Archives for category: Testing

After I posted an article about Denver yesterday, in which former board member Jeannie Kaplan asserted that corporate reform had failed in Denver, I received a note from Mike Petrilli of the Thomas B. Fordham Institute. I responded and copied Kaplan, since she wrote the post. I think it is an interesting discussion. Please read it from the bottom.

From Mike Petrilli at 9:26 pm:

If it’s everything in the chain below, yes.

On Fri, Oct 9, 2015 at 9:18 PM, Diane Ravitch wrote:

Mike, would you be okay with my posting your exchange with Jeannie Kaplan?

Diane Ravitch

On Oct 9, 2015, at 3:30 PM, Jeannie Kaplan wrote:

Sorry, Mike, but I totally disagree with you. When you make reducing the gap a main tenet of measuring success and this gap increases for ten years, you are failing. That’s like changing the rules in the middle of the game. But I guess it is all about winning, just not for the students.

And I would love to hear what you see as success in Denver. We have more segregated schools than any time since before bussing, we have the highest teacher turnover of any district in the metropolitan area, we have the most bloated administration in recent memory, etc., etc. The bottom line for me is telling the truth, and the truth is Denver is our kids are not being educated. That’s what matters. Taking tests isn’t educating. Reducing curricula isn’t educating. Pretending this is success isn’t telling the truth. My greatest fear when I was first elected in 2005 was Denver would be left as a district for those who have no other choice. I believe that is what we could easily be seeing in the not too distant future, especially when you have a board and administration that absolutely refuses to listen to anyone but themselves. And, frankly, like you, insist this is success.

Show me the (money) progress! And don’t cite 1+% gains. For all the hoopla that doesn’t cut it. The first Denver plan (2006) had very high standards and goals, none of which were ever reached. No accountability. I get asked repeatedly, where is the accountability and why does Tom Boasberg still have his job? Because when you buy the board and when you have no press, and when you have organizations like yours pretending this is working, you can say whatever you want and no one challenges. Reformers here have been very clever – a form of mayoral or superintendent control of the board without asking the voters to approve it. We are outgunned financially. Your side has poured 300,000 dollars per race into getting the six board members elected. I will expect nothing less this time.

And meanwhile, as your side hides the truth or spins it, our children and communities and Dps employees suffer. But you are certainly winning the pr battle!

Jeannie

On Oct 9, 2015, at 3:14 PM, Michael Petrilli wrote:

Hi Jeannie. What I’m saying is that we might find some common ground here. Looking literally at gaps turns everything into a zero sum game. Certainly we don’t want to root for white or Asian or middle class kids to do worse.

The right goal is to see progress across the board. Which is what Denver is demonstrating. Not that it’s perfect, but its progress is real, and promising.

Mike

******

On Fri, Oct 9, 2015 at 3:05 PM, Jeannie Kaplan wrote:

We seem to be crossing emails. Answering this I would say when reformers claim to be all about data, and when they point to reducing the gap as a mark of success, and when the Denver public schools has seen an increase in the gap in all three academic subjects based on free/reduced lunch as well as ethnic categories – white v. Black and Hispanic – for ten years it just might be time to try something different.

And if we were to stop using the gap as an indicator what might you suggest to replace it?

Jeannie

From Mike Petrilli at 3:12 pm

Yes, and then stagnation in the late 80s and most of the 90s, and then progress again with test based accountability. See: http://www.nber.org/papers/w15531

On Fri, Oct 9, 2015 at 2:57 PM, Diane Ravitch wrote:

No, read Paul Barton on the black-white gap.

Biggest narrowing was late 1970s-early 80s. Smaller classes; early childhood; economic opportunities for African American families; desegregation.

Diane Ravitch

On Oct 9, 2015, at 2:47 PM, Michael Petrilli wrote:

Well, if that’s the case, you should stop using gaps to hit reformers over the head.

However, we did see major gap closing from the late 1990s into the 2000s, probably because of test-based accountability. Though that has now plateaued.

Mike

On Fri, Oct 9, 2015 at 2:39 PM, Diane Ravitch wrote:

We know the reasons for the gaps

The reformers have cynically politicized the issue by claiming that they can close it. No they (you) can’t and you won’t, [My addition here: not with test-based accountability. Tests are a measure, not a cure.]

Diane Ravitch

On Oct 9, 2015, at 2:28 PM, Michael Petrilli wrote:

Hi Diane. Just FYI:

1. Denver is one of the few urban districts in the country with about 30% white non low-income kids (and some very high end kids) so kids are starting out far apart.

2. Both low-income and non-low income kids have grown. Low income kids growing at 1-2% per year while non low-income kids growing at 1.5-2.5% per year so the gap grows

But do we want to start rooting for affluent kids to do worse? As you’ve said, “closing the achievement gap” is good rhetoric but we don’t want to take it too literally. The goal is for everyone to learn more.

Mike

Valerie Strauss has a fascinating column about executive pay at some of our major education “nonprofits.” It is hard to see the difference between nonprofits and for-profits when you look at executive compensation. I hope she next takes a look at the compensation of charter chain executives.

According to the latest publicly available 990 tax forms filed to the IRS by the three organizations, which operate under 501(c)3 tax exempt status because of their declared educational missions:

Kurt Landgraf, now the former president and chief executive officer of the Educational Testing Service, earned for the 2013 fiscal year ending Dec. 31, 2013: $1,307,314 in reportable compensation and $42,210 in estimated other compensation from the organization and related organizations. [See the ETS 990 here.]

Jon Whitmore, the chief executive officer of ACT, earned for the 2013 fiscal year ending Aug. 31, 2014: $672,853 in salary, plus a bonus of $150,000, and other reportable compensation of $12,949, plus retirement contributions of $57,152, plus other nontaxable benefits of $18,109. That’s a total of $911,073. [See the ACT 990 here.]

David Coleman, the president and chief executive officer of the College Board, as well as a trustee, earned for the 2013 fiscal year ending June 30, 2014: $690,854 in reportable compensation plus $43,338 in other compensation from the organization and related organizations. Total: $734,192. (Coleman, a co-author of the Common Core State Standards in English Language Arts, joined the College Board in 2012, and was new in his job). [See the College Board 990 here.]

Go to her column to see the links to the tax forms.

ETS paid some members of its Board of Directors at a rate up to $103,000 a year for what is reported as approximately two hours of work a week, about $1,000 an hour. And more than three dozen top executives received more than $300,000 in total annual compensation; seven of those topped half a million dollars. For example, Philip Tabbiner, senior vice president for business innovation and growth, earned $655,055 in reportable compensation. Randy Bennett, Frederiksen Chair for assessment innovation, earned $316,450 in reportable compensation and a total of $489,758. Donado Yvette, vice president and treasurer, earned $422,793 in reportable compensation.

At the College Board, senior vice president Peter Negroni earned a total of $811,873 — the majority part of a severance package. That total was more than what was listed for Coleman, the College Board president.

The College Board itself claimed total assets that topped $1 billion, and its “assessment” programs — mostly the SAT and PSAT — took in $333 million but spent $289 million, for a net of $44 million. ACT’s total assets were $530,638,419 for fiscal 2013.

The College Board spent $1,768,295 on lobbying Congress and other public officials, the form says. ACT’s 990 form reported $674,485 in lobbying expenses, and the ETS, $40,851.

Fairtest, which criticizes the misuse of standardized tests, reported total revenue of $145,332.

Average teacher pay: about $50,000.

Who adds more value to society?

Jerusha Connor, a professor of education at Villanova University, was shocked to see what happened to her daughter on her first day of kindergarten: Most of the few hours of school were spent on assessment by five different teachers.

She writes:

For anyone who doubts that education in the U.S. has become overrun by testing, consider this. My daughter’s first day of kindergarten — her very first introduction to elementary school — consisted almost entirely of assessment. She was due at school at 9:30, and I picked her up at 11:45. In between, she was assessed by five different teachers, each a stranger, asking her to perform some task such as cutting, coloring in the lines, reciting her address and phone number, identifying letters and their sounds, and counting. She then had to wait two days, while all the other incoming kindergartners were assessed, to learn of her teacher and begin the school year in earnest.

From an educator’s point of view, this approach makes good sense. Determine what it is that kids know. Then use that baseline knowledge to assemble a class.

But this was an intimidating initiation from a child’s perspective. Usually an outgoing and independent girl, my daughter was clingy and nervous on her first day of kindergarten. When I asked how she was feeling as we approached the front door of the building, she said she did not want to go to school. She did not have any friends yet. She did not know her way around the building. She worried that there would be too many people. What if her teachers were mean? What if kids made fun of her when they heard her name? What if she had to use the restroom? She was a bundle of nerves. I’m sure this testing scenario did little to quell her concerns. I have no doubt that however she was assessed, she did not perform from a place of confidence or comfort. Even under less trying circumstances, such one-shot assessments are of questionable validity.

Indeed, by the time I picked her up, she had not relaxed at all. She did not want to talk about what she had done in school, but she did say that she did not want to go back. She did not know the teachers’ names. She did not make any friends. Later that afternoon, as she played with her animals in her room, I overheard her drilling them on their numbers and letters.

She and her husband were saddened by their daughter’s experience.

My husband and I will do our best to help her unlearn what she learned about school on her first day: that it is a place where you are judged for what you know — not how eager you are to learn; that performance matters more than understanding or inquiry; that schoolwork is hard and uninteresting. We will work with her teacher (whomever he or she is) to ensure that the strengths she brings to kindergarten — curiosity, compassion and creativity — are recognized and nurtured. We will encourage her love of learning and her self-confidence; I just wish we did not have to work against the school system in doing so.

Our educational system’s drive to assess, to label and sort kids, to make decisions on the basis of data of dubious quality has gone too far, and it is time for a course correction. We must remember that “data” are social constructions, shaped by the circumstances under which they are obtained. And just as these circumstances affect the nature of the information we collect, they have bearing on other things that matter, such as a child’s first impressions of school. I submit that these impressions matter more than any purported snapshot of a child’s abilities.

The reformers’ obsession with testing is harmful to children.

Michael Grunwald, who usually writes about politics, not education, has posted a mostly admiring profile of Arne Duncan.

Bottom line: He really cares!

What he leaves out: Arne’s persistent support for privatization of public education.

He does touch on the opposition to high-stakes testing, but skirts the hot-button issue of teacher evaluation by test scores.

But he does feel bad about the instability that has occurred in Chicago since he left.

And Undersecretary of Education Ted Mitchell cries when he talks about Arne. Michael Grunwald probably doesn’t know that Ted gave up a $750,000 a year job as CEO of NewSchools Venture Fund, the leading privatization organization in education, to join Arne.

As Grunwald says, you can’t understand Arne if you don’t understand basketball. I know a lot about education, but I don’t understand basketball. So that’s my problem.

And

Maine was part of the Smarter Balanced Assessment Consortium. Students took the online tests. Complaints from students, parents, and teachers were so loud that the legislature dropped out of SBAC.

Now the the results are in, and no one knows what to make of them since they are not comparable with previous tests.

Many students opted out. The state average of opt outs averaged more than 10%, but in some schools more than 50% did not take the tests.

Laura H. Chapman, a frequent contributor to the blog, who has been a teacher, arts educator, curriculum designer, now retired, writes the following provocative contribution:

There is a well funded marketing campaign to sustain the Common Core and the associated tests.
One facet of the current campaign is designed to lower public expectations about the success of students on the SBAC and PARCC tests and to say, in effect, that cut scores on these tests will to set to approximate the operational definition of proficiency and the pass on NAEP tests, Only students who score at nearly the highest level on NAEP tests are dubbed proficient. 
There is also a bit of distraction going on, because SBAC has already announced cut scores based on its field trials in 21 states. There is another little known fact: When PARCC and SBAC applied for federal funding, they promised to make their scores comparable.
PARCC will “coordinate with the Smarter Balanced Assessment Consortium on… artificial intelligence scoring, setting achievement levels, and anchoring high school assessments in the knowledge and skills students need to be prepared for postsecondary education and careers.” Partnership for Assessment of Readiness for College and Careers. (2010, December 23). Proposal for supplemental race to the top assessment award. Retrieved from http://www.edweek.org/media/parccsupplementalproposal12-23achievefinal.pdf p. 3
Similarly, the SMARTER Balanced Assessment Consortium (SBAC) asserts: “SBAC and PARCC are strongly committed to ensuring comparability between their assessments…[including] collaborative standard setting that will facilitate valid comparisons of achievement levels (cut scores) in each consortium’s summative test…” SMARTER Balanced Assessment Consortium. (2011, January 6). Supplemental funding budget narrative submitted to the U.S. Department of Education. p. 31. Retrieved from http://www.edweek.org/media/sbac-supplemental-budget-narrative_final.pdf p. 31

Producers of the SBAC tests have their set “cut scores” to report on four levels of performance. Level 1 signals failure. Level 2 indicates “at risk of failure.” Level 3 implies “safe harbor, doing well.” Level 4 means “proficient.” For students in grade 11, only Level 4 indicates readiness for entry-level, credit-bearing courses in college.

Across the grades, in math and ELA, about 11% of students are estimated to score at 4, the highest level and the one that indicates, in the eleventh grade, readiness for entry-level, credit-bearing courses in college. So, that will be very bad news, and it will make news when all of the test scores are gathered in.

The SBAC cut scores in math are estimated to assign 67% of grade 11 students to Level 1 or Level 2, with most (40%) at Level 1. In many states, students who score at Level 1 will also place teachers and administrators at risk of being fired, perhaps with the whole school in line for closure. In addition, many schools will just assign students even more test prep in math, at the risk of harming students’ love for learning and affinities for inquiries that are not driven by tests.

The cut scores for English Language Arts are estimated to place about 59% of students at Level 1 and Level 2, with about 32% at Level 1. Students with these scores are certain to be in the same boat, receiving more test prep. In states like Ohio that guarantee “proficiency” in reading by grade three, 62% of students are likely to fall short, up to 82% if that rule and the meaning of proficiency corresponds to SBAC’s Level 4 definition of “proficiency.”

In any case, the cut score issue is not a trivial matter given the high stakes that federal and state officials have deliberately and foolishly attached to tests, tests that are not “objective” and cut scores that are not “objective” but judgments–and these detached from any concern for the consequences to individual students and all public schools.

Proponents of the Common Core and tests are worried about the political fall-out when the test scores are released in a form that allows stack ratings among all the states that sighed up for the Common Core and have advertised that they have tests aligned with the standards.

They should be worried. Gurus of spin at the American Enterprise Institute suggest how to spin that news. They suggest that the scores should be played down, that news should avoid crisis rhetoric about poor performance. They recommend framing the testing outcomes as just another step on a path “to continuous improvement” in student learning.

That soft “slow-and-steady-as-we-go message” provides cover for policy makers who want to delay high stakes decisions based on these test scores but still use them as a baseline for judging gains in performance for the following year. This delaying tactic may buy time to reset expectations for learning, but it will not stop the obsessive use of test scores and relentless test prep than now dominates life in many schools.

For advocates of the “one size fits all” standards and tests, the comparability in scores from SBAC and PARCC tests means this: Every state that signed up for this grand and nearly maniacal experiment in standardized education will be rated as winners or losers by these supposedly “objective tests.”

The governors and the state education officials who signed adoption papers for this grand experiment in standardized education may be out of office, but current officials will be questioned about the results. Handling the political fall-out will be tricky, especially with an election season heating up, budget problems in many states, and the dueling minds and messages of politicians (notably Republicans, but also Democrats) who support or condemn the Common Core and tests, and many others who have no mindful views other than spin provided to them.

All of the investors in pushing the Common Core–Achieve, the National Governor’s Association the Council of Chief State School Officers, the Gates foundation and buddies along with venture capitalists–want to keep PARCC and SBAC in place, including Common Core “aligned” tests other than those from PARCC and SBAC.

Why? Do not believe the spin about the global economy and needed skills for the 21st century workplace, and all of the other sales pitches.

Test scores with high stakes consequences are the weapon of choice for expanding market-based education. The more students, teachers, and schools fail, the faster the collapse of public education.

As Diane Ravitch and others have said over and over, the cut scores for NAEP “proficiency” are high. NCLB never defined proficiency, Race to the Top did but only in terms of college and career ready specifications, not cut scores or tests.

Achieve has manufactured a “proficiency gap” that looks impressive to a casual viewer, but it is hot air. The proficiency gap is a version of the achievement gap. The kids and the teachers are the problem. The standards and the tests are perfect.

Something amazing is happening. Parents have discovered that they have the power to bring corporate reform to a halt. They do it by telling their children not to take the test. No test, no data. No data, no punishments for teachers, administrators, schools. Opting out is a vivid demonstration of the power of the powerless.

In Washington State, an unprecedented 48,000 students opted put.

Most of the opt outs were in 11th grade.

Carolyn Leith calls this an “educational uprising.”

If state leaders don’t listen to stents, the uprising will spread.

It is important to remember a few key facts about the Opt Out Movement.

Number one: It was created and is led by parents, not by teachers or unions. In New York, where 20% of the students refused the mandated tests, the leader of the state’s teachers’ union did not endorse opt out until a few days before the testing started. The organizations promoting the opt out were grassroots, unfunded, and parent-led.

Number two: The opt out movement did not arise in opposition to the publication or implementation of the Common Core standards. It was only when parents received the results of the first round of Common Core testing that they got angry and got organized to fight the tests. Recall that 70% of the students in the state “failed” the first round of testing. Parents in districts where almost all the children graduate from high school, and where most are admitted to four-year colleges were told to their astonishment that their children were “failing.” The parent rebellion started, and State Commissioner John King could not quell it. Secretary of Education Arne Duncan referred to the protestors as “white suburban moms” who all of a sudden discovered that their child was not as “brilliant” as they thought. What an insult!

Number three: In three administrations of the Common Core tests, a majority of students has continued to “fail.”

*In English language arts 2015, only 31.3% of students reached the “proficiency” level across the state.

*Among black and Hispanic students, the “pass” rate was less than 20%.

*Students in New York City almost matched the statewide average, but in the state’s five big cities, only 11% “passed.”

*Among English language learners, only 3.9% “passed” the ELA test. Appalling!

*Among students with disabilities, only 5.7% “passed.” Appalling!

*Achievement gaps between racial groups were unchanged over three years of testing and quite large.

The math scores were better than the ELA scores, but still only 38.1 “passed,” and nearly 62% “failed.” The corresponding scores for black and Hispanic students, English language learners and students with disabilities were far lower. Read the report.

Why are most students failing the Common Core tests in New York? In the past, a majority passed. Did the students get dumber? No. The developers of the Common Core tests decided to use a “cut score” or “passing mark” that was set beyond the capability of the students in each grade. They chose to align the passing mark with the National Assessment of Educational Progress’s achievement levels. (See here and here.)

That decision was made two years ago. At that time, Catherine Gewertz of Education Week wrote:

The two common-assessment consortia are taking early steps to align the “college readiness” achievement levels on their tests with the rigorous proficiency standard of the National Assessment of Educational Progress, a move that is expected to set many states up for a steep drop in scores.

“After all, fewer than four in 10 children reached the “proficient” level on the 2013 NAEP in reading and math.”

The NAEP “proficiency” level is not a pass-fail mark. It is not a “grade-level” mark. It is a level that represents solid academic achievement. I was on the National Assessment Governing Board for seven years, and I assure you that “proficient” represented work that I would consider to be an A or A- (the highest level, “advanced” is akin to an A+).

Please note that in no state other than Massachusetts has as much as 50% of students reached “proficient.” In no state have 60% reached the “proficient” level, and state scores have been calculated since 1992.

Thus, the developers of the Common Core tests chose a passing mark that they knew in advance would fail most students and would produce even higher failure rates among black and Hispanic students, as well as dismal passing rates for students with disabilities and English language learners. Based on NAEP, there is no evidence that harder tests and higher bars lead to smaller achievement gaps.

They set the bar so high that the tests are designed to fail most students. Do students feel motivated to work harder if they fail every year?

Parents figured this out, and they didn’t see why the state had adopted tests that most children were certain to fail.

And that is why there is an Opt Out movement. Parents do not want to participate in a system that is rigged against their children. They don’t want to be part of a system where their children’s test scores determine their teachers’ reputation, livelihood, and future. They want to bring that system crashing down and restore common sense to education.

Mercedes Schneider has followed the development of the new federal legislation to replace the failed No Child Left Behind. She is one of the few people in the U.S. who has actually read every word of both the Senate bill and the House bill.

She concludes in this post that there will be no federal sanctions for opting out. The Congress has made clear–in both houses–that it does not want the federal Department of Education to take an activist role in punishing states. Will states punish school districts where parents rise up in rebellion against high-stakes testing. Schneider thinks not.

However, I now think that if the House and Senate conference committee whose task it will be to merge SSA and ECAA into a single bill decide against the SSA blanket opt-out and go with the state-level opt-out provision in ECAA, the federal government will not sanction states, regardless of state-level opting out.

In other words, if according to the future ESEA revision, states are supposed to set their own opt-out policy and include as much in the future ESEA Title I funding application, and if a state includes no opt-out provision in its future ESEA application yet dips below the 95 percent of students completing federally-mandated annual tests, the federal government is not likely to strong-arm states with federal sanctions.

I believe the federal government knows it has gone too far in strong-arming states via conditions attached to federal tests. For example, both the SSA and ECAA revisions include language to limit the role of the US secretary of education. The current US secretary, Arne Duncan, has actively promoted and defended Common Core and its annual tests; with the backing of President Obama, Duncan has lured states into adopting Common Core sight unseen with the lure of Race to the Top (RTTT) funds; he has paid for two Common Core testing consortia, PARCC and Smarter Balanced; he has made it a condition of states’ RTTT funding to use student standardized tests to evaluate teachers, and via his NCLB “waivers,” he has cornered states into agreeing to institute Common Core and its associated annual tests as well as testing teachers using test results as a condition for avoiding having the almost all schools in all states declared “failing” according to NCLB.

So, the fact that major news outlets such as the Washington Post and New York Times are doing their best to chastise those who support opting out of standardized tests is not enough to conceal what is obviously a federal blunder to make annual testing the end-all, be-all of American public education.

In its August 15, 2015, editorial, the New York Times points to possible federal penalties for New York State’s failure to test 95 percent of its students. It also notes that parents’ opting out of tests “could damage educational reform… and undermine the Common Core standards….”

Gee, that would be terrible.

It seems that the plan in New York is for state officials to put the squeeze on superintendents and principals to encourage participation in future annual tests– and to not encourage opting out. But the opt-out movement is not driven by superintendents and principals. It is driven by parents who are tired of the toll that test-centric education is taking on their children, including the artificially branding of their children as failures and the state’s allegiance to this branding…

The reality is that opting out of federally-mandated testing is not going away and likely will only continue to gain momentum across years as increasingly more children are branded American public school failures.

Test-centered American public education has had its day, and based upon the growing appeal to parents of opting their children out of mandated tests, that day has more than passed.

There was no opt-out movement throughout the heyday of test-and-punish NCLB, but there certainly is one now.

Federal and state officials need to take the hint as they formulate a non-test-centered Plan B.

Jonathan Pelto wonders whatever happened to the Common Core test scores in Connecticut. Why hasn’t Governor Malloy’s administration released them. If the scores on the Smarter Balanced Assessment are similar to other states, Connecticut will discover that half or more of its students are “failing.”

Bear in mind that Connecticut is one of the top three states on NAEP. No matter. SBAC and PARCC set their passing scores so high that most kids will fail in most states. Diabolical or insane or incompetent?

The state’s Commissioner of Education blamed classroom teachers for growing public opposition to the tests.

Pelto writes:

“It what may be the most incredible, insulting, outrageous and absurd statement yet from Governor Malloy’s administration about the Common Core SBAC testing program, Malloy’s Commissioner of Education is now blaming teachers for the fact that there is growing opposition to the SBAC testing scam.

“In their warped world where “war is peace, freedom is slavery, ignorance is strength,” these people have the audacity to blame the victims for the crimes that are of the politicians’ making.

“Forget that the Common Core Smarter Balanced Assessment Consortium Test (SBAC) is unfair, inappropriate and discriminatory.

“Forget that the SBAC test is designed to fail the vast majority of Connecticut students.

“Forget that the SBAC test is particularly discriminatory for children who come from poorer backgrounds, those who face English Language barriers and those who require special education services.

“Forget that the SBAC test results are being used to inappropriately “evaluate” teachers

“Forget that state taxpayers have paid well over $50 million for this disastrous test program just over the past two years and local taxpayers have paid tens of millions of dollars more.

“And forget that the SBAC testing has wasted hundreds of hours of instructional time, time that our children could have been getting the education they actually need and deserve.

“Forgetting all that and proving that Governor Malloy’s administration has lost all contact with reality, the Commissioner of Education is now claiming that the lack of support for the Common Core SBAC tests is the fault of Connecticut’s public school teachers.”