New York was one of just two states to post statistically significant declines on the National Assessment of Educational Progress, a biennial assessment administered by the National Center for Education Statistics.

The state’s fourth-grade math scores fell for the second straight time, from a high of 243 points in 2007 to 238 this year. Scores on the eighth-grade math test and the reading tests showed no significant change.

Just 35 percent of fourth-graders in New York scored proficient or higher on the exam, considered the only reliable yardstick for measuring educational progress in a field of flawed state assessments. On the state’s own tests, whose scores dropped last year when state officials acknowledged that they had been inflated, more than 66 percent of fourth-graders were considered proficient in math.

It was the discrepancy between state test scores and NAEP results that triggered state officials to acknowledge that the state’s test scores were inflated in the first place.

State Education Commissioner John King called today’s results “disappointing and unacceptable.” In a statement, he said new state tests aligned with the Common Core curriculum standards, set to be given for the first time in three years, would improve New York students’ performance on the NAEP.

New York’s slip does not come as a total surprise. Math scores for fourth-graders fell in 2009, too — and that was when state test scores showed improvement. Since then, fourth-graders’ state math scores have stagnated, making any improvement or even stability on NAEP unlikely.

Nationally, students performed slightly better than ever before on the exam,. On average, score rose on fourth- and eighth-grade math tests and on the eighth-grade reading exam, while scores on the fourth-grade reading test stayed flat.

Despite this year’s stumble, over the last two decades, New York has seen a narrowing of the achievement gaps between white and black students and white and Hispanic students on the fourth-grade math exam, and also a narrowing of the gap between higher- and lower-income students in eighth-grade.

Today’s scores are for New York State as a whole; New York City results will come out later along with data from other urban school districts. Predicting the city’s NAEP picture isn’t straightforward: The city’s state test score gains have outpaced the state’s, but with so many of the state’s students enrolled in New York City schools, it would be hard for NAEP scores statewide to move one direction if the city’s scores moved the other way.

That funding, which the state legislature must approve, would support several policy changes. To catch cheating after it happens, the state will broaden erasure analysis to cover 10 percent of all elementary and middle school state tests. And as a preventive measure, teachers will be barred from grading their own students’ tests starting next year. The state is also requiring the city to boost on-the-ground monitoring of schools on testing days.

Deputy Commissioner Valerie Grey presented the new security measures to members of the Board of Regents during their monthly meeting today. The committee voted to approve the measures, and a final okay is expected when the full board convenes tomorrow.

The recommendations the Regents approved today were similar to those they first discussed last month, but there were two key changes. In the first, Grey said the state had abandoned a proposal to bar teachers from proctoring their own students’ exams after consultation with other states revealed that such a policy would be “highly unusual.” To compensate, Grey said, the state hopes to require districts to strengthen test-day monitoring. That proposal was not included in last month’s list, but was added after Regent Kathleen Cashin argued that a larger presence of test monitors was needed to prevent cheating.

Gov. Andrew Cuomo and state legislators must sign off on adding the funding to this year’s state education budget. But officials today appeared confident that the recommendations would go into effect. An open question is whether local districts will be able and willing to pay for additional test-day monitoring. In New York City, officials have said that school districts should not have to foot the bill for new security requirements the state sets.

Until now, the state has said it has had few systems in place to identify cheating on its annual tests. Instead, the state has relied heavily on local districts to handle their own test scanning and scoring — and their own investigations and punishments when cheating is suspected.

Since 2008, the state has conducted erasure analysis on a small fraction of high school Regents exams, but only as a limited pilot associated with a contract with a company that supplies tests. In today’s proposal, Grey asked for $1 million to fund an expansion of that program to cover 500,000 tests, or about 10 percent of the tests administered in grades 3-8. In addition, the department is asking for $700,000 to develop a system that would identify grading irregularities on free-response sections of tests.

Separate from the test security recommendations, the Regents also voted to ask for $200,000 to pilot computer-based tests that the state has said will be required statewide in the 2014-2015 school year in conjunction with the rollout of Common Core standards.

To supply these services, Grey said the state would pursue contractors with expertise in test security. She said costs would likely increase once the state hires an independent reviewer to look at the department’s process for handling cheating allegations and investigations, a move the Regents approved at their September meeting. Today, Grey said a selection process had already identified several candidates.

Many of the proposals are being pushed through quickly so that they can take effect for this year’s testing cycle. If passed, the erasure and reliability analysis would be used on the 2012 elementary and middle school math and English language arts exams, set to be administered in April.

For the 2012-2013 school year, the committee also approved a recommendation that would prohibit teachers from scoring their own students’ exams starting with the 2013 tests. In recommending a policy that would not take effect for another year and a half, Grey said the purpose was to send a message to districts.

“We think it’s important to say that teachers should not score their own exams,” Grey said.

The proposals came out of the test security task force that Grey has led since August, which State Education Commissioner John King formed amid high-profile news of cheating scandals in Atlanta, Washington, D.C., and Philadelphia. Those scandals have heightened debate about whether test scores can be considered reliable tools to make high-stakes decisions about teacher evaluations and school closures.

King delivered the message in a letter to district superintendents yesterday that suggested he had fielded worries about the cost and content of a new initiative to bolster test security.

First, King sought to allay concerns about the news, reported by the New York Times this weekend, that erasure analysis detected some test score improprieties.

“Obviously, this was not a large-scale, in-depth pilot of erasure analysis, nor did it rise to the level of a major finding or report,” he wrote, noting that the results of only eight exams were analyzed.

But he said indicated that erasure analysis could be on the horizon as the state continues to weigh initiatives to protect against cheating — and that it would already have been in place if the state had funded the Board of Regents’ recommendation for erasure analysis last year.

“Even this small body of evidence reinforces the larger message of the Department’s comprehensive test integrity review launched in August 2011 shortly after I became Commissioner: we need to take strong steps to ensure the integrity of New York State tests,” King wrote.

State Education Commissioner John King has convened a group to review “all aspects of the state’s testing system,” according to a statement from Jonathan Burman, a State Education Department spokesman. The group, which Deputy Commissioner Valerie Grey is leading, is planning to work quickly, Burman said: It was formed in mid-July and will announce a “series of measures” to ensure test integrity before the school year begins a month from now.

The announcement comes days before the state is set to release this year’s reading and math test scores and amid growing revelations about widespread cheating in Atlanta, Philadelphia, and New Jersey. It also follows mounting anxiety among state officials about whether schools’ performance had been inflated: Board of Regents Chancellor Merryl Tisch said in February that New York wished to avoid becoming “the Enron of test scores, the Enron of graduation rates.”

Schools Chancellor Dennis Walcott said he appreciated the state’s efforts but emphasized that New York City has for years “gone above and beyond” state requirements when it comes to ensuring test integrity.

“We welcome the state examining its standards, as it has always been its regulatory responsibility to ensure the reliability and security of state tests,” he said in a statement.

But the city is confident that its test results are sound, Walcott said, citing a 2009 audit by then-Comptroller Bill Thompson that he said confirmed the city’s test scores. In fact, while that audit found no new instances of cheating, it concluded that the city Department of Education had “engaged in sloppy and unprofessional practices that encourage cheating and data manipulation.”

Walcott also cautioned that any new measures to toughen test security could come with a high price tag that cash-strapped districts could have difficulty paying.

“I hope that as the state puts in place stricter security precautions, they don’t saddle districts with unfunded mandates,” he said.

One costly precaution that the state review group is likely to discuss is erasure analysis, a practice that detects suspicious patterns of changed answers. Georgia, Pennsylvania, and New Jersey all used erasure analysis to identifying instances of probable cheating. But New York State does not currently conduct erasure analysis on test results, Burman told me.

New York City stopped conducting systematic erasure analysis in 2001 and now employs the practice to corroborate or reject cheating allegations that are backed by other evidence, according to city officials.

The study surveyed adult New Yorkers on their knowledge of the basic structure of government. The authors Eric Lane, a professor at Hofstra University, and Meg Barnette of NYU’s Brennan Center for Justice, conclude that few New Yorkers know “even a little about the Constitution.”

They connect the poor showing to New York’s response to the federal No Child Left Behind law, arguing that the focus on math and English has hurt students’ Social Studies knowledge.

And they highlight a recent decision by the state Board of Regents to cancel an annual Social Studies test for fourth- and eighth-graders as the latest symptom of that disregard, which they dramatically term an “abandonment of history”:

For years New York required social studies assessment tests for its fourth and eighth grade students. The eighth grade assessment consisted mostly of history questions, while the fourth grade assessment tested skills such as graph reading. Overall, New Yorkers did not perform well on those tests, and New York City students performed horribly…As an explanation for this problematic showing, school officials said that they pay little attention to fourth and eighth grade social studies assessment tests “because they are not among the criteria used to determine if schools are performing adequately, either under state regulations or the federal No Child Left Behind law.”41 Proving that point, in the summer of 2010 the Board of Regents addressed the problem of low performance by ending the fourth and eighth grade social stud- ies assessment requirement, assuring, in the words of one education expert, the abandonment of history, and any hope for improvement in civic literacy at all.

The decision to cancel the test, made in June 2010, received little coverage at the time. In a memo explaining the move, the Board of Regents argued that the test was a casualty of budget deficits.

More than 60 percent of New York eighth graders scored below basic level on the 2009 NAEP science tests.

New York City fourth graders did about as poorly on a national science test in 2009 as those in other large American cities, but the city’s eighth graders lag behind their peers.

More than 60 percent of city eighth graders scored below basic on the National Assessment of Educational Progress science exams. Nationally, 38 percent of students scored below the basic level, and 56 percent of students in large city school districts did not meet that bar.

The city’s fourth graders fared better. Still, 44 percent scored below basic on the science tests. In other large cities, roughly the same percentage of students didn’t score above the “basic” bar.

The Department of Education’s Chief Academic Officer, Shael Polakow-Suransky, said that the city was focusing on introducing national “Common Core” standards into classrooms as a strategy to boost achievement in science. The standards include a focus on reading and writing non-fiction and technical texts in subjects like science.

“The gap between how our students perform in elementary and middle school is a real concern, which is why it’s crucial that we continue working toward the tougher Common Core Standards as quickly as possible,” Polakow-Suransky said. “That gap can be closed if we build literacy skills in science, as these new standards do, and prepare our kids for the critical thinking and problem solving they begin to face in middle school.”

A sampling of fourth, eighth and twelfth graders around the country take the NAEP exams every two years. NAEP scores are usually reported by state, but in 2002 several large cities including New York agreed to have their own figures reported separately.

Of the 17 city school districts whose results were reported today, New York City ranked seventh in fourth grade test results and eighth in eighth grade scores. Austin, Charlotte, Jefferson County, Ky., Miami-Dade, San Diego and Boston all bested New York City in both grade levels. On the eighth grade exams, Houston also performed better than New York.

Overall, New York City’s fourth-grade science scores were lower than the national average. But when the scores are broken down by ethnicity and poverty level, each of New York’s subgroups performed about the same on average as their peers nationally. (So for example, black fourth graders in New York City performed about the same as the national average for black students in that grade.)

The city’s eighth graders, by contrast, received lower scores than their peers nationally across all demographics except Asian students.

The results also indicate that fewer city students are doing well in science than in reading and math. More than 60 percent of the city’s eighth graders scored either basic, proficient or advanced in both reading and math in 2009.

The same was true across all of the large urban districts on average. U.S. Education Secretary Arne Duncan called on cities to improve their science education.

“The results released today show that students in our cities are further behind in science than in reading and mathematics,” Duncan said in a statement. “With 44 percent of fourth graders and 56 percent of eighth graders scoring below NAEP’s basic level, these results show that large city districts aren’t preparing enough students to succeed in the knowledge economy.”

In 2009, the science exams were overhauled, which means that the new results cannot be compared to previous years’ to track progress, the exam’s administrators said.

The new science exams cover three content areas: physical, life, and earth and space sciences. Unlike earlier NAEP exams, questions “crosscut” the subjects so that a question about one content area also relies on knowledge from one or both of the others.

The average city SAT score was five points higher on the reading portion of the test, four points higher on the math, and two points higher for writing. The gains are statistically significant, but not yet great enough to cancel out several years of loses. Today, the city’s average scores to roughly where they were two years ago.

City students’ average score was 439 out of 800 on the reading section, 462 on math, and 434 on writing.

The score increases are mainly due to improved results from Asian, white, and Hispanic students. Black students’ scores stagnated, except in the case of the writing SAT, where they fell by three points.

Another factor is the increasing number of students taking the SAT, but it’s not clear what effect this group of new test takers has had on scores.

As in years past, city officials noted that more minority students took the SATs this year and that many school districts and states have seen years of declining scores because of increased participation from lower performing students.

This year, about 4 percent more black students and 2 percent more Hispanic students sat for the college entrance exam. Fewer Asian students took the exam this year than last and white students’ participation rate held steady.

The city also released information today on how many students took and passed Advanced Placement exams, which test them on college level material they’ve studied in high school courses. More students are taking the AP exams — participation is up about 11 percent over last year, with most of the growth from Asian and Hispanic students — and more are passing them.

Many more students are taking the exams in part because the city has many more twelfth graders than it did eight years ago. With fewer students dropping out, the number of high school seniors has grown by 48 percent since 2002. In the same period, participation in the AP exams has grown by 60 percent.

The schools will test early versions of new third- through eleventh-grade exams that a consortium of 26 states — New York included — is creating. The same schools will get extra funding this year to pilot the new common core standards in their classrooms.

Because New York is a “governing state” in the consortium, its education officials have already agreed to begin using the new tests by the 2014 school year. It also means that New York officials, including city Deputy Chancellor Shael Polakow-Suransky, are helping design the new tests.

The PARCC group — Partnership for Assessment of Readiness for College and Careers — won a $170 million federal grant yesterday, which it will use to build the tests.

The new exams will complement the new national education standards that New York has also agreed to take on. They will also completely overhaul the form that state standardized exams take, and when they’re given, Suransky said today.

Right now, New York students sit for state standardized tests once a year, and the state reports results months later, over the summer. The tests consist mainly of multiple-choice questions, along with several free-response questions.

The new state test will be designed with four separate parts that students take over the course of the full school year, Suransky said. The first two parts, which students will take earlier in the year, will be shorter assignments that cover material the students should have learned up to that point. The third assignment will be longer and more complex. The fourth will be a comprehensive exam measuring a year’s worth of learning and will be given at the end of the school year.

And the consortium intends to dispense with much of the multiple-choice testing that students currently sit through, Suransky said. Instead, the assessments might take the form of a research paper or long-form math problems, for example.

“Those kinds of assignments are actually closer to the kinds of tasks that teachers are using in classrooms anyway,” Suransky said. “These will function as a way to test some of the new, higher-order skills that are in the common core standards.”

Suransky and other test designers are trying to meet a federal goal to create tests that better reflect student learning. ”By far the number one complaint I’ve heard from teachers, from parents, from students themselves is that state bubble tests pressure teachers to teach to a test that doesn’t really measure what matters,” U.S. Education Secretary Arne Duncan told reporters yesterday when he announced the grant funds.

At the end of the year, students’ four test scores will be combined into a single score. And teachers will also receive reports of their students’ performance on each of the individual sections weeks after they take them, so that they can use the results to adjust their teaching over the course of the year.

In that way, the new tests are designed to replace both the annual state tests and the diagnostic tests that many city schools already give students over the course of the year to track their progress, Suransky said. Suransky and federal officials said the new exams could lessen or roughly equal the amount of time students currently spend hunched over exams.

“I would argue we actually over-test now, in many places, in ways that aren’t helpful to the child and to the school and to the teacher,” Duncan said.

There’s also the possibility that the consortium’s tests for high school students will eventually replace the state’s current Regents exams. The state’s Board of Regents have not made a decision on the fate of the high school exams yet, though Suransky said he expects them to take up the question in the next few years.

The consortium is currently in the earliest stages of designing the new tests and will likely evolve over the next three years as designers build the new exams and test their validity.

Read the PARCC Consortium’s full grant application, which lays out its plans for building the new assessments in detail, here. Section A(3), which begins on page 43, gives a good description of what the new tests will look like.

New York State made that official last week when it finalized some cost-cutting changes to the state’s high school testing program but left most exams and test dates intact.

Back in March, state officials issued a dramatic proposal to gut the high school testing program. The state could save $13.7 million annually, they said, by eliminating exams in all subjects except math, reading, and science; ending January and August test dates used to help students graduate; and no longer translating test materials into foreign languages.

After the state budget provided for part of the Education Department’s funding request, officials ultimately decided to enact a scaled-down set of test changes. Students will no longer take a social studies exam in grades 5 and 8, and students who study German, Hebrew, and Latin won’t be able to take a state exam in those subjects.

But the vast majority of the Regents exams required for graduation will remain in place, at least for now. In a memo to district superintendents and school principals, Deputy Education Commission John King warns that unless the state finds more money to fund its testing program, the cuts might be back on the table as early as next year.

One big change that did go through was an end to the practice of re-administering just parts of tests to students who need only a small boost to pass, a practice called “component retesting.” The state has been spending $1.6 million a year giving component retests to just 9,000 students.

McGraw-Hill CEO Terry McGraw III, appearing on CNBC. The full interview can be seen here.

What goes on at McGraw-Hill, the mysterious Midtown company that makes New York’s state tests? One answer: The company is not-so-quietly producing a slew of ratings lambasted for being inflated, corrupt, and totally bankrupt.

I don’t mean more state test scores. I mean credit ratings churned out by Standard and Poor’s, the ratings agency that makes up nearly half of the company’s business, according to CNBC.

Yes, that’s the same ratings agency that has been criticized for inflating the value of companies from Enron to Bear Stearns.

One of the biggest criticisms of S&P and agencies like it is that their customers have an inherent interest in being rated highly.

As Thomas Friedman Paul Krugman wrote in a recent column, summing up the conventional wisdom:

It was a system that looked dignified and respectable on the surface. Yet it produced huge conflicts of interest. Issuers of debt — which increasingly meant Wall Street firms selling securities they created by slicing and dicing claims on things like subprime mortgages — could choose among several rating agencies. So they could direct their business to whichever agency was most likely to give a favorable verdict, and threaten to pull business from an agency that tried too hard to do its job.

I’m not saying that this is a perfect analogy for the testing industry, which also sells to a customer — state education officials — with a clear stake in looking good. In New York, at least, the customer right now is hell bent on looking bad as long as it means getting the tests more accurate.

But it’s a close enough situation structurally to make me laugh — and wonder if any McGraw-Hill executives are thinking about cross-company lessons.

That’s because in addition to having the state call fewer students proficient, both the city and state saw students’ average raw scores stagnate.

For years, state and city students average scores on the math and reading exams have risen. But from 2009 to 2010, the city students’ average reading exam scores held steady at 662. This trend continued on the math test, which also saw no significant increases or decreases in students’ average scale scores.

When the scores were separated out according to students’ ethnicities, they showed the same result: a flat line.

Speaking at Tweed Courthouse today, Mayor Bloomberg said the steady scores were a sign of progress. “The numbers that really matter are the actual scores,” he told reporters, adding that the state had made the tests more difficult this year.

This is true for the math exam. In an effort to cut down on the math tests’s now-predictable nature, state officials said they broadened the subjects it covered by 30 percent. Officials said they did not do the same for the reading exam because it was already a harder test than the math exam.

Schools Chancellor Joel Klein offered another possible explanation for the scores: the tests were given later in the year, allowing for more material to be covered. In past years, the math and ELA exams were given in January and March.

“There are so many different things that could be going on,” said Harvard University testing expert Daniel Koretz, who has been pushing the state to create harder tests. “The test was at a later day; it had broader content. It could be that the effects of coaching are petering out.”

Source: New York State Education Department

The day of reckoning has arrived.

After weeks of warning that adjusted standards would mean far fewer students passing state exams this year, state education officials released the exact numbers today.

Average raw scores on the state third through eighth grade math and reading exams remained flat. But because the state decided to raise the scores required for a student to be deemed proficient, the number of students passing fell sharply.

In New York City and other big cities, the number of students passing reading exams dropped by more than a quarter — from 68.8 percent of city students passing last year to 42.4 percent this year in reading, for example.

Just over 53 percent of third through eighth-grade students statewide passed the reading exam, compared to 77 percent last year. Around 61 percent of students passed their math exams, compared with more than 86 percent last year.

Pass rates of students learning English, students with disabilities, and poor students fell the farthest. The percentage of students learning English who passed the reading exam fell by more than half, from 36 percent to under 15 percent. Just 15 percent of students with disabilities passed the reading exam, compared to 39 percent last year.

Right now, Board of Regents Chancellor Merryl Tisch and Education Commissioner David Steiner are presenting the exam results in Albany. We’ll have more updates from the state and city presentations throughout the day.

But for now, the state’s slideshow presentation is below. An analysis of this year’s test results begins on slide 13.

Thousands of city students who passed their high school completion exams last year will receive a rude awakening once they get to college: They’ll have to retake high school math — if they get into college at all.

New analysis of students’ scores by Harvard testing expert Daniel Koretz shows that many students who passed these exams have essentially been lied to about their skill level. While a score of 65 on a Regents exam technically means the student is proficient, students actually need to score above a 75 or an 80 on the English and math tests in order to have a chance of getting into college and doing well once they’re in.

The percentage of New York City students who fell in this dangerous range of scoring between a 65 and a 75 or 80 was very high in 2009, when the most recent data was available. At that time, 51 percent of students scored in this range in algebra and 32 in English.

Though their scores should have meant they were competent in algebra and English, they actually mean these students are likely to score too low on their SATs to get into college. If they do get in, they face a high probability of being required to take remedial classes.

“The word ‘proficient’ should tell you something, and right now that is not the case on our state tests,” said David Steiner, the state education commissioner.

Students who score below an 80 on the math Regents have a 28 percent chance of scoring above 500 on the math SAT. If they get below a 75 on the English Regents, they stand a 19 percent chance of getting above a 500 on the reading SAT.

The City University of New York enrolls all incoming students who score a 75 or below on their Regents in remedial classes.

A slide in the state's presentation shows that eighth graders who score a level 3 on the math test have a low chance of getting a math Regents score that will lead to success in college.

In recent years, teachers, principals, parents, and much of the city’s press have met the annual unveiling of climbing test scores with increased skepticism, if not outright incredulity.

Today the State Education Department officially caught up to them and said yes, the results were too good to be true.

At a meeting of the Board of Regents this morning, Commissioner David Steiner presented (webcast) an analysis of state tests performed by Harvard University testing expert Daniel Koretz and New York University assistant professor Jennifer Jennings. The analysis shows that even though a greater percentage of students are passing the state’s exams than several years ago, many of these students are not prepared for high school or college.

Much of the criticism has focused on the state’s tests for elementary and middle school students and Steiner emphasized today that high school scores are exaggerated as well. Many students who pass the math Regents exam, even by a margin of 15 points, flounder in college, Steiner said.

Younger students who meet the state’s proficiency standards are moving onto high school with low chances of passing the Regents exams required for graduation. Those who do pass the Regents exams, even by a wide margin, are graduating without the skill level to get into college or, if they get in, to pass their courses.

Using an analogy (tests : educators :: thermometers : doctors), Steiner said New York’s thermometers aren’t very trustworthy.

According to the testing analysis, students who score below an 80 on their math Regents exams — well over the 65 required to pass — stand a good chance of being placed in remedial math courses in college. Many won’t even get to college, as the chance of them scoring at least a 500 on the math SAT is about 28 percent.

“I am saying that that figure of 80 is a fulcrum point,” Steiner said. “That below it you are just not a student who is going to score well on your SATs. And above it you are a student who will score well on your SATs.”

For the state’s English Regents exam, that fulcrum point is a 75.

Students in grades three through eight who sit through the state’s annual math and English exams have also been getting inflated results, Steiner said.

Eighth graders who score a level 3 — meaning proficient — on the math test have less than a one in three chance of getting an 80 on their math Regents. Those who get a level 3 on the ELA tests have a one in two chance of getting a 75 on the ELA Regents.

The Commissioner said he planned to improve the tests by making them more difficult. In the future, he said, they will cover more material and have higher proficiency bars. As soon as next year, they will also be significantly longer.

In New York City, a spokesman for the Department of Education said the analysis does not put the city’s rising test scores in doubt because city students’ scores are improving at a faster rate than students’ scores across the state.

State officials have refused to release Koretz’s full analysis.

As Education Commissioner David Steiner and Deputy Commissioner John King travel around New York explaining their plans to overhaul the state exams, they’ve largely met with support. In New York City, Schools Chancellor Joel Klein has called for tougher exams. But last week, Buffalo School Superintendent James Williams told The Buffalo News that he doubts Steiner and King’s approach will really improve the state’s schools.

“I think they’re two people who don’t know what they’re doing,” Williams said. “A more rigorous test is not going to improve student achievement. It’s not going to improve the graduation rate. I think it’s ridiculous.”

The state has already begun changing the content of the reading and math exams given annually to third through eighth graders and beginning this year, students will need higher scores on the tests to pass. Steiner and King, along with State Board of Regents Chancellor Merryl Tisch, are also calling for a new, more rigorous statewide curriculum. But Williams said that state officials are completing their work in the wrong order:

“There are three phases to improving education: One, you must have a curriculum. Next, you have the instruction. Then you do the assessment,” Williams said. “The State of New York seems to have it backwards. They’re talking about changing the assessment, but we don’t have a curriculum.”

The chart shows how neatly eighth graders’ scores on the tests predict their future academic success. The higher the score they get, the better their shot at graduating high school with a Regents diploma — the only kind that will count come 2014.

But the reader pointed out that the chart also includes a more frightening statistic: Students who score at a level considered proficient by every measure, a 3 out of possible 4, only have a 55% shot of getting a Regents diploma.

That 3 represents the low range of possible 3′s, which are handed out after the state converts raw results known as scale scores into a 1, 2, 3, or 4. A higher scale score equivalent to a 3.5 gives a student an 81% shot of getting a Regents diploma. Students who get a 4 out of 4, meanwhile, graduate 93% of the time..

Nevertheless, a 3 is considered enough to pass. And to move from one grade to the next, a student needs only a 2.

Concerns about the poor rigor of state tests began as cries in the dark by critics of the Bloomberg administration, elevated to the level of academic-but-still-not-official concern, and have now come close to gospel. New tests are being developed as part of the federal government’s push for “Common Core” standards, but they won’t be administered for several years.

UPDATE: The new people in charge of making state tests, State Education Commissioner David Steiner and his deputy John King, have taken the new gospel on the road. Steiner recently took the get-tough message against “grade inflation” to Rochester and Buffalo.

]]> http://gothamschools.org/2010/07/12/what-it-really-means-to-score-proficient-on-new-york-tests/feed/ 5 http://gothamschools.org/2010/06/15/on-first-day-of-regents-exams-test-jitters-spill-onto-twitter/ http://gothamschools.org/2010/06/15/on-first-day-of-regents-exams-test-jitters-spill-onto-twitter/#comments Tue, 15 Jun 2010 16:52:37 +0000 Maura Walz http://gothamschools.org/?p=40733 Pre-exam anxiety and post-exam elation and regret are in the air today, but those feelings are also streaming through Twitter.

By mid-morning today, the first day the city’s high schoolers are sitting for their Regents exams, thousands of tweets included the word “Regents.” A Twitter search paints a portrait of how students spend their time studying for and stressing out about their tests before they take them and how they celebrate after they finish. And it even includes a rare tweet from inside the exam hall.

“Good luck to everyone taking the Regents this week, including myself for my FINAL chance,” wrote one student. Jitters abound, though some students are entering the exams with confidence:

Some students warn that Twitter can abet cheaters, while others plan their cheating strategies:

And some students are just sick of the whole thing:

Here’s an assortment of students’ reactions to the exams:

Public school principals were told this morning how many of their students passed the state’s annual math and English exams and from what we’re hearing, the numbers aren’t pretty.

One principal wrote in to say that the percentage of his students who scored so low they didn’t meet promotion criteria has quadrupled since last year. On the English exam, his percentage of low-scoring students is more than ten times higher. Almost all of his special education students and most of his students who are recent immigrants didn’t pass the exams.

“It’s not like the kids have gotten dumber or the teachers worse, it’s just the tests are being looked at differently,” the principal said.

A Department of Education official confirmed that because the city and state set higher score cutoffs this year, fewer students will meet the standards for promotion to the next grade. As a result, the city expects that more students will be required to attend summer school  this year.

“We are committed to raising the bar for our students, so we’re using preliminary results on this year’s tests to set higher promotional cut scores than last year,” said DOE spokesman Matt Mittenthal. “We will guarantee a seat to every student who requires summer school.”

“We’re going to have a huge summer school program now,” the principal said. “No question about that.”

Schools haven’t received their students’ raw scores — they only know whether a student met the promotion criteria or didn’t. See below for the DOE’s cutoff scores.

Rather than rely on a group of teachers and state officials to examine tests for grade tampering, the city will begin scanning students’ multiple choice answer sheets next year. State officials said scanning tests will let them perform a high-tech cheating check called “erasure analysis.”

That means officials will be able to look for instances of teachers changing students’ answers by counting the number of times each student erased a wrong answer and bubbled in a correct one.

Next year, only six tests that students frequently take in order to get diplomas will be scanned, but in 2012 all Regents exams will be.

Board of Regents Chancellor Merryl Tisch said instituting scanning part of a larger plan to increase the public’s trust in the state’s results.

“The audits that have been done on the exams have led people to ask about the results in some instances and the scanning is the first step towards maintaining the integrity of the results,” Tisch said.

Scanning will also give state education officials more information about the tests than they’ve ever had.

Currently, the state depends on school districts to send in their students’ Regents exam results, but does not collect data about how different kinds of students perform on the exams or how they fair on certain types of questions.

“It’s not that we’re going to be grading the tests at all,” said SED spokesman Tom Dunn. “It’s that we’re going to know a lot more about how 220,000 kids who took the algebra test did on question four or how they did on questions that looked at equations.”

Dunn said it was still unclear how New York City high schools would get their exams to the one scanning center in the city, but that the state and city have more than a year to work this out.

The EMSC Committee approve the recommendation below:

The Department realizes that there will be costs associated with scanning Regents Answer sheets. To that end, we propose the following two step approach:

1. Starting with the June Regents Administration in the 2010-2011 school year, require that all Regents Exams in the following titles be scanned for Department analysis: Comprehensive English, Integrated Algebra, Global Studies and Geography, U.S. History and Government, Living Environment, and Earth Science. These exams are all used for the awarding of a Regents Diploma.
2. Commencing with the 2011-2012 school year (June Regents Administration), require that all remaining Regents Exams be scanned.

The National Assessment of Educational Progress, commonly known as NAEP, or the nation’s report card, is given every two years to students across the country. In New York City, 2,300 fourth and 2,100 eighth grade students took the NAEP reading exam last year.

While their peers in New York State have not seen real changes to their reading scores in over a decade, some New York City students saw gains. City fourth graders’ scores have increased an average of four points in the last two years and many more of them are meeting the standards that signal proficiency or basic understanding. In 2002, 48 percent of fourth graders scored basic or above on the exam and in 2009 62 percent were in the range.

The score increases have put New York City on the list of large cities where fourth grade students’ scores rose in the last two years. It’s a short list: Boston, Washington, D.C., and Houston are the other three.

The city’s eighth grade students have not made the same leaps. In 2003, the first year reading data for these students was separated from state-wide data, their average score was 252. Six years later, it’s exactly the same. Atlanta and Los Angeles were the only cities to see eighth graders’ scores rise in the last two years.

Eighth graders’ results on New York State’s own annual reading tests don’t match up with their NAEP scores. According to the state’s tests, city eighth graders’ scores rose 10 points between 2007 and 2009. Fourth graders’ state scores were more in line with their NAEP results, as the state showed them making a nine point score jump and the national exam put the increase at four points.

The difference between the state’s results and NAEP scores is a constant that has gotten more attention in recent years and has forced state officials to call for more difficult tests.

In a PowerPoint Department of Education officials sent to reporters, the city emphasized that its students’ scores have risen while scores across New York State have not, signaling that the wide gap between the city and state is narrowing.