell is. And schools can’t just ignore high school science if they expect kids to go to college or to compete globally. I would not have been eligible for the state university I attended without three years of high school science.

But what to make of the fact that NAEP science scores rose at exactly the same time that schools were supposedly decreasing time allotted for science instruction?  Maybe the increased focus on language arts has increased students’ comprehension so much that they’re doing better on the comprehension-dependent science questions.  Or maybe we can conclude that we don’t have to add science to AYP in order to see scores rise. Or maybe two years of test data is not enough from which to draw major conclusions?

Posted by Beth at 03:19 PM | Permalink | Comments (0)

A new report confirms what we already know: NAEP’s definition of proficiency is too high, as shown by the fact that only a handful of Asian countries could do better than the U.S.  On the other hand, here is general agreement that many states are setting their own proficiency bars too low.  The extreme example is Mississippi, where 71 percent fewer are “proficient” on NAEP than on the state assessment. And Mississippi has plenty of company---every state has a gap between the percentage of students deemed proficient on NAEP and the percentage on the state assessment.

So, Goldilocks, what’s a just right standard of proficiency?  And what to do about the current inconsistency? The "sunshine and shame" tactic of comparing NAEP results to state standards is getting policymakers' attention now but will get tiresome after years of little or no change. Allowing states to do what they want, including lowering proficiency standards, is equally problematic for a Congress and administration touting an education law that is supposedly built upon high standards and accountability.

I am not sure what "just right" is, but the AFT is recommending what could be a compromise until everyone figures it out: grants to consortia of states to develop common standards, assessments and curricula as a step toward consistency and higher quality.

Posted by Beth at 03:12 PM | Permalink | Comments (1)

That seems to be the reasoning of Fairfax County, Va., and other Virginia school systems that announced yesterday that they would comply with the provisions of NCLB and administer the standard English language arts assessment to their English language learners.

After a months-long standoff, we get this summary from a local superintendent: "It does not make educational sense or academic sense, but the U.S. Department of Education does not make sense either."

(And yes, Kevin, sometimes schools make educational decisions based on funding needs.)

Posted by Beth at 02:14 PM | Permalink | Comments (0)

Seems like the folks in Houston don’t like the state standardized test and want to move to end-of-course assessments. Makes sense to me, though I would think that this would only apply to high school courses, rather than elementary or even middle school. A potential downside, depending on your view, is that this may lead to some folks asking for more NCLB-related testing in high school. Now, the state standardized test probably covers a few grade levels worth of subjects like biology, chemistry and earth science, while the end-of-course test would only assess, well, content for that course.  Seems like you’d want to make sure that students mastered the range of high school content, not just what students happen to be taking in, for example, 10^th grade.  The upshot is that it would mean that the focus would be on course content rather than test preparation. 

In other test related news, a bunch of states have figured out a way to give a common Algebra II test.  That nicely aligns with the AFT’s NCLB recommendation that the feds should provide grants to voluntary consortia of states to develop common academic standards, assessments and curricula.

Posted by Beth at 08:54 AM | Permalink | Comments (0)

Virginia is scheduled to administer its Standards of Learning test in about three weeks.  Fairfax County and others have said that they won’t give the standard test to their English language learners.  They claim they’ve got a test that is fairer.  The U.S. Department of Education says fair is following the rules that have been on the books since 2002.

Last Week, Republican and Democratic members of congress introduced a bill asking for “common sense flexibility” in dealing with this issue.  Sound familiar? That’s because Secretary Spellings used it back in 2005 when she talked about her approach to NCLB. 

Did we mention that Secretary Spellings is a resident of Fairfax County and sent her daughter to the public schools there, and that Fairfax could lose $17 million in NCLB funds if it does not follow the rules?

So, who is going to blink first? I’d guess Spellings.

Who is uncharacteristically silent on this issue?  Eduwonk.  He’s on the VA Board of Ed, so he must have an opinion.  As a blogger, he’s in the business of sharing his opinions all of the time.  C’mon, Eduwonk, tell us what you think.

Posted by Beth at 11:39 AM | Permalink | Comments (1)

What follows is a comment from a parent of a special education student on a recent post.  I've heard people argue that the term "high-stakes testing" is all about the consequences for adults rather than children.  It's true that failing an NCLB-mandated tests doesn't necessarily mean a student will be held back or given an F.  But this impassioned, informed, unsolicited comment should put to rest the idea that NCLB's tests aren't sometimes "high-stakes tests" for children. 

I would love to have a face to face or I would settle with a over the phone talk with the President, and anyone else that really wants to know how the No Child Left Behind works.  Coming from a parent of Special Education children I hate it.  When it comes to testing they have to be tested on there grade level not the level of learning they may actually be on.  What a load of bull!!  Do you understand how that makes a child feel when they can't complete a test because they don't even know what they are doing?  It's very degrading.  Now be that parent when that child comes home upset because they didn't do as well on the test as the other kids do.  Then to hold that school responsible when that child grade level didn't do well because you tested a child on something they have no clue about is also wrong.  How do you truly see how a school is doing?  Anyone really wanting to sit down and talk about this, feel free to email me.

Posted by John at 10:26 AM | Permalink | Comments (4)

This weekend brought another call for states to create tests as "rigorous" as NAEP, in the form of a New York Times editorial titled "Real Tests for Real Children." Despite the headline, the editorialists seem to be out of touch with the reality of NAEP tests and cut scores.

They write:

Instead of moving toward the educational excellence that the country needs to compete in the global economy, many states opted for dumbed-down tests and deliberate sleight of hand to create the fraudulent appearance of progress.  As a result, states that perform well with their own watered-down exams do shockingly poorly when their students take the far more rigorous federal test known as the National Assessment of Educational Progress.

The first sentence expresses dismay that states are gaming the system to deal with AYP, though the editorialists ignore AYP's inability to measure progress accurately. The second sentence links national standards, national tests, and federal funding to NAEP proficiency levels and NCLB's 100% proficiency requirement.  Combining NAEP proficiency and NCLB's 100% target is a recipe for disaster.  (We've written about this before, but it's one of those bad, Hydra-like ideas that keeps coming back.)  

The AFT's recommendations for reauthorizing NCLB include grants for voluntary consortia of states to develop common academic standards, curriculum and assessments to provide more consistency in the definition of proficiency and growth across participating states.  We continue to support high standards and rigorous tests. 

But here's why relying on NAEP proficiency levels is over the top.  In 1995 the National Education Goals Panel published a study suggesting that it is unrealistic to expect 100% of students to reach NAEP's "proficient" level.  Researchers converted scores for the top-performing countries on an international exam (called the IAEP) to NAEP-equivalent scores and found that even the top-performing countries fell far short of 50% proficiency.  Taiwan, for example, finished first in the world in 8th grade math and science, but just 41% of Taiwan's students scored at the "proficient" or "advanced" levels on this test.

Another NCES study, linking state NAEP results with TIMSS scores, showed that some states competed very well with the top countries. Only Singapore out-performed Wisconsin, for example, but in the 19 times that Wisconsin's students have taken NAEP exams, the state has never broken the 50% percent NAEP proficiency threshold.

It seems that no country or state has ever had 50% of students performing at the level of NAEP proficiency. Yet the Times seems to be recommending that we shoot for NAEP proficiency for 100% of students.  The Times' editorialists should understand that they are asking America's students to fly twice as high as any other group of students, anywhere else in the world, has ever flown.  That isn't a noble call for rigororous tests, high standards and accountability.  It's an awful, unrealistic recommendation that would put a federal "failure" stamp on large numbers of teachers and schools -- even if America's students outperform those in every other nation.

Posted by John at 12:02 PM | Permalink | Comments (0)

Posted by Beth 

Regarding the latest NAEP data release, I’ll start by saying that it is tragic that the racially-identifiable achievement gap is not closing.  All involved in education should redouble efforts toward this goal.

I’ll also speculate that maybe the teaching to the test and the repetition of  rote skills to students in the poorest schools—who happen to be predominantly Black and Latino—don’t lead to the higher order thinking skills needed to achieve proficiency or better on NAEP.

This is of course speculation, but the achievement gap is closing in 4^th grade, but not budging in high school.  I’m hypothesizing that some of the “drill and kill” test prep and lower order thinking skills forced onto the poorest schools may lead to pretty good scores in elementary school, but leave students ill-equipped to fully comprehend high school level texts.  Maybe what we need is a rich, content-laden curriculum for all students, as outlined in this American Educator article, “Building Knowledge.” Here’s an excerpt from author E.D. Hirsch:

I believe inadequate attention to building students’ knowledge is the main reason why the reading scores of 13- and 17-year-olds on the National Assessment of Educational Progress have not budged in years. I believe this neglect of knowledge is a major source of inequity, at the heart of the achievement gap between America’s poor and non-poor. I also believe that if this idea about what is limiting students’ comprehension isn’t understood and aggressively addressed, reading scores won’t move up, no matter how hard teachers try. And the public debate will wrongly continue to pillory teachers and public schools for stagnant achievement scores.

Posted by Guest at 10:30 AM | Permalink | Comments (3)

When international test results put American students in the middle of the pack on international tests, John Stossel was up in arms, laying the blame on teacher unions and the public school "monopoly."

But don't expect to hear anything from Stossel about Child Poverty in Perspective: An Overview of Child Well-Being in Rich Countries, recently released by the U.N.  It shows that the U.S. did indeed rank in the middle of the pack (12th among 21 countries) for children's Educational Well-Being.  But it also finds that the U.S. is 17th in Material Well-Being, 21st in Health and Safety, 20th in Family and Peer Relationships, and 20th in Behaviors and Risks. Stossel was outraged when American students were in the middle of the pack, but he'll be silent about the U.S. being at the bottom of the list in our treatment of children.

So why did Stossel attack earlier and why won't he attack now?  Because the "blame the schools" approach fits his preconceptions.  Unions are bad.  Public services are less efficient than private enterprises.  Therefore, if students aren't doing as well as they should be, we need to get rid of unions, break up the public school monopoly, and insert private-sector values into our schools.

But you won't hear from Stossel about this study.  With the U.S. ranking 18th in children's Material Well-Being, he won't call for a better social safety net or a higher minimum wage. With the U.S. ranked 21st in Health and Safety, he won't call for universal health care. If solving a problem might require more government involvement, Stossel and his ilk simply pretend the problem doesn't exist.

All this is not to say that the performance of American students on these tests shouldn't be better, or that we don't need changes in our schools to improve teaching and learning.  But it is folly to think scores will soar if we do not address out-of-school factors that affect students' lives.

Posted by John at 12:03 PM | Permalink | Comments (0)

Posted by Beth 

A timely news bit from the Baltimore Sun informs us that snow days lower test scores. Seriously, researchers have looked at the data and found that when there are a number of snow days in January and February, and the exams are given in March, students perform poorly in comparison to years with no snow days.

Their solution? Wait for global warming to produce smarter kids? No. Live in a place where it doesn’t snow? Nope. In southern California in the 1980s, we had the occasional afternoon off because the smog levels were so high. Really.

The researchers suggest waiting until the end of the year to give the assessments. Makes sense to me, especially if the tests are assessing fourth grade knowledge and skills, for example. You’d think you’d want to give the kids a full years’ worth of fourth grade instruction. Problem is, the test developers say that they can’t turn the results around in time to plan for the following year if test administration dates are moved past early spring. I’m not really buying it. Plus, c’mon, everyone deserves a snow day now and then.

Posted by Guest at 12:14 PM | Permalink | Comments (0)

We've had a series of long comments on NAEP assessments and how they're used.  The gist of the comments--I think I have this right--is that scoring, accommodations and characteristics of NAEP change, so it is not a reliable device for making year-to-year comparisons. 

The question of NAEP's reliability becomes more important with the announcement of proposals to raise the stakes associated with NAEP--linking it to national standards or publishing comparisons of NAEP proficiency levels and states' proficiency levels.

Here are a few quotes to give you a flavor:

Comment:  The NAEP trends report emphasizes that the results are comparable over time, but there are indeed several critical changes with the 2004 analyses. For example, for the first time with NAEP trends, the data were calculated using nonpoststratified sample weights (p. 107)....[W]ith the NAEP trends test in 2004, they actually changed the primary sample unit to stay within state physical boundaries (p. 97) and for the first time used ONLY age-eligible students (p. 93).

Comment:  Tests should be updated to reflect scientific advances in measurement, but it is difficult to know if a child has grown an inch when you keep changing the yardstick.

Comment:  Comparisons of 2005 data (and even 2002 or 2003) to previous years are simply not valid.  Consider three major changes since 1998, two since NCLB was passed.... 

Comment:  Can we afford to place the fate of our educational system on tests that, with only supposedly minor changes in procedure, can produce such different results?

And this commenter purports to have some information on the thinking of Rep. George Miller (D.-Calif.), chair of the House Education and the Workforce Committee. 

Comment:  Rep. Miller believes that there may have been some hanky-panky with the public-private or ESL data.  Similar statistical adjustments to NAEP main and trend assessments have obfuscated the effects of NCLB in reading and math.

Posted by John at 11:30 AM | Permalink | Comments (0)

Posted by Dan at AFT. 

(Michele and I have invited Dan to do a series of posts stripping away the rhetoric and looking at some hard data on student achievement and other measures of student progress.  Overall, we hope to show that while there is plenty of room for improvement in our schools, there is also quite a bit to celebrate -- John)

Kids today.  They can’t read, write or think for themselves.  If only we could return to those halcyon days when all kids knew the three Rs.  If only public schools weren’t such dismal failures.

We hear this refrain so often that it has become part of our public discourse, reflexively accepted as fact.  For example, a recent Atlantic Monthly article echoes the conventional wisdom: "Despite serious efforts to improve the performance of U.S. students, it has been flat since the 1970s."

Where do these perceptions come from?  Certainly, the right-wing, pro-voucher, anti-public education attack machine has a lot to do with it. For example, in his widely criticized but heavily viewed "Stupid in America" segment, John Stossel features a typical quote from education researcher Jay Greene:  "We’ve doubled per pupil spending,* adjusting for inflation, over the last 30 years, and yet schools aren’t better."

Stossel then adds:  “He’s absolutely right.  National…achievement scores are flat....”

Stossel, Greene, and the rest of the pro-privatization crowd love making these statements, but are they true? 

For now, I’ll leave you with a couple of statistics from the National Assessment of Educational Progress (NAEP) long-term trend test, widely considered the “gold standard” for measuring achievement over time:

•    Between 1971 and 2004, African-American nine-year-olds increased their reading scores from 170 to 200.  A rule of thumb is that ten points on NAEP roughly equals one grade level, so that’s an increase of about three grade levels.

•    Between 1973 and 2004, African-American nine-year-olds increased their math scores from 190 to 224—an increase of almost three and a half grade levels.

Is that progress?  Absolutely.  Can we do even better?  Absolutely, we must. But I sure prefer today's dystopia to the "good ol' days" of years past.

*Greene’s school spending claims are also misleading, but we’ll get to those later.

Posted by Guest at 01:32 PM | Permalink | Comments (1)

As we've recently pointed out, there's a demonstrably false statement about NCLB floating around and certain politicians seem to be providing the hot air to keep it afloat. 

The false claim is that National Assessment of Educational Progress (NAEP) results rose more in the last five years (or since 2001) than in the previous 28 years combined.  A sharp rise did occur, but it was between 1999 and 2004, and that's an important distinction because NCLB wasn't enacted until 2002.  The legitimate post-NCLB NAEP results for 4th-graders show little movement in reading scores.

Secretary Spellings and President Bush have spread the falsehood.  Apparently, it was also sent out through ED's e-propaganda campaign.  And the lie or error has taken on a life of its own.  It's been:

• included in Wikipedia's "No Child Left Behind" entry [UPDATE: Thanks to Sherman Dorn for modifying the Wiki entry to reflect objections to ED's use of NAEP data.]
• unquestioned by the Pittsburgh Post-Gazette, the Greensboro News-Record, Bloomberg, CNN, the Tavis Smiley show, and the Greater Danbury (CT?) News-Times
  
• bragged about in a Honolulu Advertiser op-ed written an ED official,
• bragged about by another ED official in a letter to the editor of the Columbus Dispatch
  
• posted on the Web site of the American Embassy in Italy,
• used to beat up on Senator Ted Kennedy, and
  
• regurgitated by a Republican member of Congress.

Despite the conventional wisdom among some political hacks, repeating a lie does not make it true.  There has been no dramatic increase in NAEP reading results since 2001 or since NCLB was enacted.

Posted by John at 03:04 PM | Permalink | Comments (2)

[Thanks to Dan at AFT for help with the NAEP results described below.]

The U.S. Department of Education (ED) is no longer employing the rhetorical device known as a "half-truth" to puff up No Child Left Behind (NCLB) and the President Bush's sinking popularity.  Apparently, at least on ED's Web site, the half-truth is being replaced by the untruth (in bold below).

"Today, with these principles and the hard work of America’s dedicated teachers and school administrators, we can see that the law is working. Significant progress has been achieved over the last five years:

• The achievement gap is narrowing—scores are at all-time highs for Hispanic and African American students, who are catching up to their white peers;

• 9-year-olds have made more reading progress since 2001 than in the last 28 years combined; and

• In math, 9-year-olds and 13-year-olds earned the highest scores in the history of the Nation’s Report Card."

The statement above simply is not true.  Even after you put all the spin from the Bush administration's PR machine on it, it's still not true.

This misuse of test scores to justify NCLB goes back to the 2004 campaign.  And just last week, we noted that President Bush and Education Secretary Margaret Spellings had used imprecise language to create a politically convenient confusion between two five-year periods.  The first is February 1999 to February 2004, a period that includes about three years before the enactment of No Child Left Behind and a dramatic rise in certain test scores.  The second period is January 2002, when the law was signed, to January 2007, when the Bush administration is celebrating the law's fifth anniversary.  During this period, dramatic test score increases are harder to find. But blend the two five-year periods together, throw a few exclamation points into Secretary Spellings' speeches, and -- voila -- No Child Left Behind is "working!"   

But you can't say that NCLB is working (or, to be fair, that it's causing harm) if you look only at NAEP's legitimate post-NCLB testing periods for 3rd (oops) 4th grade reading.  Scores went down a point between 2002 and 2003, up a point between 2003 and 2005.

The lie error occurs in an announcement of a News You Can Use piece that will be aired on PBS.  Yes, it's news you can use but only for political purposes, kind of like those fake TV news stories ED created under Secretary Rod Paige.

The political appointees at ED did get one fact right: This month really does mark the 5th anniversary of NCLB.  But there isn't much cause for celebration unless you're one of the political appointees.  We hear they're going to be partying like it's 1999.  Or 2001.  Or whatever.

Posted by John at 07:55 AM | Permalink | Comments (0)

Through chutzpah, ignorance or cynicism, President Bush and Education Secretary Margaret Spellings (who should know better) recently pointed to education successes that occurred before NCLB -- and tried to attribute the success to themselves and NCLB. 

President Bush's Saturday radio address included this nugget:

"Since No Child Left Behind was passed, we have seen major improvements in student achievement all across America.  In reading, nine-year-olds have made larger gains in the last five years of the test than in the previous 28 years."

And Education Secretary Spellings echoed the claim in her speech marking the fifth anniversary of NCLB, also linking the test score rise to the law:

"With No Child Left Behind, we set the goal to have every student reading and doing math on grade level by 2014. And it's working!  The Nation's Report Card showed our younger students made more reading progress in 5 years than in the previous 28 combined.

Let's skip over the exclamation points and political rhetoric.  Here are the facts, courtesy of the NAEP long-term trend results.

There's a big rise in 9-year-olds' scores.  But when did it occur? Between 1999 and 2004, with the testing conducted between January and March.  Most of the time lapse between the two tests cited by Pres. Bush and Sec. Spellings, occurred BEFORE January 2002 when the law was enacted.  And even more time passed between the signing of the law and when it began to affect classrooms. 

Many indicators suggest student achievement is on the rise, and that's good.  But the trend predates NCLB.  And Pres. Bush and Sec. Spellings can continue to make the preposterous claim that NCLB is 99.9% pure or needs only a few tweaks.  But a good chunk of those "large gains in reading" likely came when candidate Bush was saying that federal education policy reflected "the soft bigotry of low expectations" and "No Child Left Behind" was just a doodle on Karl Rove's notepad.

Posted by John at 05:30 PM | Permalink | Comments (1)

Joseph Berger of the New York Times writes about the inclusion of English language learners (ELLs) in state assessments by profiling the schools in Port Chester, NY.  Berger makes an attempt to discuss both sides of the issue, but he comes off as skeptical of the New York state's policy of excluding student from academic assessments until they have been in the public schools for three consecutive years.  (I mean, the title of the article is, "Immigrant Children Shielded from State Assessments--But for Whose Protection?")

Research shows that it takes ELLs five to seven years to fully acquire the English skills needed to handle academic assessments in English, so it seems that New York state was acting properly in this regard.  And, missing from the New York Times artilce is the fact that the U.S. Department of Education (ED) permitted New York to use an English proficiency assessment--instead of an academic assessment--for these students and only recently reneged on its deal with the state. It's not as simple as making the state look like the bad guys here--ED changed its policy rather abruptly.

To get more of the back story, read the testimony of Maria Neira, New York State United Teachers Vice President and AFT Vice President, to the state's Assembly Standing Committee on Education.

Posted by Michele at 12:30 PM | Permalink | Comments (0)

NCLBlog welcomes Matt at AFT to our humble blog. 

"Accurate, but misleading." Those were the words used by the National School Boards Association’s Patte Barth to describe how the Washington Post chose to characterize--rather, mischaracterize--the performance of U.S. students on an international math assessment.

In a letter to the editor in Saturday’s Post, Barth writes:

The table that compared the United States with other countries in performance in fourth-grade math gave the false impression that the United States ranked 12th out of 13 countries, and it suggested that the only country the United States outperformed was Cyprus.

Not shown were 12 countries, including Italy, Australia and Norway, that were outperformed by the United States, leaving the impression that U.S. fourth-graders were next to last in the world. In reality, the United States is among neither the highest-performing nor the lowest-performing countries in fourth-grade math.

A fuller picture would also reveal that U.S. performance varies by subject and grade level. For example, U.S. fourth-graders performed well in reading; they were outperformed by their peers in only three of the 34 countries that took that test, and the Americans were above the international average.

In contrast, U.S. high school students were outperformed by their counterparts in 20 of 30 countries in math, a performance that was below the international average.

Posted by Guest at 10:09 AM | Permalink | Comments (0)

Posted by Beth 

One thing that is missing from most of the reports about the latest urban NAEP scores is this bit, tucked into the end of the EdWeek story:

In the NAEP urban-science study, scores may have been influenced by districts’ policies toward excluding students with disabilities and English-language learners from taking the tests and allowing test-takers special accommodations during the exams. The number of excluded students, and those allowed accommodations, varied greatly from district to district, the NAEP report reveals. The Austin district, for instance, which had the highest 4th grade scores, excluded 9 percent of all the students tested at that grade level either because they were in special education or English-language learners. By comparison, the second-highest-scoring district at the 4th grade level, Charlotte, excluded only 3 percent of those students. The nation-wide average exclusion rate was 3 percent.

Seems to me like this should be a big part of the story. If District A is supposed to be compared with District B, and District A excludes three times the percentage of students as District B, the results aren’t completely comparable. I don’t know what the correct exclusion allowance should be--I’d probably err on the side of generous--but a standard should be set for all participating districts.

Posted by Guest at 03:17 PM | Permalink | Comments (0)

After reading this morning's WaPo article about students who receive good grades, but then score poorly on state assessments, I was reminded of this story, told by former Texas Federation of Teachers President John Cole in the American Educator:

Let’s start with my initial years as a teacher in Corpus Christi. Schools in Corpus Christi weren’t desegregated until 1976. So when I started teaching in the late 1960s, we had three sets of schools—one for whites, one for blacks, and one for Hispanics. I taught in the Hispanic junior high school. We were blessed, I suppose, in that we got the textbooks right after the white junior high school was finished with them—when we finished with them, we sent them over to the black junior high school.

The school had a lot of dedicated teachers, but as an institution, the public school system didn’t really care too much about what went on in the school that I taught in—or in the other schools that Hispanic and African-American kids attended. For example, there were no standards for coursework. We had valedictorians from some of these schools who couldn’t get into college because they hadn’t taken the right courses. There were many places where kids took the same remedial math course four years in a row under a different name. They never got to algebra, never got beyond arithmetic. When I started teaching, I was told that four percent of the kids at my junior high went on to graduate from the Hispanic high school in Corpus Christi; the other 96 percent dropped out along the way. As far as I could tell, not one person cared if I ever taught a lick.

Cole's story is a sobering reminder of why it's important to "keep score" in education through a standards and accountability system.  He also explains why, when proposed, the new Texas accountability system had the support of the union:

We kept pumping in new money so that between 1984 and 2000, state aid per pupil increased by 24 percent (in constant dollars). And we didn’t just add new money—we redirected the state’s resources so that low-wealth school districts and school districts with high concentrations of disadvantaged children received the bulk of it. It was a revolution. We also created a minimum standard for teachers; it was suddenly much harder for administrators to hire unqualified people and call them teachers. Starting in 1986, all teachers had to take a basic reading and writing test; if they couldn’t pass it, they lost their teaching certificate. But we also increased salaries, spectacularly so in the poorest districts, so that when new teachers were hired, we were able to attract teachers who met the higher standard.

The Texas accountability system is hardly perfect, but it shows what can be accomplished when the resources are available--something to keep in mind as we approach reauthorization of NCLB.  As Senator Kennedy recently remarked in a recent floor speech on Senate education priorities:

This year, we’ll revisit the reforms contained in the No Child Left Behind Act . . . The law charted a sound course for American education four years ago, but it’s time for us to reshape our commitment and provide better solutions for schools to respond to the challenges identified by the law. These reforms are right and we’re ready to work with President Bush, as we did five years ago. But given the many failures of implementation by his Administration and the meager commitments to education reform in his budgets, the President has a high hurdle to cross to demonstrate that he is seriously committed to these reforms.

Update: Kevin Carey at The Quick and the Ed on the WaPo article here, Eduwonk here.

Posted by Michele at 01:45 PM | Permalink | Comments (3)

Update: Education Week($) is reporting that at least two more states are caving into ED’s requirements for testing English language learners. But, just to make things fun, while ED is saying that states can’t use portfolios or use a single test to assess language proficiency and reading NOW, they’re also saying that they are looking into the possibility of allowing them to be used LATER.- Beth

Posted by Beth 

That is the often-repeated (and often effective) threat from the U.S. Department of Education (ED).  In a recent case, it was a threat directed toward New York state. For years, the state had been giving an English language proficiency test--instead of an English language arts content test--to English language learners (ELLs) who have been in U.S. schools for less than three years. Recently, ED said that New York was out of compliance with NCLB, and that all but first-year arrivals had to take both assessments. The state caved and agreed to comply.

Read testimony from Maria Neira, AFT Vice President, on New York State United Teachers’ take on why the decision is unfair and not educationally sound.

Posted by Guest at 02:45 PM | Permalink | Comments (0)

Update: Sherman Dorn adds to the debate, pointing out that many parents think having schools do test prep for the SAT and ACT is just dandy.

Of course I can't resist getting in on the "teaching to the test" blog debate (see here, here, here and here).  I think Matthew Yglesias' comments on Craig Jerald's recent post are dead-on:

He [Jerald] brings good evidence to bear that real teaching is a more effective way of improving test scores than is simple test prep. That, however, isn't evidence that schools are not, in fact, doing what The Wire portrays them as doing.

And, in fairness, Jerald concedes that, "my evidence doesn't prove drill-and-kill teaching to the test isn't happening in lots of places."  It is, in fact, happening in too many places, driving at least one teacher out of the profession. 

I was "off-duty" last weekend, attending a bachelorette party in Baltimore with some friends (most of whom are teachers), and they were lamenting the overemphasis on testing in their schools.  Right or wrong, they attribute this development to No Child Left Behind and, since I work with the AFT's legislative department, they want me to fix it!  Here's hoping.

On a happier testing note, the Harvard Education Letter($) has an interesting article about formative testing and its value to teachers and schools.

Posted by Michele at 07:45 AM | Permalink | Comments (0)

So, if most state assessments are "crappy," as Sara Mead argues at The Quick and the Ed, why is it again that they should be used to determine whether a school is making adequate yearly progress under NCLB?  And, if the tests don't accurately measure student learning, why would be want to use them to judge teacher performance?

Posted by Michele at 04:04 PM | Permalink | Comments (0)

Over at The Life that Chose Me, Dick has been blogging extensively about how his high school is handling administration of the Georgia Alternate Assessment (GAA), designed for students with severe and profound disabilities.  It's not a pretty picture.  His post on the staff training for the GAA stands out:

The anxiety level amongst participants had to be noticibly higher, as evidenced by the frustration of the presenters.  And this is justifiably so, because of the gap between performance of some students and the grade.  For instance, the presenter wanted to use 3rd grade examples because that is what the knobs at the state department used.  The difference between a student functioning at an 11 month old and an 8 year-old is very significant.  However it is nothing compared to the gap that exists between a student functioning at an 11 month-old level and a 17 year-old.   This is the distance many of us high school teachers have to jump.

While I understand the importance of having students with disabilities participate in state assessments, after reading Dick's posts, I have trouble understanding what their scores would really show about their abilities.

Posted by Michele at 08:45 AM | Permalink | Comments (4)

Check out this summary of the latest research by David Card and Jesse Rothstein on the effects of neighborhood and school segregation on student achievement.  Using SAT scores, the authors find that "neighborhood composition matters more than school compositon."  One possible explanation for the lack of school effects is the prevalence of within-school segregation.

Posted by Michele at 04:30 PM | Permalink | Comments (0)

Posted by Beth

Recently, the New York State Education Department decided to add a "mixed or other races" category to its current list of racial and ethnic categories for reporting,* and the U.S. Department of Education has proposed regulations requiring states to include a multiracial category.

I’m of two minds about this. On one hand, kids are increasingly not fitting into one neat category. They (and their parents, teachers and schools) should not be forced to choose one box over another, and increasingly, one box is inaccurate.

On the other hand, this is a civil rights issue. If there weren’t achievement gaps between the various racial and ethnic groups to expose, there would be no need to disaggregate the data along these lines. The federal government has taken on the collection of racial and ethnic data to protect civil rights in employment, housing, education, etc., and the fact that there are still big gaps between the groups in each of these areas means that their job is not done. Adding a mutiracial group might mask some gaps.

Obviously, ED’s proposal has implications for NCLB AYP, but in this decision, the stakes are higher.

*Contrary to popular belief, NCLB does not require that states disaggregate student achievement data by the categories of Black, White, Hispanic, Asian and Native American, but by the "major racial and ethnic subgroups" as decided by individual states. Some states have a multiracial category; most don’t.

Posted by Guest at 12:08 PM | Permalink | Comments (1)

Posted by Beth 

There too much testing in Texas and, rightly so, folks are upset.  The Texas Federation of Teachers surveyed its members and found a couple of things that might resonate outside of the state:

• Most members report that over 50% of class time is devoted to preparation for the TAKS Texas' testing program.
• Over 85% of members report that test preparation and drill and kill activities result in a significant loss of instructional time.
• More than 90% report that quality of education in areas not covered by TAKS has bee adversely affected by the emphasis on TAKS scores.

Despite this dissatisfaction with the current testing system, Texas educators don’t want to totally get rid of standards and accountability.  More than half of teachers say that they would fix the TAKS testing program, but that there is a need for statewide standards and testing.

In reaction to the testing survey, the Texas Federation of Teachers developed a seven point plan for testing reform.  Their ideas all make sense, and could certainly be applied to other states.

Posted by Michele at 12:16 PM | Permalink | Comments (0)

First grade teacher The Rain, who writes at I Thought a Think, posts about how he uses formative assessments to place his students in reading groups.  He also writes:

There's a myth out there that teachers are anti-test. It's not true. All teachers test and assess; the difference is in what we do with the information.

Go over and visit!  And maybe you can figure out why he includes a graphic for Tom Wolfe's The Electric Kool-Aid Acid Test.  (I tried to read that book a few years back, and let's just say it has not withstood the test of time.)

Posted by Michele at 08:30 AM | Permalink | Comments (0)

What, pray tell, is the NCEXTEND2?  Why it's North Carolina's test for the two percent of students with disabilities allowed to take a modified assessment under NCLB and still have their scores count for AYP purposes. Now say that five times fast. Picture the poor teacher having to explain that concept and say "NCEXTEND2" multiple times in a parent teacher conference. Whew!

All kidding aside, North Carolina should be commended for being one of three states--Louisiana and Kanasas are the other two--with the fortitude to develop modified assessments before federal regulations regarding their specifications are finalized.  (Yes, the final IDEA regs did come out in August, but Ed Week reports that we should expect additional regs on the two percent in the new year.)

Ed Week found that developing such assessments is turning out to be a difficult task for states, in part because it is hard to identify exactly which students should be allowed to take them.  I imagine developing test items that are the same as the regular assessment, but cover less breadth and depth must also be a challenge.  Never could quite get my mind around that one.

But that name-Holy Acronym Batman!  Maybe that's what you get when you allow the psychometricians to name the test.

Posted by Michele at 01:25 PM | Permalink | Comments (0)

Posted by Beth at AFT 

Who knows why it took ED two years and three months from the time they released a draft regulation to the time the regulation was finalized. Some (not the AFT) might say that’s enough time for an English language learner (ELL) to learn English.

Most states have been operating as though the draft regulation was law, so it is a good thing that the final regulation has so few changes. The three main provisions stand:

ELLs can be exempt during their first year in U.S. schools from taking and English language arts assessment (but not an English language proficiency assessment).

During their first year in the U.S., ELL students’ scores can be exempt from counting toward AYP.

Students who exit the ELL subgroup because they’ve attained English language proficiency can have their scores count in the ELL subgroup for AYP calculations for up to two year.

These changes align with Secretary Spellings "common sense" approach to NCLB, though the AFT has many other ideas for policies to improve the education of ELLs.

Posted by Guest at 10:24 AM | Permalink | Comments (0)