Out of frustration, they’ll march

Anthony Cody organized disaffected teachers

Like no other, Diane Ravitch – author, polemicist, historian, Twittermeister – has galvanized classroom teachers to oppose the Obama administration’s vision of ed reform. And if she’s the James Madison or Thomas Jefferson of the rebellion, Anthony Cody is the movement’s Sam Adams.

With bulldog tenacity, the 24-year Oakland Unified science teacher and teacher coach has challenged teachers to speak out and take action – and put his organizing talents behind his words. Later this month, his two-year campaign will culminate in the Save Our Schools March and National Call to Action on Saturday, July 30 in Washington, D.C., with a rally at the Ellipse followed by a march to the White House.

“We have really tried over the last two years to engage the administration in dialogue .. ,” Cody told me. “So, yeah, we do feel like we need to protest at this point, because, you know, we really expected much better from this administration, and we still do.”

In a recent video interview (here for the transcipt), Cody laid out his grievances – chiefly Obama and Secretary of Education Arne Duncan’s embrace of the chief tenets of No Child Left Behind – and his goals for the march. He hopes that at least a few thousand teachers and sympathizers (he’s low-balling the number) will brave the potentially swampy Washington weather to attend. He and other volunteer organizers have lined up Ravitch, author Jonathan Kozol (Death at an Early Age), actor Matt Damon, and activist Deborah Meier (co-blogger with Ravitch in Bridging Differences) as speakers. Bookending the rally will be a conference on July 28 and 29 and a congress on July 31 to plan future actions.

The protesters will make four demands: equitable funding for all public school communities; curriculum developed for and by local school communities; teacher and community leadership in forming public education policies; and an end to high-stakes testing for student, teacher, and school evaluation. It’s clearly the last point that has fired up teachers and motivated them to travel to Washington. Cody argues, as do others, that fear of having their schools labeled as failures under NCLB has created an obsession with standardized tests, particularly in low-income schools where the curriculum has been narrowed to exclude subjects other than those that are tested, primarily math and English language arts. Low-income students, he says, “are losing the rich education that they really need. They’re losing the chance to be challenged, to think critically, and we’re developing a split education system” in which wealthy schools escape NCLB sanctions and can do the deeper learning denied poor schools.

Teachers who are feeling under siege are furious over the requirement in Race to the Top that states include student scores on standardized tests to evaluate teachers. Duncan and Obama have made more conciliatory statements this year, talking about the need for a richer curriculums  and multiple measures to evaluate teachers. But Cody says he hasn’t seen their more nuanced positions reflected in policies.

“I see them making rhetorical nods to how much they honor teachers; but you do not honor us by tying our pay to test scores, by tying our teacher evaluations to test scores,” he said.

In an explanation “A One-sided Dialogue on his popular blog “Living in Dialogue,” Cody lays out his disappointment and grievances. Eighteen months ago, he started a Facebook group Teachers’ Letters to Obama, in which he encouraged teachers to vent their frustrations in writing, for personal delivery to the White House. More than 100 did.

Then he and others pushed for a conference call with Duncan for them to explain their problems with NCLB. It did happen, after months of planning with a dozen teacher representatives from across the nation. But the call left Cody and others dissatisfied, with most of the half hour, Cody said, spent listening to Duncan instead of the other way around. The idea of a march took off.

Over the past two years of back and forth, Cody and I have sometimes agreed to disagree, but I admire his energy and savvy in creating his grassroots effort. He said he created Letters to Obama for less than $500. The Save Our Schools March has a $150,000 budget, with more than half raised from donations of $25 to $100. Both national teachers unions have kicked in $25,000 each. On Thursday, Ravitch will participate in  a fundraising webinar with a goal of raising $5,000.

“For the last decade, ‘education reform’ has been defined as No Child Left Behind, and the current administration picked that up, and is continuing to run with it,” Cody said. “We  are determined to show a different face for education reform.”


  1. If we can’t agree at the state or federal level what constitutes a good curriculum do we really have a public school system?  Shouldn’t I be able to send my children to any public school and be confident they would get a good education?  Isn’t that why we strive for equitable funding for all schools?  Doesn’t the public deserve to know what’s happening at schools so that corrective action can be taken?  There may be better ways to slice it and dice it, but the bottom line is that the education of children is a high stakes endeavor.  Just take a look at the difference in life trajectories based on educational outcomes.

    Report this comment for abusive language, hate speech and profanity

  2. Unfortunately, the honest and well-intended goals of SOS will come across as fear of accountability. I agree with Paul that shining a light on what kids can and can’t do it useful for focusing action.  This system only works, however,  if the tests  are credible and reflect a public consensus on what academic proficiency means.  Unfortunately, neither of these conditions exists.  Further, the state has no system for following student performance over time.  This is like going to the doctor and working without a medical record to guide action.  In medicine, there is a body of evidence and a public conversation about what tests represent, and a well-defined battery of tests that lead to a differential diagnosis.  No such structure exists in education.
    Consider two other comparatively well-accepted forms of testing. When I go to the DMV, I take both an isolated-concept multiple choice test and an authentic performance test.  This is high stakes because the outcome is meaningful to me and to the state.  Even though there are some bad drivers out there, most people agree that this system works reasonably well: it is quick, gives instant results, is tied to natural outcomes, and the public’s business is done in public.
    In academics, the  AP program is closer to the high stakes testing environment that worries so many people.  But by comparison, the AP program is the closest thing we have to a national curriculum. The curriculum is developed by well respected practitioners in various fields, we know who they are, courses are regularly audited, thousands of colleges and universities accept the results, and so there is a consensus of support behind the validity of this program.  None of these validating factors are true in our current testing environment.
    Testing in the current state is simply a proxy for a lack of public consensus on the goals of education.

    Report this comment for abusive language, hate speech and profanity

  3. Love the comparisons to the Founding Fathers! Thank you for this post, John.
    I’m also putting in a plug for Parents Across America, which is deeply involved in organizing the SOS event. PAA formed in 2010 to demand — with an open letter to Duncan and Obama — that the parent voice be heard in national education policy, and organized more formally with a launch event in New York City in February 2011. I’m a founding member who went to NYC for the launch meeting, along with parents from Chicago, New Orleans, Denver, Florida, North Carolina, Seattle, Oakland and more. www dot parentsacrossamerica dot org
    The phone “conversation” between Arne Duncan and the teachers organized by Cody is really emblematic of how dissenters against the national direction of ed policy feel our views are received (or not). As John writes, it was (I’m told by the disappointed participants) a one-sided conversation — Duncan talked, the teachers listened. It was a very literal way to make it clear to people that they’re being unheard.
    John, I’m curious about what kind of advocacy you feel would be effective, for those who feel unheard by the voices in power and those who speak for them in the press etc. — and also, as an editorial-board veteran, if you think there IS any way to reach editorial boards that have established their positions and seem (to the outsider, and even the former insider, since I’m a newsroom veteran) to brook no other views. With the newspaper business in its death throes, I know I’m giving editorial boards more power than they probably have, but I do think they still carry weight.
    And also, semi-related: Readers here who have not heard about the interesting Jonah Edelman fiasco might be intrigued to learn more about it. Edelman, son of revered children’s advocate Marian Wright Edelman, founded (with his mother) the organization Stand for Children (SFC) as a grassroots advocacy group in Portland. As many see it, SFC has become corrupted into an advocate on a national scale for billionaire-funded corporate education reform and privatization, and an opponent of teachers. At the recent Aspen Ideas Festival, Edelman gave a talk on how SFC moved into Illinois, bought off legislators and manipulated the teachers’ unions there into supporting legislation that severely weakened teachers’ job security and rights. Edelman’s talk has caused quite a stir on multiple levels; he has fruitlessly apologized and backed off some off his remarks. On the Parents Across America website, go to the blog and see “Jonah Edelman on Outfoxing Teachers’ Unions” and also “Stand for Children: a Hometown Perspective of its Evolution.” (Again, www dot parentsacrossamerica dot org.)
    That situation gives a good idea of some of the issues inspiring advocates to march in D.C.

    Report this comment for abusive language, hate speech and profanity

  4. Neither the DMV written tests nor the AP tests have stakes attached in a manner that’s comparable to K-12 public schools’ standardized tests. The comparison doesn’t work.
    It would work if individual DMV offices were judged based on the number of would-be drivers who passed the written test and/or their scores — with the DMV staff threatened with firing if the test results didn’t meet the set standard, the DMV office threatened with being closed down entirely, and the DMV staff offered pay increases if the test results met the set standard. This could get really interesting if it applied to the driving tests, too!
    And it would work if there were stakes attached to AP test results for schools and teachers the way there are to standardized tests. There are artificial stakes attached to AP tests in one area — that’s the Newsweek ranking of U.S. high schools created by Newsweek/Washington Post ed columnist Jay Mathews. That feature bases the school’s ranking on the number of AP tests taken per student — but not on their test results. And it has led numerous school districts to do what it takes to get the maximum number of students to take the AP tests — but since the results don’t matter, it’s not necessary to get them to succeed on the AP tests, just take them. Since AP tests carry a fee of $86 each, that’s expensive, but some do it. (The College Board drops the fee only for students who qualify for free/reduced lunch, which only covers the very, very poor.) I’m told that Bellevue, WA, schools — which are always among the very top in the Newsweek rankings — pay for all students to take the test, though since these are teens, there are always going to be refuseniks.
    Anyway, the AP situation does show how the stakes attached change behavior in a way that, many argue, doesn’t actually contribute to improved education. But neither the AP nor DMV comparisons works as an analogy to the stakes attached to K-12 standardized tests under No Child Left Behind, Race to the Top and the general currently popular education reform policies.

    Report this comment for abusive language, hate speech and profanity

  5. I’m not one to channel other people’s opinions to make my own point but it seems apropos that this was published on the same day…

    Report this comment for abusive language, hate speech and profanity

  6. I wish Ravitch would make up her mind as to what she (and by extension her followers) wants in a school system in concrete ways that can be implemented – as opposed to just being for some amorphous and anodyne “community of teachers” and reverting to the situation pre-NCLB.
    In her recent NY Times letter (http://www.nytimes.com/2011/07/10/opinion/sunday/l10dialogue.html) she claims to admire the Japanese and Finnish systems.  Does she have any idea what she is talking about? Any.idea.at.ALL!??  The Japanese system has a national curriculum set from on high, has about 40 students per teacher, requires every student to be in a uniform, is 220 days long with homework over the breaks, and does not require school attendance after the 9th grade.  The Japanese system has a nationwide set of exams for entrance into 10-12 HSs (which are NOT free).  The HS’s are ranked and you have to have a GPA + test score above a certain amount to get into a certain class of HS.  The test score is more important than the GPA.  (High stakes testing *indeed*, but for the student not the teacher.)  If you can’t make it into one of the public HS’s you tried for, you have to pay about twice as much to go to a private school.  Nonetheless, Japan’s k-12 drop-out rate is about 4-6% (vs. the US’s 25%). Books are bought by the student and as a consequence are cheaply made (sort of like the SAT prep books here) and stick to the subject with no pretty pictures showing diversity.
    The result is a very highly educated work force such that every HS graduate has calculus and/or statistics which Japanese manufacturer’s put to use by having factory floor work groups do  statistical quality control to maintain quality.  Another consequence is that 9th graders are a lot more mature than US students (“mature 9th graders” is an oxymoron in the US) since they have to think about academics and their future earlier since the HS you go to determines to a degree the university you go to and the life you will have.  Another consequence is that HS students may travel an hour on the train to their HS while another HS might be only a block or two from their house but either they couldn’t get in there or the further one is “better”.
    The other thing Ravitch harps on is national poverty rates as a determinant.  The poverty rate in Japan is 15%, in the US it is 12%.
    If the Japanese system is one of the school systems Ravitch admires why is she against a national curriculum, and why is she against high stakes testing for students and teachers?  Or maybe she didn’t bother looking up anything but just grabbed an academically high achieving country at random and added her standard rhetorical flourish about “community” and “respected teachers”.  We’re talking “research” as in _Google_ folks, not years of study and self-denial locked away in the library of the Sorbonne.  Japan changes as does the US so I may be a bit out of date, but not substantially.  I also added above some stuff I know from talking to Japanese and American people who experienced the Japanese system.
    Teachers are getting unfairly beat up as a cause of US education problems when they are in many cases almost as much the victim as the students.  Ravitch makes things worse with her retrograde conservatism and just plain *awful* lack of research.  She should quit relying on telling teachers what they want to hear about some mythical golden age before NCLB and start looking at what other countries do that we might realistically be able to do.  Accountability and  testing are here to stay – deal with it and make them better for everyone.

    Report this comment for abusive language, hate speech and profanity

  7. Anthony Cody says “You do not honor us by tying our pay to test scores, by tying our teacher evaluations to test scores.”

    Okay.  If student test scores are not to be used for teacher evaluations, then what criteria should be used for teacher evaluations?

    Test results can’t be all or nothing  based on raw scores alone, especially for teachers with students who enter the class year with low skills.  But is Mr. Cody suggesting that student test scores should have nothing to do with teacher evaluations?

    Nothing at all?

    Really now… if several teachers have similar class demographics and over a period of years some of those teachers consistently lead students to higher proficiency than other teachers, isn’t that data that all education leaders (teachers and administrators) want to know?

    John, what criteria does Mr. Cody practically suggest?

    Chris Stampolis
    Trustee, West Valley-Mission Community College District

    Report this comment for abusive language, hate speech and profanity

  8. Movements generally succeed or fail based on how well they can articulate a coherent, compelling message to a wider public, and this one does not look as if it is off to a good start.  As much as I sympathize with Mr. Cody’s frustrations (for other reasons, sometimes), his four key points are not going to mobilize any significant constuencies outside the ed establishment.

    “Equitable funding for all public school communities”  When has this ever happened?  Even in the states with the most rigorous “equalization” policies, parents figure out ways to supplement the funding for their own children’s education.  Even in the Soviet Union, and certainly in present-day China, the people with access, connections, and resources have always found a way.  What, specifically, would Mr. Cody et al propose to do to finally make sure that funding for every child is “equal”.  (If that should even be the goal.  We know that in states that have really pressed “funding equality”, students in high-poverty areas often receive higher amounts of public funding per student when all sources are taken into account, because public policy at the state and federal level directs specific pots of money directly at those populations.  Then the question becomes, “How much extra funding should those students receive for the overall funding to be “equitable”.  Mr. Cody’s own district spends a lot more per pupil than my own suburban district, presumably because the aggregate needs of Oakland’s students are greater.  But do they spend enough extra for their funding to be considered “equitable”?)

    “Curriculum developed for and by local school communities”.  Who are these local “communities”?  Would Mr. Cody’s concept include parents?  Teachers?  Site councils? Or just the administrators in the local monopoly district (which sometimes makes great decisions in small, homogenous suburban districts, but often makes disasterous decisions when done on a large scale for diverse student populations)?  I’m no fan of Common Core.  I actually think California’s current standards are very good.  But those are just standards.  Choosing among all of the curriculum options that purport to meet these standards is no small undertaking.  If Mr. Cody will take it out of the hands of the often ill-informed (and even corrupt) central district administrators I have dealt with, I’m interested.  But who, then, would make these decisions?  I’ve seen curriculum decisions handled quite expertly at the level of individual schools (typically private schools, where that is permitted and where the job is easier), but it is a whole other ball game when an entire “community” is in the driver’s seat.

    “Teacher and community leadership in forming public education policies.”  I like having teachers involved, since they are at the front lines dealing with students every day, but how would they be represented?  By their unions, with all of the non-student-centric (or even student-destructive) “policies” that often dominate their agendas?  And again, who would represent the “community” that he has in mind?  The mayor and city council, perhaps? Mr. Cody does not sound like an advocate for any kind of parent-oriented School Choice, so he must have someone in mind to micromanage education, in the absence of any kind of choice-based accountability.

    “An end to high-stakes testing for student, teacher, and school evaluation.”  As much as people understandably complain about using those test scores for teacher and school evaluation, Mr. Cody would end high-stakes testing for student evaluation?  Yes, there is widespread dissatisfaction with standardized tests among the education cognoscenti, but what evaluation system would Mr. Cody propose?  If he is leaning toward generalities about “portfolio evaluations” and “authentic performance measures”, those are not going to cut it, because no one can agree on what those mean when applied beyond the individual classroom, to large groups of students.  Compared to many other countries, we actually have a notably “low-stakes” testing regime for most students.  There are many legitimate criticisms of how we develop, administer, and intrepret our standardized tests, but if Mr. Cody thinks he can develop a critical mass movement for eliminating standardized measures for evaluating student achievement across multiple student populations, he is seriously alienated from  the public at large.

    If teachers like Mr. Cody want to seriously influence the public at the grass roots (or even astroturf) level, they are going to have to articulate positions that people outside their profession can understand and discuss.  The public can understand tangible things like per-pupil funding, number of school days, teacher salaries, and so on.  But for the more esoteric topics like curriculum, assessment, and “equity”, I’m afraid that teachers are going to have to fall back on the tactics that have worked so well for them in the past:  big-time payola and lots of legwork for the politicians who are willing to do their bidding.

    Report this comment for abusive language, hate speech and profanity

  9. The comparison of DMV tests may well be a valid one, but with a different point of comparison.   DMV tests are not federally devised, funded, administered or enforced.   DMV is a state operated entity.   Marches on Washington weren’t deemed necessary before the Federal Government intruded itself into the K-12 scenario at every turn.   It is late, but where is the logical call in the education community to eliminate the Federal Dept. of Education which is the fountain head of the destruction of a system which was the envy of the world prior to ESEA?   Precious time, effort and frustration on the part of both parents and teachers continues to be spent to address the problems created by   Washington.   How about a march on Washington calling for the abolition of the Department of Education, repeal of ESEA, and the accompanying thorns in the flesh of education such as NCLB could be left to whither on the vine?   Unless the source of the agendas designing the straight jackets of education are removed, all the marches and swatting at Federal gnats won’t solve the problems inherent in nationalized schools. 
    Unemployment insurance payments for the bureaucracy and accompanying regional networks
    associated with it, would be minor compared to the funding they consume.  In times of deficit
    crises, it would be a positive all way around.

    Report this comment for abusive language, hate speech and profanity

  10. Regarding the publishing of school test scores:  Accountability is something that no one enjoys, but everyone needs.   Our children deserve teacher and school accountability.

    Report this comment for abusive language, hate speech and profanity

  11. Michael G. — I don’t think it’s a matter of Diane Ravitch “making up her mind”.  Has she ever offered concrete solutions? (Oh, right, we need to end poverty before all kids can learn. Thanks, Diane. We had no idea that poverty mattered.)  Her current career seems to be focused on making educators feel better about themselves.  All of her platitudes undoubtedly generate lots of speaking engagements at the hundreds of available industry gatherings — who doesn’t want to be told that they are actually fabulous? — and ginning up a profession-wide persecution complex may also yield some nice “professional development” consulting gigs.  But when has she ever had anything detailed or constructive to offer?  By all means, every one of us should take every opportunity we have to acknowledge and thank the good teachers that we know for the work that they do, but a national campaign to enhance teacher self-esteem is not going to help anyone (except, of course, Ms. Ravitch and others who have latched on to her celebrity).  Teachers, as members of a profession, will see higher regard from the public when the system they are part of shows more regard for the students and parents who should be the focus of the system.  Many teachers already understand that being the daily “face” of an unaccountable, monopolistic system is no way to enhance one’s professional stature. 

    Report this comment for abusive language, hate speech and profanity

  12. Jim Mills – I completely agree with you (as always).  But I get so annoyed at the success of Ravitch’s demagoguery that I feel compelled to point out she has no idea what she is talking about whenever her name comes up.  Some variant on OCD I suppose.  Maybe the pharmaceutical industry could help.
    (BTW, in my earlier post I meant “grades” 10-12 HSs not that there are only 10-12 HSs in Japan.  Japan’s school system goes K-6, 7-9, 10-12.)

    Report this comment for abusive language, hate speech and profanity

  13. The problem with our current testing models is that they measure the wrong things in the wrong ways for the wrong reasons.  Teacher performance and accountability are too complex to be measured fairly via student test scores.  Well trained scientists should know that there are too many uncontrolled variables to draw meaningful conclusions from this data.  Assessment of teachers needs to be qualitative.  Efforts to quantify the complex suite of behaviors exhibited by successful teachers by using their students’ test scores is reductionist.  Effective teaching involves a highly nuanced set of human interactions within the classroom, and the skills necessary to orchestrate successful educational outcomes are not measurable by multiple choice tests.

    Report this comment for abusive language, hate speech and profanity

  14. So many good comments this morning — all insightful even where they disagree.

    One reason I started the Contributors section in the blog was the recognition, as Caroline notes, of  the shrinkage of the traditional op-ed page. There needs to be more forums where people can come together in civil dialogue. I am not as pessimistic, though, and disagree with the perception of teachers and activist parents like her that they’re outgunned by well-funded groups like Stand for Children and StudentsFirst. I see only voices of the Counter-Reformation on Huffington Post, and I am impressed by the power of Facebook, Twitter and blogs like Anthony’s to let a thousand flowers bloom.  Valerie Strauss of the Washington Post carries the water of the opponents of ed reform. And Caroline, you are as prolific a writer as I have seen and have the weight of your Parents Across America behind you. That should get you in the door to what you call the MSM.

    Michael G, I appreciate your criticism of  Saint Diane, a wonderful writer but also a historian with amnesia who is often too quick and glib with facts. Because she is persuasive, she has the potential to do, as her blog says, bridge differences. Instead, she has become another divisive force in a polarized world of education politics, the anti-Rhee. That’s a shame. It’s clear what she doesn’t like (at least lately) but less of what she does . Nor does she recognize the accurate criticisms of the union politics and their defense of contracts with indefensible provisions.

    But as for unions, CapitolReader, thanks for the link to Jay Greene, who also makes the mistake of seeing everything through the prism of union vs reform politics. In doing so, he misses the importance of the march, as a heartfelt expression of classroom teachers’ anger and frustration. It’s a grassroots effort, not a union effort, notwithstanding the contributions of AFT and NEA (nothing compared with their political donations).  Hear them out: They have something important to say about what’s happening in their classrooms.

    Which gets to your point, Chris. I won’t begin to speak for Anthony, and our 10-minute interview (as limited by YouTube) didn’t get into his views on reform. My bet is that Anthony is a terrific teacher who conveyed to students the wonders of science through hands-on learning and has guided a generation of teachers in Oakland through his mentoring. But, as my post implied, I have basic disagreements with him on issues of the components of and the importance of teacher evaluations, school accountability and charter schools. Education has become a battleground, and kids have become long-suffering Poland in World War II. Anthony, too, has chosen sides.

    Report this comment for abusive language, hate speech and profanity

  15. a letter I sent to the Times a while back:

    Dear Editor,

    Your reporter, Seema Mehta says Arne Duncan was in California on a “15-state listening tour.” [U.S. Education Secretary Says California Students In Peril, May 23, 2009]

    Yet, I discover that Duncan warned, challenged, said, repeatedly told, said, said, told, and warned.

    Then Charles Weis, “superintendent of Santa Clara County schools and the president of the Assn. of California School Administrators,” raised a gnawing concern.

    So Duncan demurred, and Supt. of Public Instruction Jack O’Connell later said he feared.

    So, Duncan challenged, said, assessed, slammed, also said, said [again], then met privately with state officials to discuss.

    We are then told that Duncan called for, said, noted and also spoke.

    Finally, in response to a child’s question he modeled mastery of the English language by explaining, “We do pretty good.”

    Obviously, he wants CA to do gooder. But when does his listening begin?


    Richard K. Moore
    Huntington Beach, CA

    Report this comment for abusive language, hate speech and profanity

  16. Here’s a key comparison.  Our public primary and secondary schools do a relatively poor job compared to many other countries.  Yet our universities are the envy of the world, head and shoulders above those in other countries.  For some reason our 12th grades are weak while our 13th grades shine — in teaching the same kids, at least among the college-bound.
    The obvious difference is the great variety of choice among colleges.  If viewed as unsuccessful in educating students, a college will lose attractiveness and may face real organizational consequences through loss of applicants, or a decrease in their quality.  A few even close.  Even where colleges are in pretty good shape and have a good reputation, that pressure is always there in both a negative and positive way — to enforce accountability at some level, and to reward successful innovation.  This is true whether they are public or private institutions.
    These are important debates about how to achieve and measure educational quality in public schools.  But so long as most public schools operate with effective monopoly franchises over most of their students, there will always be a lack of incentives for local school administrations and unions to make tough or innovative decisions to benefit students, and there will always be a lack of ability for any state or federal government to make centralized decisions about accountability that entirely make sense.  The argument between “you’re not performing” versus “you don’t really understand our circumstances” can go on forever, since both sides will always have some version of the truth they can defend.  Meanwhile, more generations of students (especially the poor and disadvantaged) go through the system without getting the capabilities to which they should be entitled.  And we all lose, but especially those students.
    We can cut through much of this by giving parents more real choices about where their kids can go to school — whether it’s charters, vouchers, more freedom of choice within and between districts, or whatever.  The key is that good performing schools need the ability to prosper and reward their employees, and poorly performing schools need the ability to fail and cost their employees pay and jobs.  Parents will make that determination as the customers, just as they evaluate the quality of most other products and services to which their kids are exposed.  (If you can pick a dentist, I think you can also pick a middle school.)
    Only in that way can we line up a full incentive to perform along with the local knowledge and abilities that teachers and administrators have about how to do so.  In that system, the excellent teacher becomes a precious asset, a truly skilled administrator’s worth becomes clear, and a poor performer in either category places everyone’s livelihood at some risk.  That’s as it should be, and it doesn’t require a federal mandate to make it work.

    Report this comment for abusive language, hate speech and profanity

  17. No, it doesn’t get me or my views in the door of the MSM, and the weakened and battered MSM still carry weight. Nor does the fact that my friends, ex-colleagues, acquaintances and such are included among the voices of the MSM get me in the door.

    When I see every large paper in the state run essentially the same editorial supporting the Parent Trigger and blasting the “powerful teachers’ unions” – with no indication that they comprehend the key issues — how does this pack mentality happen? Do they just all read the same press release and obediently go “OK! If you say so!” …? Those aren’t the people I thought I knew and worked with.  

    Meanwhile, the dissenting voice does NOT have an equally prominent way to respond — there simply isn’t one.  If I sent a press release, would they all read it, obediently go “OK! If you say so!” and write up my views as a “voice of the newspaper” editorial? How does that happen; how do they decide, so often in lockstep?  There are so many questions.

    Report this comment for abusive language, hate speech and profanity

  18. No, Carl D., you’re missing the main point. College and universities choose who is admitted; our most challenged students, those most severely afflicted by poverty, the troubled and damaged, and the “intentional non-learners” don’t apply.

    Our nation has a child poverty rate far higher than that of every other industrialized nation. That’s the situation that poses such a challenge to our public schools.

    Yes, more and more low-income students are attending our colleges, but they’re the higher-functioning students (as selected by the colleges).

    Report this comment for abusive language, hate speech and profanity

  19. Thanks for your thoughts.  I think the studies of lottery charter schools (where students are admitted by lotteries due to more demand than there are spaces) rebut the point you are making by showing that choice can greatly benefit the low income population, perhaps more so than other students who are more advantaged.  You might look at economist Caroline Hoxby’s research on NYC, for example, where many charters have specifically adapted their programs to meet the needs of kids with low-income working parents.  A greater benefit for disadvantaged students also makes sense because higher-income families can move more easily to find better schools, and use their own resources to supplement what’s provided by the public system (which in CA can be pretty poor at times even in “good” districts).
    Your response would also suggest that our public schools do quite well in teaching college-bound students, since you point to the qualities of the students as the determining factor.  However, some commentary seems to suggest to the contrary that our college-bound students are more poorly-prepared than in the past, require more remedial courses and so on.  Is there evidence you can share as to how the U.S. compares to other countries in preparing (say) the upper half of its student population?

    Report this comment for abusive language, hate speech and profanity

  20. Boy, Ravitch’s big problem is her “vagueness.” Right.

    RE: Jim Mills

    ” I like having teachers involved, since they are at the front lines dealing with students every day, but how would they be represented?  By their unions, with all of the non-student-centric (or even student-destructive) ”policies” that often dominate their agendas? ”


    RE: John Fensterwald

    “Nor does she recognize the accurate criticisms of the union politics and their defense of contracts with indefensible provisions.”

    So, just what are these “non-student centric” and “indefensible postions, ” gentlemen? And please be specific.

    Report this comment for abusive language, hate speech and profanity

  21. I agree that low-income students benefit from being in a school setting from which their more troubled, challenging, high-need peers are excluded.  I think that’s an important lesson we can learn from charter schools — if only they and their supporters would be honest about it and stop trying to deny and conceal it.

    That’s a benefit to those particular students in those charter schools that are functioning successfully. I think there are so many downsides to charter schools that the movement needs a total do-over, however. That’s a quick summary of my view.

    Carl D., your own praise of U.S. higher education speaks for itself as to “how the U.S. compares to other countries in preparing (say) the upper half of its student population” for college.

    As to the claims that our college students are more poorly prepared than in the past:

    In the past, only our top students and elites were expected to go to college. There wasn’t even an expectation that it was the norm for all students to graduate from high school, let alone the faintest notion that all students should go to college.

    For that matter, I don’t know of any other developed nation that doesn’t track its students into entirely separate college vs. vocational programs.

    Today, we in the dreamy-eyed U.S. claim to expect all students to go to college — and we brand the students (and their K-12 schools) failures if they don’t go to college.

    It’s obvious that that huge cultural shift would result in a larger number of unprepared, less-academically able students going to college. Since no other nation has succeeded in changing culture, educational systems,  and human ability deviations to enable all students to go to college (if any have even tried, which seems unlikely) — it would be astounding if the U.S. were able to succeed at that impossibly pie-in-the-sky goal.

    So, of course a greater number of unprepared students are starting college. It would be logically impossible for that not to happen.

    Report this comment for abusive language, hate speech and profanity

  22. http://www.nytimes.com/2011/07/18/education/18rand.html?_r=2

    The above is a link to a NYT article re one more failed attempt to “incentivize” teacher performance via test based “accountability” measures. Those teachers who are not self-incentivized and who are not accountable to their own personal and professional  standards leave the profession. Almost one of two teachers leave the profession voluntarily in the first five years of teaching. The classroom is a very uncomfortable place for adults who lack the internal motivation to teach.
    That keeps being demonstrated over and over, yet the facts sway few opinions in the world of the neo-liberal public school critics.
    The highest scientific body in the nation, the National Research Council, and scores of other scholars back the positions of Diane Ravitch that the abusive testing regime imposed on students and teachers is having  narrowing, destructive, and corrupting effects on the schools.
    The Obama Administration along with the state of California have dedicated almost half a billion dollars to replace the current tests because they are simplistic and unreliable. Then people want teachers to have their evaluations, employment rights, and compensation tied to the scores on these same lousy tests. And if their unions move to defend teachers from the travesty they are “afraid of accountability.” Get real.

    Report this comment for abusive language, hate speech and profanity

  23. The idea that there was some golden age of American education where every child got a better education than our schools present today is utterly ridiculous.
    We didn’t measure kids who were disabled, or the wrong color. Girls who learned to type were well educated. Kids who cared about school and who were the correct demographic in the right neighborhoods did OK. Other kids fell through the cracks and no one cared.
    In the 1980′s, the elite kids took algebra in 8th grade – maybe 8% of the total school population. Today in California we are expecting every 8th grader to take algebra. The net result is that far more Calfornia kids take – and pass – algebra than did 30 years ago, and at younger ages, even though not every student does. (Perhaps, however, a policy in between these two might be optimal.)
    30 years ago, minority kids who enrolled in elite math classes were often rescheduled into woodshop and Consumer Math. 30 years ago, minority kids might attend a school that didn’t even offer those elite classes. It’s only since the mid 1970′s that disabled kids were fully entitled to an appropriate education.
    I look around and I see nothing that was better in education Back In My Day. Not. One. Thing. My daughter’s education is more rigorous and more engaging than mine, by far – and I was in the top 2% on the SAT.
    We have a long way to go for equity for all and perhaps more relevantly, for getting our kids to really want to be educated, and to find ways to work with the kids who don’t come ready and excited to learn. But it’s not that we’ve lost greatness, only that our sights are higher than they have ever been… and rightly so.

    Report this comment for abusive language, hate speech and profanity

  24. Gary: Including state policies that unions helped to pass and now defend: 2 year tenure;  step and column pay system that rewards longevity  and often meaningless academic credits, not accomplishment; strictly last in first out layoff system combined with bumping system still in place in many districts; defense of an ineffective evaluation system and reflexive resistance to any use of student test scores in the evaluation; inflexible scheduling that discourages innovation.

    Report this comment for abusive language, hate speech and profanity

  25. Mr. Fensterwald:
    I would like to refute every one of the points you have raised:
    1) 2 year tenure is not “tenure”, that is, a sinecure. It is protection from being fired at will and receiving due process instead. If the administration cannot exercise its prerogative to establish non-performance, why should the teacher unions be blamed?
    2a) step and column pay system: the system is not much different than any other way of advancing in any organization or public agency. How else are school district going to retain experienced teachers? And why should teachers be demonized now for that system? Teachers are not, as far as I know, solely responsible for its creation.
    2b) …system that rewards… not accomplishments: what kind of accomplishments do you have in mind?
    3) FILO: that is the primary method of keeping personnel anywhere there is a skilled workforce and no “at will” firing. Why do you want to replace it with an arbitrary system? Do you want to return to good ol’ days when teachers were fired when pregnant?
    4) defense of an ineffective evaluation system: as far as I know, every teacher is unhappy with the way they are evaluated, not because of its arbitrariness but because of its lack of feedback and collaboration. And many teachers have said so in many reports, articles, etc.
    5) reflexive resistance to any use of student scores: teachers are very aware that current testing does not correctly assess student achievement because of what they see in their classroom day in and day out. So why should they agree to be evaluated by an almost meaningless test score that is highly dependent on whether the student had a good day?
    6) inflexible scheduling that discourages innovation: a good teacher spends a minimum of 7 hours at the school site and anywhere from 2 to 4 hours daily offsite working on lessons, preparing, grading, etc. And often at least 6 to 8 hours on a weekend day. No, nobody told me: those are the hours my wife puts in as a first grade teacher. What more do you want? Is your work schedule as grueling?
    The test. In California, the test is a normalized test that defines the average as its proficient cut off point. And it has been so since 2002. Did teachers design the test like that? Of course not. Can they collectively move the “below average” (aka non-proficient) students to “above average” (aka proficient)? They can move some, but not the entire population because such a test is designed to be a zero sum game: if someone moves to the right of the average, someone else must move to the left.  It is in the interest of the state to not let the public know that their children are being graded on the Bell curve. Therefore, I believe you have no idea what the actual score distribution is because you have never seen it. If you were a true journalist you would have already demanded to see those score distributions to make sure that usage of the test is valid. (For instance, did you know that the “failing” API of 800 used in California can only be achieved if the school has more than 50% of students above average? This means that at least 50% of schools in California are failing at any given year. Would you say that is reality or a test artifact?) But, no, like every other pundit I’ve ever read, you have taken the word of the test makers/promoters/educational bureaucrats that the test is a fair test and truly measures academic achievement and every student who takes it does as well as s/he could do on it (guess you have never been in a high-school classroom when the CST in math is being administered). Is this a manifestation of mathematical intimidation? Or the even more common, math anxiety?
    In my experience, nobody in a position to do anything about it seems to want to touch this issue because it would literally bring down the house on their heads. Plus it provides a very convenient vehicle to pay lip service to the idea that everyone can be above average (and every one can go to college and grow up to be President).
    Meanwhile, people attack Ravitch for not offering concrete solutions while defending a testing regimen that is based on the same use of statistics as eugenics (yes, the Bell Curve!). And they are shocked, shocked, when testing scandals like Atlanta, DC, Philadelphia, etc., come to light. And they blame security not the test administration, design, etc.!
    How does that happen, indeed.

    Report this comment for abusive language, hate speech and profanity

    • Manuel: I guess we journalists — or whoever provides your information — are failing you, because you are simply wrong. California Standardized Tests are not done on the bell curve and are not normalized. It is a standards-based test with the official goal being proficiency for all kids. There is nothing in the way the test was constructed to prevent all students — 100% for each school and each district in each grade and content area and each subgroup — from scoring above the proficiency or advanced cut score. Some schools come close to having all students proficient. You are incorrect in saying that 50 percent of schools must “fail” every year — be below 800 on their API score. 800 is the state’s goal, but if every student were proficient in a school, the API would actually be 875.

      As for a few of your other points, California is one of a few states that have 2-year tenure (actually1 1/2 years, since the decision must be made by March of the second year). I’m not opposed to granting due-process rights, but with those rights must come the assurance that the teacher deserves that life-time protections. As important as the length of time is an evaluation process that is fair to the teacher, including the opportunity to get feedback and get better. Many teachers hit their stride after three or more years. With so few administrators in our schools to do thorough evaluations, they need more than a year and a half. I favor having teachers help design evaluations, participate in doing them and even score them, though dismissal should remain the principal/district’s prerogative.

      I don’t hold up pay scales for sheriffs, prison guards and other public employees as the right gauge for teachers. The proposed evaluation system that the UTLA is fighting tooth and nail in Los Angeles Unified is a system worth examining. It would include peer and administrator observations, school and community involvement, school leadership, parent surveys and student test scores. I would minimize the importance of the latter — 15 to 20 percent of the weight — but certainly would include them and other measurements for subjects not tested. (Teachers, you suggest them.) Teachers who year after year simply cannot move their students toward proficiency and mastery — whether from below basic to basic or basic to proficient or proficient to advanced (if they have a class of high achievers) are not being effective. There needs to be objective measures to guide improvement or, eventually, dismiss the teacher.

      I hope you have a better justification for first-in, last-out than saying critics want to return to the era of dismissing teachers who are pregnant. Many of your younger peers are frustrated the current system, which they are right in saying fails them.

      Report this comment for abusive language, hate speech and profanity

  26. Here’s a pretty good example of how impossible it is to put the horse back in the barn when the press reports bull$#!+ as fact.
    Every major newspaper and its cousin has reported on the claim that today’s 12th-graders are idiots and can’t pass a geography test.
    Kevin Drum of Mother Jones actually violates the sacred “check it and lose it” rule and takes a close look at the test. But trust me, the debunking of this mass BS about what morons our kids are will go nowhere — a lie travels halfway around the world before the truth can get its pants on and all that.
    This site sabotages all posts that include links, so I’ll try to make this one easy to piece together.
    [http://] motherjones [dot] com [/]  kevin-drum/2011/07/are-you-smarter-12th-grader
    (Why is it we love stories about what morons our kids are? Does any other culture love to assail their own kids as morons?)

    Report this comment for abusive language, hate speech and profanity

  27. Considering how the favored reforms of the past decade have yielded a steady stream of failed pay schemes and incentives, unworkable and misplaced notions of accountability, mixed results and minimal growth in “achievement” measures, a demoralized teacher workforce, and systemic scandals, I would think that the SOS March and its agenda might be received with a bit more generosity of spirit by some of the commenters above.  Who’s defending the status quo now?  Anthony and others in this effort are advocating for a vision of a public education system with principles and priorities that have been missing for the past decade and beyond.  Anthony’s blog and his body of work in Oakland and in various professional organizations all provide testament that he has specific ideas about improving education, and bringing real accountability for all education stakeholders – especially those who make the flawed decisions that teachers are forced to implement.   Is there an implied defense of failed federal policies in the criticisms here?
    I would also add that while it is fair to ask “what’s your solution?” when someone criticizes a policy, the answer has nothing at all to do with the validity of the criticism.  I can hear if a symphony is poorly played, and I can criticize it, but the fact that I don’t suggest how to improve the orchestra doesn’t mean that I’m wrong about their performance.   I bring that up NOT to suggest that Anthony Cody or Diane Ravitch are short on ideas for how to improve education, but rather to suggest that we are so bombarded with bad ideas in the policy sphere right now that it’s practically a full time job just pointing out how utterly stupid it is to tie bonus pay to test scores.  How many failed systems, how many testing scandals, how many behavioral psychologists, economists, and educational measurement experts will it take to convince some people?  It’s getting to be comparable to trying to talk to creationists about evolution.   So, even when we summon the time and energy to articulate a better way, we end up dragged back into this mess.  Getting back to Anthony, he has been part of writing and leadership efforts through Teacher Leaders Network and Accomplished California Teachers to write about and talk about better evaluations, better compensation systems that DO NOT strictly rely on seniority and accrued units.  If his critics would take the time to be as well informed about him and his ideas as he is about everything he’s marching against, if they would read more widely and learn about the GOOD ideas and more promising innovations that educators are putting out there, we might have a more substantive conversation.
    Disclosure: Anthony is a colleague and friend.

    Report this comment for abusive language, hate speech and profanity

  28. Our sheriff’s department also uses a step-and-column type payment scale, and Last In First Out for layoffs. It’s a pretty typical arrangement.
    I think it’s interesting that people think a 2 year probationary period for teachers is too short. Did they want to keep ineffective teachers in place longer than that before letting them go?
    Meanwhile, teacher’s unions are pushing also for small class sizes, for anti-bullying measures, and for the DREAM Act.

    Report this comment for abusive language, hate speech and profanity

  29. David: Your group, Accomplished Teachers of California, of which Anthony is a member, did create a comprehensive, smart alternative plan for teacher evaluations, which I wrote about a year ago — http://toped.svefoundation.org/2010/06/28/smart-report-on-teacher-evaluations/ — as you know. It has not gotten the attention it deserved, but the CTA has hardly embraced it (I do credit President Dean Vogel for moving the ball forward on the issue a bit, against resistance.). Until it takes becomes serious about  this and other issues, like the step and column pay system, holding back the development of  the next generation of teachers, it will be subject to just criticism and bad ideas imposed by misguided legislators. And yours will be a voice in the wilderness.

    Diane Ravitch could but chooses not to nudge the union and encourage teachers to take a more progressive position. It’s a poor excuse to say that she’s too busy being dragged back into this mess.

    Report this comment for abusive language, hate speech and profanity

  30. I’m speaking up in defense of Ravitch too. This kind of criticism is desperation nitpicking. It beats out-and-out viciousness (which is surfacing elsewhere), but is still graceless, fruitless and amounts to grabbing at straws.
    On a much smaller scale, I’ve gotten the same grief during the many years that I’ve been a researcher and critic of education reform fads: “Don’t you have anything better to do?” “Don’t criticize unless you can present a better solution.” “Why don’t you go volunteer in a school?” — and lots of snide-edged comments about being prolific, an effort to characterize that as somehow suspect.
    And even with a tiny potato like me, these folks research what they can about my personal life and try to dig something up. Green Dot/Parent Revolution’s PR firm, the Rose Group of Culver City, sent newspaper editors a letter describing where my kids go to school in an attempt to characterize me as a hypocrite. Edison Schools did similar things 10 years ago.
    I hate to think what these desperate and ethics-free forces are trying in efforts to discredit Ravitch, a nationally prominent and highly effective voice. It’s hackneyed to repeat the Gandhi quote, but we’re now in the “…then they fight you…” phase. The next phase, of course, is: “Then you win.”

    Report this comment for abusive language, hate speech and profanity

  31. Although the tests are not set on a bell curve per se, they are renormalized every year. Questions that too many students get correct are dropped from the exam. If every child rated proficient on the exam , no one would say, “Hooray, our students are learning and our teachers are teaching these important concepts well.” Instead, they’d say, “Man, we must have made the test too easy.”
    These exams are rigorous and have far higher expectations than we had 30 years ago. I’m in favor of that. But, we should appreciate that our bar for 4th grade proficiency (et al)  has risen considerably since we were students, and thus giving us a very false sense of decline. Sample exams are available online (I won’t link to them because a link will hide my comment), and I encourage all adults interested in California education policy to sit down and work through a few of them.

    Report this comment for abusive language, hate speech and profanity

  32. My understanding, EL, is that minor changes are made yearly to introduce new questions and retire some of the questions but this is done to ensure comparability from year to year. The fact that elementary API scores in particular have risen over the past decade are certainly to teachers’ credit. There is no effort to hold back rising test scores of students by making tests harder.

    Report this comment for abusive language, hate speech and profanity

  33. Mr. Fensterwald: thank you for your reply.
    You are failing me because I chose to believe exactly what you wrote until I went to look at the data itself and aren’t journalists supposed to be watchdogs for the rest of us?  Yes, I took the officials at their word until I realized that the scores ranges defining the achievement levels were not linear. Nothing like, say, 90% and above equivalent to advanced, 80%-90% proficient, etc.. Then I started looking for histograms of the scores. Nothing. Until finally someone in the know told me that the data was presented in tabular form in the Technical Reports published by the CDE in the web.
    After I plotted them all (yes, it took some time, but it was doable), it was clear as day that ETS has done a remarkable job in creating a standardized test masquerading as a criterion-referenced test. The distribution of ELA CST scores for all grades has been remarkably constant in 10 years of administration. The math CST is similarly good up to grade 7, higher grades go to hell for a variety of reasons, but chief among them the requirement that all students must pass Algebra I to graduate from high school.
    If the SBoE was really interested in 100% proficiency, they would have done something about this poor level of achievement by 2007. Instead, they upped the ante and authorized in May 2010 similar cut-off points for the STS, and CMA tests. And compounded it by allowing the CSTs to be the tests until 2014 by extending ETS’ contract.
    I can perfectly understand why you won’t take my word for this: it threatens all that you have come to rely on. And I could not believe it myself either until I looked at the data. So there is a simple solution: take a look at the Technical Reports yourself. For instance, the ELA scores for 2007-2010 are in pp. 605-607 (actual pages are 621-623) of this document, California Standards Tests Technical Report, Spring 2010 Administration. You can see with the naked eye from these tables that the bulk of the responses are in the center. If you want the graphs, I can certainly email them to you, and you can do spot checks against the published tabular data.
    Again, once upon a time I trusted the education officials to do the right thing. But since the reports I got for my children’s CST tests did not make sense, I decided to verify the claims. I treated the problem as an observation of a natural process: collect the data, graph it, and see what model applies. Once I graphed the scores, it was remarkably easy: they were all bell curves. To top it off, ETS and SBoE staff made it even easier: all CSTs have the same point as the proficient cut-off point: the average of the distribution. From there, it is very easy to determine the API of an elementary school that fits the model, 755. Which presents a challenge to those who call themselves journalists: where did the 800 level come from?
    And, yes, there will always be schools that get most of their students to be advanced. They are either true outliers (in the statistical sense, without which you will never get the spread necessary for the bell curve to be met), are cheating (Atlanta got caught because they went overboard, and in California nobody is looking so it is just a matter of time), or their socioeconomic condition puts them in the outlier population all the time.
    Let me be clear on this: all tests will have a natural tendency to develop a bell curve if the population sample is big enough. The important thing is to decide where to put the cut-off points. The state has erred in setting points that define “basic” as “failing”. And it also has erred in maintaining the “realiability” of the test so that not school will score above proficient, again, because it set as a zero sum game.
    The real problem is that the CST design, implementation, and results did not get a sufficiently public airing because, as unbelievable as that sounds, even faculty members at a prestigious California university have never checked on this either. But it is not surprising since work of this type is not funded and people would like to think that experts know what they are doing, right?
    Alas, they are human beings, and carry the same bias and problems the rest of us mere mortals do.
    The rest of my points have been covered by others. I may give them a whirl later but right now I have to cut this short.

    Report this comment for abusive language, hate speech and profanity

  34. John,
    el is correct that items that are too easy or too hard have little discriminating power and tend not to show up on the test. In theory, it would tend to make the test objectively somewhat “harder” over time– as compared to its early versions — if the population achievement increased a *lot* overall.  However, it is important to note the “in theory” qualifier. We are far far away from that, and the process is exactly as you describe, that items are retired over time due to age and replaced with items of similar difficulty. If at all, there is some evidence that the cut scores of California has been silently(!) diluted by the California Department of Education over time, presumably without the knowledge or blessing of the State Board. In other word, in practice, scoring on the test became somewhat easier over time. Just the opposite of what the “in theory” predicts.

    Report this comment for abusive language, hate speech and profanity

  35. I have to ask if anyone reading this can admit that Manuel is making a credible argument to support his claims: has anyone else here ever looked at those technical reports he reviewed and reached a different conclusion?
    Given all the hoopla over the recent damning fed report on SIG and the multiple formulae we have defining “low performing” and the potential impact of the looming parent empowerment regulations, I think it is past time we take another look at what constitutes a failing school. Thank you for all your hard work, Manuel.

    Report this comment for abusive language, hate speech and profanity

  36. I have looked at some of them in the past, and I re-looked at the recent report just now. To tell the truth, I am not sure what Manuel is trying to say, as he is somewhat incoherent. However, without dwelling on every particular, one should simply look at appendix 10.A (p. 603 & up) and see that the achievement of cohorts over time (a) indeed approximates a bell curve, as is expected(!) and (b) the center of the bell curve generally shifts slowly to the right over time, which is consistent with increasing achievement. In other words, his claim that the proficient cut-score is defined exactly at the peak of the bell curve and doesn’t change is simply incorrect — 8-10 years ago the cut scores were beyond the peak of the distribution, and now most of the peaks are already beyond the proficient cut scores (about 350).
    As to Manuel’s other arguments — sorry, but I can’t make much sense of them. He seems to be looking for conspiracy while what he sees is simply an artifact of the design. These are scaled scores and it is only expected that the scaled cut scores will stay about constant. Perhaps reading something about the level-setting process (California used the Bookmark procedure: Lewis, Mitzel, & Green, 1996) and about the equating process, earlier in the report, might help.

    Report this comment for abusive language, hate speech and profanity

  37. Ze’ev, thank you for looking into it.
    However, I am puzzled by the label “incoherent” and other remarks. Let me try it again, but this time in more detail (or is it louder and simpler?).
    For those not trained in the fine art of plotting data and unable to notice fine detail in their mental picture of a data table (I’m good, but not that good), I transferred the tables from the 2002 to the 2010 administrations into Excel (I wanted to use a widely available data analysis package) and plotted the number of scores at a given score vs the score. The data bins are a bit rough (15 points wide) over the score range (150 to 600), but they produced understandable plots, especially when “bar graphs” are used. The plots, traditionally called histograms, do not exactly fit a Gaussian distribution (aka the Bell Curve), but they “are good enough for government work.” The next task is to find a best fit for these plots. I exploited the fact that Excel has a function to do just that (NORMDIST and its associates). I was aided by the ranges listed in p. 18 (p. 34, actual) of the 2010 report, which, incidentally, have not changed since the first official administration of the CSTs in 2002. Conveniently, the proficient cut-off point is the same for all tests: 350. That’s darn close to the average of all the ELA CSTs distributions, so it is good enough for me. Next for the spread: the best value for the standard deviation turned out to be 60. The model, much to my surprise, also fitted the math CSTs for elementary grades, so I was a happy man. From there, I can calculate what the expected number of test takers is for each achievement band and compare it to the actual result. Or I can calculate what the expected score should be for a population that fits the model. Or I can draw conclusions about the state of math teaching and A-G completion. Etc.
    It is fair to ask: is there any large variation on the average? Not that I can tell. Over the years, the ELA CSTs histograms for grades 2-5 are almost rock solid with small variations (one caveat: the infamous 3rd grade drop is just a slight shift to the left of the average, but all in all it still a Gaussian). ELA CSTs for secondary schools also do not look so bad when considering that most students have figured out that the tests are meaningless to their grades and, consequently, many of them just don’t care how well they do. Still, English is the everyday language and the test makers have managed to design a test that still delivers a good distribution.
    This doesn’t apply to math in the secondary schools because math, in contrast to English, is not an everyday language, especially when we are talking about algebra (How many of you gentle readers can do ratios in their head?). The CSTs for math up to the 7th grade still fit the model but once students take the Algebra I CST the resulting scores do not fit it. The distribution heavily skews to the left and is no longer symmetric about its average. These effects are even worse for non-8th graders, that is, those most likely who are repeating the grade. And this is not a fluke, as the distribution has remained the same for all years where the data has been disaggregated by grade (2007-2010).
    So what to do with this information? Well, we are told by educational bureaucrats that the tests are curriculum-referenced tests and as such compare the student against the curriculum, not against other students as is done in a standardized test. But this is precisely what California has done. Plotting the data shows that the distribution of scaled scores, which is what parents get in the mail in August and forms the basis of the API, has not budged at all. I would expect that if the aim is to achieve 100% proficiency, the peak of the histograms would have moved far to the right by now because teachers (yes, those who actually deal with students) would have adjusted their teaching to conform to the “released questions” styles, made sure they covered every single item in the curriculum, done drill-and-kill to no end, etc. But, no, the shape of the histograms seem eternal. And the teachers I’ve shown these graphs to are not happy. Not happy at all. Worse, this means that the 100% proficient goal will never be met (unless California pulls an Atlanta, but I digress). But what can anyone do when seemingly everyone dismisses these findings as “incoherent” or “uninformed”?
    These distributions can only happen if the test was designed to conform to a standard distribution. The state admits that the questions chosen for the test all have a “probability” of being answered after being tried out in a chosen population (the details are hairy so I won’t go into that). That’s the hallmark of a normalized test. So the CSTs are not what Mr. Fensterwald tells as are supposed to be. The data does not match what he was told, pure and simple. And this, folks, is not, IMNSHO, incoherent.
    (Point of fact: the 2002 and 2003 administrations of the CTS were the “shakedown cruises” of the tests. Thereafter, whoever (ETS? CDE?) defined what was in the test got into a pretty good groove and have, in my opinion, batted nearly 1.000. Indeed, the averages in those early years were lower than 350, but unless one looks at the histogram, one cannot tell how close they came to fitting a Gaussian. I continue to be impressed in how well this scheme has work and realize that is exactly why “they” don’t want to go to a test that has no Gaussian shape but rather the mythical “everybody achieves” “J” distribution: “they” would have to develop the math for this new approach. Maybe Ze’ev knows of a better method since that is his bailiwick.)
    Thank you for reading.

    Report this comment for abusive language, hate speech and profanity

  38. Manuel,
    Your arguments seem to be split into two groups. One has to do with the design of the test, and the other with drawing conclusions based on your perceptions of the first. Let me respond to them in that order.
    Assuming that various student abilities are normally distributed — an assumption most believe is true — any well-made test will exhibit that underlying pattern. The difference between a norm-referenced test and a criterion-referenced test is simply how “one draws the line.” California drew those lines during its level-setting in early 2000s and those lines are supposed to stay fixed over time. Norm-referenced test redraws those lines every so often. Consequently, if everything works as it should, the distribution shape does not change but, if learning occurs, that distribution shifts rightward over time relative to those lines (i.e., cut-scores). As STAR reports the size of the group in each “bin,” the higher bins tend to get more populated over time if all goes well. And that is precisely what one can see in the data.
    When it comes to variation, the standard deviation (SD) measures its empirical value. If the test-forms over the years are held reasonably consistent, one would expect the SD to remain stable. If, however, something changed — e.g., student ability stopped being normally-distributed, or teacher/school effectiveness became more uniform, or major cheating took place, SD might change significantly. The fact that ELA SD has been about constant is, in fact, expected. If we were to see SD shrinking, we would interpret it as either schools & teachers becoming more uniform and consistent, or as a sign of massive cheating. If SD were to grow, we would interpret is as an indication that the variability in school effectiveness got bigger and some schools are doing much better in managing the effectiveness of instruction, while other are left behind. Interestingly, we see the SD pretty steady in most ELA results (although growing in some grades, e.g., 2,4,8) but in math the SD increased in almost every K-8 grade. Seems like the variability of  the quality of math instruction has seriously increased in our school. Not very surprising to those who watch this area.
    The behavior above is predicated on the normal distribution of student achievement, and the way test is constructed. When you pull the cohort apart by achievement — as we do in math starting in grade 8 — this assumption is no longer valid, and the results look much more confusing and literally not “normal.” That is what accounts for the “strange” results you see there. Finally, the scaled cut scores are *expected* to stay more or less constant — that is why they are “scaled.” In early years they were not so steady and still fluttered a couple of points from year to year, but this seems to have settled finally.
    Now I come to your second type of argument, your opinions. For example, “ELA CSTs for secondary schools also do not look so bad when considering that most students have figured out that the tests are meaningless to their grades and, consequently, many of them just don’t care how well they do.”  You have no basis for this statement and, in fact, the data does not support you. If it were a matter of HS students “figuring out” that results don’t matter, we would see a large increase in SD. Yet we do not. “Still, English is the everyday language and the test makers have managed to design a test that still delivers a good distribution.” — what is the meaning of that? Based on what? And why the conspiratorial tone?
    Further we see things like “Plotting the data shows that the distribution of scaled scores … has not budged at all. I would expect that if the aim is to achieve 100% proficiency, the peak of the histograms would have moved far to the right by now.” Wrong. First, the peaks did move quite significantly to the right over the years, as you repeatedly seem to ignore. Second, what you describe is the end goal, which we have not reached yet.  Is anyone surprised by it? Hiding it?
    I am not going into the details of every clause of the second half of your post, but essentially all of them draw incorrect or unsubstantiated conclusions. Perhaps I would simply suggest you read a bit about statistics and psychometric before you throw unsupported, and often contrary to empirical date, accusations. Until then, I suggest to keep using IMHO rather than IMNSHO.

    Report this comment for abusive language, hate speech and profanity

  39. John:
    I believe we have gone over much of this before.
    There is NO tenure for k-12 teachers in CA. There is “permanent status.” This entitles a teacher to due process rights under law and negotiated agreements. The law is subject to change and school management, or the public, has every right to lobby to get it changed.  The law is not that protective of teacher rights.
    Two year permanent status was a political trade off made when probationary teachers lost due process rights and now can be “non-reelected” without cause. That was the silver bullet reform du jour of the time. Prob teachers are typically  evaluated two times a year for two years. Management has the right to evaluate them every day if they want. How many times does it take to determine if they are to gain perm status?
    Here we have all the chatter about “underprepared teachers” and you propose eliminating the incentive to garner more college credit. The iconic Finnish system has all teachers with MAs. Why is it a bad idea for our teachers to do the same? Teachers typically have to get approval of units taken by management to insure they are not meaningless. Where is the breakdown?
    Step and column is in place because every other method has been a failure. It is fair and objective. See link above to latest abortive effort in New York to create performance based system.
    There has been outrage recently that schools in highly stressed communities were overburdened with rookie teachers. Now, improvement can only come by making it easier to get rid of senior teachers and keep newer teachers. The same reason applies it as does for step and column. With no fair criteria in place teachers were subject to arbitrary and capricious dismissal, nepotism, and routine dismissal when they could be replaced with cheaper teachers.
    And it is not “strictly” seniority based. Districts can and do “skip” over teachers who have been identified to be credentialed in high needs areas.
    I’ll let you argue the test score question with the National Research Council. The NRC (as well as the AERA and Amer Psych Assoc) says there is no research base to use student test scores for teacher evaluation. The NRC has (along with Ravitch) also taken the stand that test driven education has not had a positive effect on achievement and that it is having a narrowing and negative effect on curriculum. The only true supporters are the two totally unqualified reporters at the LA Times. (I don’t think this brand of yellow journalism will save the LAT.)
    Then there is the “law” that putting high stakes on social science data results in corruption. See Atlanta.
    I don’t know what “scheduling” you’re talking about. Schools have “traditional” and “block.” What other “no-cost” silver bullet did you have in mind?

    Report this comment for abusive language, hate speech and profanity

  40. Our union contract at the San Jose Mercury News had a pay-scale system based on seniority too, and only 13 weeks’ probation. When I say “our” contract, I mean mine and John Fensterwald’s, though as far as I know he started there after I left.

    Report this comment for abusive language, hate speech and profanity

  41. The California Standards for the Teaching Profession are used and are effective.  Student test scores were never designed to evaluate teachers.  As we are obsessing about “evaluating teachers,”  in my district, the Academic Decathlon program is being dismantled and electives are disappearing. 

    I have an idea. If we are going to make kids take chemistry, then offer a plastics class.  If we are going to make kids take math, then offier accounting because that is one of the top fields of the future.

    Report this comment for abusive language, hate speech and profanity

  42. El,

    I agree that we have a better educated student population today that we did in our days in school.  I also took Algebra in 8th grade and there were few of us taking Calculus in 11th grade.  We see many more students successfully passing AP Calculus today than there were in the 80s.

    Today’s leaders are forced to confront percentages of proficiency – for everyone – that didn’t have to be acknowledged 30 years ago.  I’m certain we have more young people finishing high school better prepared for higher education.  A challenge is we also have more young people finishing school less prepared for higher education.  The population has increased more quickly than society can handle, but the number of four-year public university spots has not matched population growth.

    Thanks for the dose of reality about demographics and how much actually is better today.  Perhaps it’s not as comfy today for those who attended relatively sheltered suburban schools.   But overall, California has made lots of improvements.  And, we need to achieve even more rapidly.

    - Chris S.

    Report this comment for abusive language, hate speech and profanity

  43. John writes: “Teachers who year after year simply cannot move their students toward proficiency and mastery — whether from below basic to basic or basic to proficient or proficient to advanced (if they have a class of high achievers) are not being effective. There needs to be objective measures to guide improvement or, eventually, dismiss the teacher.”

    Huzzah John!  Simply huzzah!

    And teachers who year after year DO move their students towards proficiency and mastery should be commended.  Society needs those professionals to continue their success.

    Report this comment for abusive language, hate speech and profanity

  44. Ze’ev,
    thank you for replying and confirming that the CSTs’ achievement levels are set on a bell curve that has remained fixed over the years. Indeed, the lines are fixed. But the test can be “corrected” over the years so the outcome fits the lines. It happens when the “curve” is used to grade large classes at universities. Why not in this case?
    In the graphs I made, I see no evidence that there is much of a right-ward shift in the distributions, but maybe my standards are more stringent than yours. Now, you claim “that is precisely what one can see in the data.” Wonderful. With that as a hopeful sign I dove back into the reports. Bingo! Each of them has a table on the means and standard divisions. Let’s make it easy and allow me to pick the 5th graders’ ELA as any CST should fit your paradigm. I read from the 2004, 2007, and 2010 reports:
    Year  Mean SD
    2002   328   46
    2003   332   47
    2004   338   54
    2005   340   56
    2006   342   57
    2007   343   54
    2008   348   52
    2009   356   57
    2010   359   54
    A shift in the mean of 20 points over 9 years when the SD is about 50 is what you call a “dramatic shift?” My standards are clearly more stringent than yours (and this is empirical data straight from the reports). Especially when the proficient point is set at 350, which translate into a very low percentage of “proficient and above” for the early years of the program.
    In regards to the SD, I doubt that massive cheating or uniform teacher effectiveness would increase the SD. It should be the opposite since the spread in scores would be less as all of them would be aiming at a high score. Conversely, if the teacher effectiveness were not uniform across the state, then I agree that it would increase. The plots I made of the math CSTs scores do not show much variability between administrations at a given grade. To me, that’s evidence that efforts across the state by teachers to conform to the CSTs are not producing much “progress”.
    The taking apart of the cohort I believe is correct: all that is being done is following the 8th graders as they progress from 7th grade math to Algebra I, and so on. As such, they are still with their peers and 8th graders are being compared only to 8th graders that the schools have deemed as “ready” to take Algebra I. That’s “following the cohort,” isn’t it? And I did not call the results strange. I said they went to hell. I just left out the description: the peak of their distribution moved leftward while the majority of test takers clustered in the “basic” classification. That’s not a very good “normalized” distribution. Either the test was poorly conceived or the students have learned poorly (why that is is left to the readers’ imagination and political persuasion).
    As for my opinions on why students do so poorly, there is no published peer-reviewed research backing my assertion but my experience at my kids’ high-school indicates that a non-self-motivated population (i.e., not a gung-ho academic magnet) will not really care about the CSTs because they know it doesn’t matter one way or another to their class mark. Many of them freely admitted it at a meeting organized to inform them that if they did not increase their math scores, their art-oriented magnet program could be closed. That got their attention. In regards to the English CST, if you are able to speak, read, and write Standard American English, you will do average in the test and there is always a probability that you’ll do advanced as well as far below basic. Not so in math once it is Algebra and above. (An examination of a particular high-achieving 6-12 magnet in a certain large California district proves it but I won’t go into that here.) That’s my experience with my own children and others. Of course you can dismiss it, but that is not going to change the reality at my ground level which is not, I assure you, Palo Alto.
    As the table above shows, admittedly for one single grade and test, I don’t see the “peaks … move quite significantly to the right over the years.” What is “significant” to you? 0.5 SD? 1.0 SD? If it is anything above 0.5 SD, I ain’t ignoring ‘em ’cause they ain’t there.
    As for not yet reaching the 100% proficient, we are three years away from the NCLB deadline and we’ve been in this CST rat race since 2002. Yet, slicing the data differently, the percent of 5th graders (yes, I like this group because it is the transition between elementary and middle school) in the ELA above proficient has been, according to the STAR web site (the data is also in the reports, but let’s use the web site), as follows:
    2002: 31%
    2003: 36%
    2004: 40%
    2005: 43%
    2006: 43%
    2007: 44%
    2008: 48%
    2009: 54%
    2010: 58%
    Yes, there has been a small shift right-ward, but it isn’t dramatic at all (and it doesn’t even translate into a significant increase in the number of students in the next-to-last bin: 712, 900, 311, 355, 1436, 691, for 2005, 2006, 2007, 2008, 2009, 2010, respectively, and keep in mind that the size of the cohort ranges from 485,000 to 431,000 students). Is this progress? And at this pace, we’ll need, what? Ten more years to reach 100% proficiency? This “progress” is not what I expected. And neither do the feds. What has the state done about this “progress” other than release sample questions with a disclaimer that reads “Test scores cannot be projected based on performance on released test questions”? And you wonder why I am suspicious? How would you feel if you were stonewalled when asking for the data from CDE and you can’t get it from your local district because “This information is the property of the ***** District and is intended solely for internal use by authorized employees with a reasonable need to utilize it in order to perform those duties assigned by their supervisors. This information is not to be reproduced in any form or viewed by non-authorized personnel nor distributed, including via any electronic means, including email, to any authorized or unauthorized person or organization.” This is supposed to be public information, isn’t it? So why all the legalese and threats? Given this state of affairs, who wouldn’t be conspiratorial and accusational?
    And I know enough about statistics to know when I’ve been sold a bill of goods, ’cause, darn it, that Ph.D. in Experimental Physics does come in handy sometimes, so I think I’m entitled to the IMNSHO. As for psychometrics, isn’t that another word for “someone who does testing?” Or are you talking about the folks that figure out ways to pass a test without actually knowing its content? I’ve heard both definitions from practitioners in that very same field and I think I’ll go with their take on it until someone turns Asimov’s musings into reality.
    Anyhow, I think I’ve abundantly made my point. Thank you, again, for responding and pointing me in other directions. This has truly been a very good educational experience. (And thanks also to Mr. Fensterwald for making this space available.)

    Report this comment for abusive language, hate speech and profanity

  45. Hi – I am no stats expert, but the difficulty with looking at CA scores at grades 2-11 from 1998 – present is that the test altered to Criterion referenced mid-way through, thanks to NCLB. The criterion referenced tests are just that – there is no norm referencing, and worse – no attempt to develop external validity. If you talk to staffers at the CDE, they will tell you that there are skews at 3rd grade and (I think) 9th grade that they wanted to fix – especially the 3rd grade ELA), but weren’t authorized to. 
    Many districts – I worked in one that did this – use the State test results to place students in program levels. This is done, even though neither the State (formally) nor districts – have developed correlations between scaled scores on the CA tests, and AP/SAT/ACT results. Now that CALPADS is almost gone, any opportunity to really run post-hoc analyses of the validity of tests is gone. When using the SAT9 testing I realized that students at around the 80th percentile mark were those heading to UCs! 
    I’ve been told that the score for which one should aim for is now Scaled at 400 or above. But, since API is so generous to schools there is no reason for schools to aim there for the individual kids. The best assessment I found over the years in the US (I used to teach in UK) for reading comp, and vocab, was the 1920s Gates-Maginitie! Cheap, quick to administer, easy to explain to students and parents, and has excellent psychometric standing! 
    I used to blame CDE staffers until I talked to them and realized how constrained they are by the Legislature and the SBE. Until the politicians get out of the picture, along with the other powerbrokers – CTA (?), textbook publishers, – the children are really the pawns. 

    Report this comment for abusive language, hate speech and profanity

  46. Very interesting back and forth between Manuel and Ze’ve on the underlying nature of the stats behind our STAR Standards Tests (CSTs). I’ve been out-of-state this week, but regular readers of TopED will recognize I frequently weigh in on technical issues dealing with CA’s statewide assessment system.
    I guess I need to contrast the Norm-Referenced Tests (NRTs) that were the staple for national El-Hi testing in the 60′s through the 90′s to the so-called Standards-Based Tests that have become the staple for all statewide testing programs in the 00′s. The changeover was initiated in the very late 80′s starting with the GeoBush the First education summitt at Charlottesville Va in late 89 and it’s call to reframe how K-12 tests were constructed and used. That summitt came shortly after a physician activist (J.J.Cannell) accused the testing industry producing NRTs to be in cohoots with school administrators to mislead the American public via a testing system that allowed all states to claim they were “above average.” That controversy led to Cannell being published in places like TIME magazine and interviewed by Mike Wallace on 60 minutes with his allegations [which became known as the Lake Wobegon effect, after Garrison Keiler's book claiming all kids from his community were "above average], and was the seed for a call for change from a normative system to a standards-based non-normative system. It took 10 full years to work through that call for a major change essentially in the underlying assumptions undergriding K-12 testing, including a 1997 re-authorization of ESEA that “urged” states to adopt standards-based testing systems, presaging the mandatory change to the new type of tests by NCLB in 2002.
    The underlying statistical rationale for the NRTs in the 60′s through the 90′s was to measure “current” achievement levels for kids as comparisons to achievement levels for all other kids at the same grade level — i.e., the fundamental score from NRTs was a “percentile” score that provided a student comparison to other kids [i.e., norm-referenced]. Norms did not absolutely have to be based on Bell Curve distributions as suggested by Manuel, but in fact they were simply because student achievement tended to be exhibited by Bell Curve or normal distribuitons (as pointed out by Ze’ve). But the biggest difference between NRTs and the new-fangled Standards Based Tests that were desired was that NRTs attempted to measure “current” achievement while Standards-Based tests were to be designed to measure “aspired” achievement, or what “should be achieved” for the US to be competitive with other countries which were beginning to outpace the US education system per various international comparisons. The change to attempting to measure what “should be achieved” was a major major change, and all the features of the NRT system that was the self-contained zero sum system that Manuel described had to be jettisoned to get to a Standards-Based system that reflected what we wanted kids to know and be able to do to be competitive with other countries.
    The result of this philosophical as well as statistical change led to the concept of “proficiency” being a goal for what we wanted kids to achieve, and for that goal to be fixed against a description of content rather than a description based on scores for all students (i.e. the “norm”). The concept of “proficient” is not intended to be an average like the NRT 50th percentile or grade level score was — rather it was intended to be a fixed amount of knowlege or achievement that would remain fixed over time [or at least until we collectively decided to raise or lower that standard -- that collective decision is essentially a political decision in our society, in the case of Caliifornia it is rendered by our State Board of Education based on recommendations from the SPI and California Department of Education with staff work done by the testing vendor which is now ETS]. To the best of the abilities of all involved in the evolution of the STAR CSTs in California, that is what the STAR system now reflects and has reflected since it was fully installed in 2003. [Previous to 2003, CA's STAR tests were based on the norm-referenced Stanford Achievement Tests published by Harcourt, which was used as a platform to get to the Standards-Based CSTs in 2003.]
    Given this history, most of Manuel’s criticisms are based on normative concepts that undergridded the jetisoned NRTs. The fact that the CSTs exhibit more or less normative distributions is an artifact of underlying achievement levels across all kids. It is not connected to the cut scores that determine proficiency or the other scores generated by the CSTs (far below basic, below basic, basic, advanced). The CST scores are designed to be a fixed degree of knowledge, as best we can measure. Now, educational achievement measurement ain’t perfect, and there are in perfections in the system. Sue Moore pointed out the grade 3 E/LA anonaly that we’ve know about since 2003 or 2004 and for which there was a contentions “task force” meeting in late 2005 to try to resolve — my take on that meeting was that (1) folks couldn’t entirely agree on the content or statistical cause for the anonaly, and (2) neither CDE staff nor SBE folks had the will to issue a recall or revision of CST scores for that grade level and content area that would have been necessary to correct the anonaly. So, we’ve lived with the grade 3 E/LA glitch in the STAR system for the last 8 years even though folks have known about it from early on. But, to put a context on this glitch, it primarily affects only one grade level and one content area out of some 25 CSTs in the STAR system of tests — as a K-12 testing system designer and developer for more than 40 years, I am fond of saying nobody is perfect and virtually all systems of tests have glitches in one or more subtests and/or grade levels. I know the grade 9 Stanford Achievement Math tests had glitches that were never resolved, and I can tell you the 1993 California Achievement Tests published under my direction had selected subtest glitches — such is the nature of an imperfect world and designers and developers do not always get what they hoped to get when they publish a K-12 assessment system.
    The facts are that most of Manuel’s statistical observations are an attempt to map STAR CST data backwards onto the NRT system of test development, and they are not reflective of how the STAR CSTs were actually developed. That the STAR CSTs do show some similar statistical characteristics does not logically translate to claims that the CSTs have the same properties as the NRTs which led to the effective demise of NRTs, with such demise quite consciously effected by the desire to measure a fixed concept of standards-based (or content based) achievement that kids should have rather than a comparative or normative based concept of achievement that kids currently do have.
    As a last word, let me respond to Manuel’s comments re what a describes a psychometrician. Us cats are definitely not folks who give tests — there is a career as a psychometrist that describes that function, a very noble occupation. Using the word derivation, a psychometrician is a “mind measurer.” Fair enough, I guess, though to date we have had to resort to indirect measurement of what is going on in the minds of kids that yields that concept called academic achievement. And that indirect measurement is the tests that psychometricians design and construct, with all their limitations and inperfections. And those tests generate a whole bunch of criticism, especially the idea that “evil” multiple-choice test items are the bane of existance for anyone affected in one way or the other by the results of the mind measuring that psychometricians attempt to do. The existance of the statistical psychometric occupation is probably threated by the emerging field of neuropsychology with all its brain scanning reasearch — it ain’t that far a stretch to think that academic achievement might be measured by periodic brain scanning MRI sessions rather than paper-and-pencil (or even computer-administerer or computer-adaptive) tests that reflect the current state-of-the-art for educational measurement. But, that notion has ethical concerns as well as scientific hurdles that need to be addressed. [Full disclosure: I have a daughter with a Ph.D. in Neuropsychology, gratefully applying her research skills to the area of human emotion rather than educational measurement.] However, the definition of a psychometrician that I prefer is one that says a psychometrician is a folk who is good with numbers . . . . but doesn’t have the personality to be an accountant. Given all the criticism of educational measurement in the context of use for high stakes accountability purposes, one needs all the personality one can muster.
    Doug McRae, Retired Test Publisher, Monterey

    Report this comment for abusive language, hate speech and profanity

  47. Couple of quickies, as the discussion with Manuel seems to meander. I don’t plan to respond to him beyond this.
    -  Manuel: “achievement levels are set on a bell curve.” Wrong. Achievement levels are not set on a bell curve. Achievement levels had been set by a level-setting process (*Bookmark”) that is criterion referenced. The *underlying student achievement* of the whole cohort has a bell curve type of distribution.
    - Manuel: “dramatic shift” and other “interpretive” arguments. I never used this term. Please don’t put words in my mouth and then attack them. There has been a significant shift in achievement over the last decade, both in terms of statistical significance, and in terms of its educational importance. Whether they fit your sense of drama is up to you. Please be careful with your reading, as you seem to misread and/or misinterpret much of what I wrote. As another example, you completely misinterpret my words about SD — I wrote that cheating would narrow it, you claim I did the opposite. And so on, and so forth, in many other places.
    @Sue Moore: We (Calif.) did not alter the criteria (i.e., cut-scores) because of NCLB. We simply chose that our STAR category of “proficient” would be the one we will use as “proficient” for the purpose of NCLB. Further, we cannot compare results from 1998 because we did not have scaled scores in 1998. It took some time to develop and validate them, and we have had them since 2001-2004 (depending on the subject). They are comparable since then, and very useful at that. Incidentally, we will lose all this longitudinal track record with the future advent of common core testing.
    As to the 3rd grade ELA cut-scores, you are correct that they seem to have been set a bit too high in the 2000 level-setting process. This was identified a very long time ago to CDE and instead of simply admitting it and coming forward to the State Board with a request to re-run the level-setting for that grade — a pretty simple exercise — it took them 3-4 years even to admit it. When CDE  finally did acknowledge the problem of grade 3, the internal bickering between CDE and SBE and the loss of most of competent people from both CDE and SBE leadership doomed any effort to fix it.  In my opinion you are correct to view this persisting 3rd grade “kink” as a symptom of California’s dysfunctional and incompetent educational leadership over much of the last decade, but I would disagree with you that the blame lies chiefly with the state board. In my book the blame is more or less equally divided between incompetent CDE leadership and weak SBE members.

    Report this comment for abusive language, hate speech and profanity

    • We’ve gone into much more depth on testing that I had expected. It’s been a useful crash course. Thanks, Doug and Ze’ev, for sharing your knowledge. Manuel, I appreciate your delving into an area beyond my expertise — whether or not you accept Doug’s and Ze’ev’s points.
      On other issues:
      Caroline: While I was at the Mercury News, we went through a series of wrenching layoffs. Seniority was given a 25 percent weight in deciding who would be let go, which I considered to be appropriate, since we had outstanding young reporters and editors. Readers certainly appreciated that they continued in their jobs, as did I. Parents should have the same expectations as readers; the loss of outstanding teachers based totally on seniority is a tragedy.
      We had a step pay scale at the Merc, which was generous at the time compared with most newspapers, and I appreciated the hard-nose negotiations by my union leaders and the benefits they secured for us. But I was hired well above scale, which made my move from the East Coast possible. During the salad at the Merc, merit raises far outpaced union scale and were a reason the Merc was able to attract and keep some of its outstanding journalists.
      Yes, Gary, we have been over some of these points, and I will continue to raise them. Call it tenure or permanent status, teachers beyond two years get substantial due-process rights, so, as I said, we should be confident they are good for the long haul because it will be much more difficult to fire them from then on. We obviously disagree over your phrase “not that protective” of teachers’ right. I would say “overly protective.”
      I certainly don’t equate teacher preparation with an accumulation of more credits toward a master’s, particularly when the master’s has no relation to the teacher’s subject. There is simply no correlation between post-BA credits and teacher performance, but we continue to pay teachers on that basis. An objective basis, yes. A wise basis, no. New York’s decision to discontinue a flawed bonus pay system, awarded to all teachers in a school based on the school’s test results, does not discredit the need for both a more effective evaluation system that looks at a full range of teachers’ leadership and skills, and a pay system that recognizes those accomplishments. I’m certainly not advocating that value-added test scores comprise the biggest part of an evaluation. But many of the same studies that have found flaws in the value-added method acknowledge its usefulness as a component in assessing overall performance. The much maligned LAT reporters made the same argument.

      Report this comment for abusive language, hate speech and profanity

  48. And as I’ve said, John, judging teachers based on their students’ test scores is the exact equivalent of judging newspaper reporters/columnists based on the readership for their work. Reporters covering planning commissions, pension reform and LAFCO (not to mention education) — out. Reporters covering celebrity gossip get lifetime jobs.
    Though of course by the same standards we apply to teachers, circulation and profits should also be part of the evaluation system.
    And also, an issue with merit raises is that they are based very heavily on the employee’s ability to sell him/herself, not so much on his/her actual quality as a journalist. The meek do not inherit the Earth; the flashy and supremely confident do. I could give many examples and I know you could too. (One of our star reporters in my day was so flamboyantly inaccurate that our semi-joking instruction when I was on the Metro copy desk was: If she says San Jose is in California, double-check it.)
    (Full disclosure that I got overscale as an enticement to take on a job that was not considered desirable, in a time of crisis.)

    Report this comment for abusive language, hate speech and profanity

  49. …and another analogous observation is that when I first arrived at the Mercury News in 1982 (as a bright-eyed beginner myself), the prevailing management attitude was one of contempt for the veterans on the staff. They were viewed as relics from a time when low local standards prevailed, outdated anachronisms now that the Mercury News had gone world-class. Filling the staff with rising stars from elsewhere was considered smart management.
    But after a few years, that attitude was reversed — partly because the former newcomers started to become veterans, but also because it was impossible to do local coverage competently when those who knew the local landscape were treated with disdain and excluded from decisionmaking — something that started to become embarrassingly apparent.
    The other issue with at-will employment is that, obviously, it discourages risk-taking. That’s an issue of one kind in the newspaper business and of another kind in public education.
    In any case, we know that a teaching force with at-will employment does not make for better schools and better-educated students — in the large-scale real life examples we have in front of us, the opposite is true. The right-to-work states consistently show the lowest academic achievement, and the strongest union states consistently have the highest-achieving schools.

    Report this comment for abusive language, hate speech and profanity

  50. I like back-and-forth discussion. I’m not trying to get the final word*, but just correcting a misunderstanding.
    I’m not saying that union power necessarily CAUSES high achievement — in fact, I constantly complain that the press en masse fails to grasp the principle that correlation doesn’t equal causation, so it would be a lapse for me to fail to grasp that principle myself.
    Union power correlates with high achievement; the absence of unions correlates with low achievement.
    That doesn’t mean that strong unions cause high achievement and that lack of job security and employee rights for teachers causes low achievement.

    But it decisively, resoundingly, conclusively (add synonyms ad infinitum) refutes the ongoing refrain from the ed reformers that eliminating teachers’ collective bargaining rights and job security would lead to higher achievement.
    *John, if you want lively discussion in your comments section — as I infer from your recent tweet that you do — you might rethink trying to embarrass commenters, as you just did me. That tends to discourage discussion.

    Report this comment for abusive language, hate speech and profanity


  1. Connecting the Dots: Top news stories for Wednesday, July 20 | California News

"Darn, I wish I had read that over again before I hit send.” Don’t let this be your lament. To promote a civil dialogue, please be considerate, respectful and mindful of your tone. We encourage you to use your real name, but if you must use a nom de plume, stick with it. Anonymous postings will be removed.

2010 elections(16)
2012 election(16)
A to G Curriculum(27)
Achievement Gap(38)
Adequacy suit(19)
Adult education(1)
Advocacy organizations(20)
Blog info(5)
Career academies(20)
Character education(2)
Common Core standards(71)
Community Colleges(66)
Did You Know(16)
Disabilities education(3)
Dropout prevention(11)
© Thoughts on Public Education 2015 | Home | Terms of Use | Site Map | Contact Us