Student scores in evaluations

In a decision with statewide implications, a Superior Court judge ruled that Los Angeles Unified must include measures of student progress, including scores on state standardized tests, when evaluating teachers and principals.

But Los Angeles County Superior Court Judge James Chalfant will leave it to the district, in negotiations with its teachers union and administrators union, to determine what other measures of student performance might also be included, how much weight to give them in an evaluation, and how exactly test scores and other measures should be used.

Chalfant’s decision would appear to strengthen Superintendent John Deasy’s push to move forward with a complex value-added system of measuring individual students’ progress on state standardized tests, called Academic Growth over Time. Deasy wants to introduce AGT on a test basis in a pilot evaluation program next year. But the unions remain adamantly opposed to AGT; Chalfant said the use of AGT as a measure of student progress is not his call to make; and today, hours before Chalfant is to meet again with parties in the lawsuit over evaluations, Los Angeles Unified school board member Steve Zimmer will propose barring AGT from staff evaluations. The school board will vote on his motion later this month.

Chalfant released his tentative decision on Monday. (Update: On Tuesday, after a hearing with all parties, he made the ruling final.) But the carefully crafted, 25-page ruling is not likely to change much, if at all, and may become final today, after the school district and unions get a final chance to make their case at a hearing.

The ruling is a victory for Sacramento-based EdVoice, which filed suit on behalf of a half-dozen unnamed Los Angeles Unified students and their parents and guardians. EdVoice’s lawsuit claimed that the Stull Act, the 40-year-old state law laying out procedures for teacher and administrator evaluations, requires school districts to factor in student progress on district standards, however they decide to measure it, as well as scores on the California Standards Tests (CST) in evaluations and that Los Angeles Unified was ignoring the requirement – as do most school districts.

Chalfant agreed and, in his decision, quoted Deasy, who, in testimony, acknowledged the district doesn’t look at how students do academically when evaluating teachers.  On Monday, Deasy praised the tentative decision, and called for  the district, his employer, to move quickly to act on it. “The district has waited far too long to comply with the law,” Deasy said. “This is why LAUSD has created its own evaluation system, and has begun to use it. The system was developed with the input of teachers and administrators.”

Next step: negotiating compliance

Chalfant’s tentative ruling proposed that attorneys for EdVoice and the parents propose a plan for compliance and that they and the district try to negotiate specifics over the next month. Whatever they agree to would still likely have to be negotiated with United Teachers Los Angeles and Associated Administrators Los Angeles.

Bill Lucia, president and CEO of EdVoice, praised Chalfant’s decision. While acknowledging that the emphasis given to student progress could become a sticking point in negotiations between the district and teachers, he said the ruling makes clear “there is no status quo going forward.”

“It won’t be OK to sit on their hands,” Lucia said. “The district must come up with something different that passes the laugh test and makes a sincere effort to honor the statute requiring that evaluations look at whether kids are learning.”

EdVoice took no position on whether the AGT should be the tool by which to measure student performance in Los Angeles. But, Lucia said, the district must consider other measures ­– whether student portfolios or other district tests ­ – in the evaluations of teachers of courses in which CSTs aren’t given, such as first grade, art and seventh grade science.

Signal to other districts

Chalfant’s ruling would apply only to Los Angeles Unified, although other Superior Courts could cite the ruling. Nonetheless, Lucia said that the message to other districts is that “a district cannot omit the progress of kids in job performance of adults.” The goal, he said, “should be a better determination of effectiveness that allows limited resources to be targeted to those teachers needing the most improvement.”

Attorneys for UTLA and the district could not be reached for comment on Monday.

UTLA argued in its brief that a dispute over requirements in the Stull Act belonged before the Public Employee Relations Board, not a court, and that any requirement for the use of test scores or other measures must be negotiated.  But Chalfant wrote that first and foremost, the district must comply with state law, regardless of the contract it reached with the unions.

The position of the district, on behalf of the school board, was confusing. Last year, in defending the  pilot program using AGT, the district  said it had the authority to impose the terms of evaluations without union negotiations. Even though Deasy testified that test scores and student progress weren’t part of staff evaluations, the district fought the EdVoice lawsuit.

In its brief, the district asserted that the use of AGT in the pilot satisfied the law’s requirement to use state standardized test scores – even though they have yet to be applied, with consequences, to any teacher. The district also asserted that it uses results on district and state tests and other student measures to set goals for teacher instruction and measure improvements in the classroom.

But Chalfant ruled that that’s not sufficient. “There must be a nexus between pupil progress and the evaluations. No such nexus currently exists.”

“This does not mean that there must be a box on a form which directly addresses pupil progress,” he wrote. “It does mean that pupil progress must be reflected in some factor on a written teacher evaluation.”

Whether pupil progress – AGT alone or in combination with other student growth measures ­ – counts 20 percent or 30 percent of an evaluation, as Deasy has advocated, must be decided through negotiations, unless the district asserts a right to impose AGT unilaterally.

Villaraigosa’s Stull Act amendment

In 1999, when he was state Assembly speaker, Los Angeles Mayor Antonio Villaraigosa sponsored an amendment updating the Stull Act  to require the use of CST scores in teacher evaluations. Villaraigosa submitted a brief supporting this position.

Chalfant incorporated some of Villaraigosa’s points in explaining the rationale for his decision. In 2009-10, 99.3 percent of teachers evaluated received the highest evaluation rating, with 79 percent meeting all 27 measures of performance. This despite that the district “has one of the lowest high school graduation rates in the State, and an even lower percentage of students are college ready.”

“These failures cannot be laid solely at the feet of the District’s teachers,” Chalfant cointinued. “Students must want to learn in order to do so, and some students can never be motivated to learn. But the District has an obligation to look at any and all means available to help improve the dismal results of its student population. One means of improving student education is to evaluate teachers and administrators based on the overall progress of their students.”

This entry was posted in Evaluations, Tests and tagged , , , , , , , on by .

About John Fensterwald - Educated Guess

John Fensterwald, a journalist at the Silicon Valley Education Foundation, edits and co-writes "Thoughts on Public Education in California" (, one of the leading sources of California education policy reporting and opinion, which he founded in 2009. For 11 years before that, John wrote editorials for the Mercury News in San Jose, with a focus on education. He worked as a reporter, news editor and opinion editor for three newspapers in New Hampshire for two decades before receiving a Knight Fellowship at Stanford University in 1997 and heading West shortly thereafter. His wife is an elementary school teacher and his daughter attends the University California at Davis.

33 thoughts on “Student scores in evaluations

  1. Paul

    “Students must want to learn in order to do so, and some students can never be motivated to learn. But…”
    There you have it.
    Given a choice of professions, what rational person would submit to being evaluated on the basis of test results over which he or she has little direct control? I can present interesting and thorough lessons, but I cannot force students to attend class, to process the information, to complete in-class practice and homework, to study for classroom tests, or to do a good job on state tests that carry no consequences whatsoever for the students themselves.
    Combine this poison pill evaluation ruling with Darrell Steinberg’s, Tom Ammiano’s and San Francisco Unified’s curbs on student discipline and you make success a matter of luck rather than a matter of ability and effort. Do we really expect more people to sign up for the profession under these conditions? The ruling might be legally correct, but all it does is to discourage smart people from being public school teachers.

    Report this comment for abusive language, hate speech and profanity

  2. Sue Moore

    John: Have you heard of any other organizations waiting for this ruling before filing against other districts/the State to enforce Stull Bill, etc.?
    Paul: Some students are not motivated to learn, and as they grow older may well have other issues such as substance abuse, gang pressures, home conflicts, economic pressures, etc., however, one red flag is when students succeed in some classes, then there is a fairly noticeable alteration in their approach to another classroom. Or, when a similar cohort succeed in one teacher’s class, but a similar cohort languish in other classroom.
    I can fully understand, having seen some of the work given to my kids, why they might not be motivated. They are not being asked to learn, they are asked to regurgitate whereas another teacher would have engaged them and prepared interesting approaches to the same material.
    There isn’t a magic answer – but we have to move forwards.

    Report this comment for abusive language, hate speech and profanity

  3. CarolineSF

    We have to move forward, not in the wrong direction.
    A lot of the support for these education reform fads seems to be based on the notion that *we have to do something — anything at all*. I think the rational view is that making huge mistakes is worse than doing nothing.
    If we judge physicians based on how many of their patients get well, we punish those who care for the most seriously ill and high-risk patients. In the nation with the second-highest child poverty rate of developed counties, how is judging teachers on that basis going to benefit anyone except those whose interest is in deprofessionalizing and harming teachers?

    Report this comment for abusive language, hate speech and profanity

  4. Chris Stampolis

    Paul, I think you have almost completely missed the point.
    We don’t pay public school teachers to present lectures for the sake of lectures.  As Khan Academy online shows, there is a low inherent monetary value to the price of a good lecture, though there is a high social benefit.  Mr. Khan has set the consumer price cost to learn quality information at ZERO.
    Rather, we pay teachers to educate students and that means, yes, the teacher has to figure out how to inspire students to learn.  If a teacher does not have the communication and social science skills to understand the complexities of this decade’s student dynamics, that teacher is not a good fit for that particular assignment – or possibly for the profession of teaching in a public school classroom.  Teaching is hard work.  A teacher must exhibit organization, diligence, subject proficiency and a complex, deep understanding of the neighborhood community around the public school to which the teacher is assigned.  So, yes, the residents of California do expect that test scores are one measure against which a District evaluates a teacher’s performance.  We’re not angry at teachers as individual people.  We all know public school teaching is a very difficult job.  At the same time, we expect student proficiency to rise and for career professionals to figure out what is needed and to communicate those needs to students, parents, fellow educators and community members.  In detail.  In specific.  At high levels of multicultural communication.
    Knowing much about a subject or obtaining a teaching certificate is only an early prerequisite to being a good teacher.  Showing up on time to work and providing a well-prepared lecture is only part of being a good teacher in 2012.  Connecting with students, understanding family dynamics, developing community awareness, spending time to analyze individual student performance and fostering student success to heighten proficiency all are part of career commitment in modern California.  Yes, that’s a high bar to reach!  Today’s public school teachers must be classroom teachers and astute social scientists!
    Teaching in this era involves development of professional relationships with parents and creative connections with students.  It requires most teachers to be experts in multicultural communications – and to demonstrate multilingual skills as well.  Responding to Paul’s comment, the system can “force students to attend class” or at least to demand accountability when they do not attend.  The system can provide consequences when “in-class practice and homework” are not complete.  And there are plenty of ways to provide student incentives for state testing.
    In almost every school district in California, new hires to the teaching profession should be required to demonstrate language proficiency in English and at least one other language before receiving a job offer.  There are plenty of bilingual young people in California.  If we require second language skills for all new teachers, we’ll enhance site effectiveness in coming years and require less off-site translators to travel from District offices to school sites for parent conferences.
    And, it’s wonderful to have a discussion about successful teachers.  This is not a simple time to succeed in a public school.  But, it’s an exciting opportunity for society to define what we expect of the next generation of teachers and how the teaching corps of California must change as we replace retiring Baby Boomers in the next 10 to 15 years.
    Chris Stampolis
    Trustee, West Valley-Mission Community College Distict
    State Board Member, California Community College Trustees (CCCT)
    PO Box 270, Santa Clara, CA 95052
    408-390-4748  *

    Report this comment for abusive language, hate speech and profanity

  5. Pingback: Judge says student scores must be in teacher evaluations | Parents for Better Schools

  6. David B. Cohen

    If standardized test scores were removed from the equation and replaced with more meaningful measures of student learning, it would be so much easier to make progress here.  It’s our obsession with supposedly “objective” measures that facilitate sorting, ranking and rating that will be our undoing.  No high performing private schools are trying to go this route (as far as I know – let me know if I’m wrong).  None of the highest performing states or districts needed such methodology in the past.  None of the top-performing nations are using test scores to evaluate teachers.  If we focus on training and professional development for administrators and teachers, provide the resources everyone needs to do their job, and have schools and districts working together with their communities to be transparent about their work and open to feedback, it would not be necessary to rely on low-quality standardized dreck.  I realize I just offered up a lot of “ifs” – but taking steps in the wrong direction won’t help at all.  James Popham, an expert on teaching, learning, and assessment, said that student tests for teacher evaluation would only work if two conditions were both true: the tests must be instructionally sensitive, and the teachers must know what will be on the test.  Popham says that with state tests, neither condition is true.  Tests are not instructionally sensitive because many students can answer many questions on the test prior to the start of instruction, and their ability to answer more of the questions next time depends on factors other than the teacher’s instruction.  Additionally, as Popham sees it, there’s too much guesswork about which of the standards will be tested.  I would add that, as a high school English teacher, I do know that most of the standards I’m supposed to use are not on standardized tests.  For all those reasons, and more, I hope that LAUSD and UTLA can find ways to uphold the law and the ruling without making things any worse than they are with regards to testing. In the long run, a revising or replacing the Stull Act would also be a welcome change.

    Report this comment for abusive language, hate speech and profanity

  7. navigio

    Well, I’m sorry, but the issue is not as clear-cut at the headline and article imply. First off, the judge said exactly the opposite as it relates to district standards (44662(b)(1) is part of the stull act):
    “Pursuant to section 44662(b)(1), the District is not required to use student test scores in measuring pupil progress toward District standards for purposes of teacher evaluations. “

    Where the connection must exist is on the measure of progress toward State standards. However, even there, the ruling is not as clear cut as implied. Firstly, it places much of the onus on the district to decide how to do this. Secondly, it bases the applicability of CSTs as a measure on the fact that the district created AGT (and thus it must think CSTs are relevant). This implies that if the district eventually finds AGT lacking, and/or does not use them, the applicability of CSTs would dissolve. Thirdly, the judge admits that just because current CST data is not available in time for the end of year evaluations, historical (read ‘older’) data can still be used. This means a year X teacher evaluation will be based on CST results from year X-1. Maybe not a huge deal, but something that will likely introduce the basis for a fair labor practice dispute. This is important because he made a point in this ruling that PERB had no jurisdiction over whether the stull act was being complied with or not, but will likely have jurisdiction over how that lack of compliance is addressed (essentially it means the union must agree with any metric used).

    And finally, as expected, the judge dedicated a section to the term ‘reasonably related’. And he concluded that the term only means there must be some connection, not how the measure is implemented, or more importantly, whether that connection even be direct.

    “The Stull Act does not say how pupil progress should be factored into the evaluation, leaving it to the school district’s discretion. It is perfectly appropriate for a school district to incorporate pupil progress into the evaluation indirectly, through other factors in assessing the teacher or principal.”

    This seems to be saying that even in the case of state standard progress, some indirect measure can be used to determine whether students are making progress.

    To be honest, I do not see this as a clear win for EdVoice, and it could very well be a loss if only for the fact that test scores were explicitly exempted as a requirement for progress toward district standards, and were even given indirect and union-approved status for state standards.

    That said, any bets on the likelihood that AGT will simply be proposed as the compliance metric?

    And its odd that no concerns about conflict of interest on Deasy’s part were raised given his relationship with EdVoice and the fact that he is the figurehead of the district-side negotiations against the unions.

    Report this comment for abusive language, hate speech and profanity

  8. John Fensterwald - Educated Guess Post author

    I believe you misread the decision, Navigio. The judge did say that using test results per se is not required to establish a teacher’s impact on student progress on district standards, but some measure or measures of student progress still must be used. Those measures could be district assessments, success rates on AP, student portfolios — the district must choose  something or a combination.

    The use of CST scores as part of teacher evaluations is clearly required under the Stull Act. That was the easy part of the decision.  The AGT method is one approach, but if the district in the end chooses not to use it, the requirement to use CST scores remains.

    Judge Chalfant talks about indirect and direct use of test scores. I didn’t understand what he meant, and a lawyer involved in the case didn’t either. Perhaps it will become clearer over time.

    Report this comment for abusive language, hate speech and profanity

  9. Pingback: Ed News- June 12, 2012 « tigersteach

  10. navigio

    Sorry if I was not clear John. My first point was that ‘scores’ are not required to assess progress toward district standards. The judge said this explicitly. It is true that some measure of progress is required, but it need not be scores. I think this is an important distinction, especially given the use of ‘scores’ in the title. And although you did make the distinction between district and state standards in your article, I dont think it would have been clear enough to someone who had not actually read the decision.
    I agree with you that the portion dealing with state standards is less than clear, however, he did say
    ‘but [...] the District has discretion in whether pupil progress must be direct or indirect in this evaluation.’
    which, to me implies that he is giving freedom to the district as opposed to the alternative of restricting its behavior. In the context of how this relates to CST scores and their relationship to student progress toward state standards, it seems one possible reading is that something other than CST altogether might be used to divine progress toward state standards. In other words, even though CST might be one measure, another, indirect one would be as viable in terms of the question of compliance. In my opinion, he actually went out of his way to introduce the notion of indirectness so I think he was trying to make a point with it. The mention of district discretion is clear.
    In any case, the gist of the entire ruling is essentially to focus on compliance and not methodology. He made the point repeatedly that he was not concerned with the how but with the whether. He basically said the district has not even pretended to include any level of student progress in the evaluation and thus is clearly not complying with the stull act. He also made the point that progress can be defined in many ways. And then he seemed to say, how they comply is really up to the parties involved. If a lawyer involved in the case did not understand that portion, it’s not surprising we may disagree in our interpretations.   :-)

    Report this comment for abusive language, hate speech and profanity

  11. Gary Ravani

    I read a joke recently that could well be applied to public education policy proposals and actions recently taken in CA, by the courts, some members of the legislature,  the popular media, and certain self-described education “reform” groups. It would go like this: What’s the difference between education policy in CA and the Titanic? The Titanic had a band.
    Bill Lucia of Ed Voice states: “there is no status quo going forward.” To the contrary “status quo” is exactly what this ruling, and Ed Voice’s lawsuit could bring. CA enacted test based accountability in the mid-1990s and then in the early 2000s NCLB doubled down on those wrong headed prescriptions.  And now through some weird alchemy, awful NCLB style policies are somehow to become “reform” gold by heaping yet more scapegoating, euphemistically called accountability measures, on to teacher evaluation systems. Isn’t over a decade long enough for a set of policy prescriptions to become the “status quo?”
    Just what part of the lack of reliability and validity asserted by the National Research Council for AGT/VAM style methodologies is it that people like Deasy don’t understand? There is no way to create reliable cohorts of comparable student groups in schools. There are too many variables. Changing any variable used in the AGT/VAM statistical models, and the models are created by subjective human beings, creates wild differences in teachers ratings over time.
    CA and the federal government are spending upwards of billions of dollars to create new assessments, partly because the assessments need to be aligned with newly adopted standards, and partly because the current assessments are narrow, require little critical thinking, and are invalid and unreliable at either end of the performance spectrum–for students let alone teachers. The assessments were not designed to evaluate teachers. It’s like trying to take your body temperature using your bathroom scale.
    Warren Buffet often talks about a “Circle of Competence.” That is, he understands that just because he has a skill set that provided him with great success in investing it does not follow that he has expertise in everything. (Then he went ahead and put much of his fortune under the control of Bill Gates and his Foundation. Gates does not appear to appreciate the limits of  a “Circle of Competence.” Go figure.) Public education is subject to the influence of many who are outside the “circle of competence” when it comes to education policy.  These folks can spare me the lofty rhetoric about what teachers need to be able to do and be  “what is part of being a good teacher in 2012.”  There are all kinds of elected boards with all kinds of legitimate policy decisions to make. Public hospital boards for example. You do not find these boards exhorting the hospital’s doctors the cut a little to the left or a little to the right in the surgery. In education we get tsunami’s of those exhortations.
    It can be hoped that Navigio’s “analysis” of the judge’s rule is correct and this ruling is much more limited in scope than many believe/hope. Data from valid and reliable tests could be one of a number of “artifacts” of student performance no more, and probably much less, than any other artifact: portfolios, writing sample, locally developed assessments, etc. that could be used in teacher evaluations if that is what is collectively bargained.

    Report this comment for abusive language, hate speech and profanity

  12. Gary Ravani

    Well, Mr. Muench, that is a question that goes directly to the argument over Price v. Value and those who value the first to the exclusion of the second.
    First, the people of CA pay very little for public education relative to the rest of the United States, 47th in per pupil spending of the 50 states.
    This harms poor and minority students most, but as results on the ACT and SAT show with CA above the national average, our more affluent and college bound do OK. Schools in affluent neighborhoods are as good as schools anywhere. That can’t be construed as a waste of money.
    The solution for the poor and minority students is to bring those conditions in affluent neighborhoods that support educational achievement to their neighborhoods. Make sure, as much as possible that kids start school on an even playing field, which is not at all the case now. Then there’s “summer loss,” which is much more sever for poor and minorities. Make sure all kids have maximum access to the cultural advantages that middle class kids do. Then we have: parent’s access to living wage jobs building family stability: decent and affordable housing; access to health dental, and vision care; access to high quality child care and early childhood education.
    To this point are you building a mental picture on how to evaluate teachers on the above conditions?
    Remember ETS, CA’s testing vendor, conducted a study on differences in outcomes on the NAEP. ETS concluded that about a third of score variability could be attributed to school related factors. Other researchers (Linda-Darling-Hammond) cite 7-10% of test achievement directly associated with teacher inputs.
    So, now we can begin to refine the evaluation model: be sure it assigns 7-10% of accountability to teachers.
    Some other points to provide context.
    The US economy continues to lead the world, of course. Annual GDP is about three times larger than the next country in ranking: China. The US has the most productive workforce in the world and 90% of those folks, who went to school here, went to public schools. CA itself has about the 9th biggest economy in the world, just ahead of # 10: the entire nation of India. The US has lost several spots in the “international competitiveness” rankings. Education had nothing to do with that, it was “instability” (aka. greed and recklessness) in financial markets.
    The leading competitive nations are mostly Nordic (aka, social democracies) where all of the support systems I noted in paragraph #4, above, are pretty much taken care of. Finland leads the pack, both in competitiveness and international test scores. Of course Finland has the least number of children living in poverty of the industrialized nations while the US has the second highest number of 29 nations, next to Mexico. Not a pretty picture.
    If its international test scores you want the US has the highest in the world in schools that have 10% poverty or less. A way to make the good old USA #1 in international tests would simply be to insure no school had more than 10% of students living in poverty. Simple. See paragraph #4 again, for a good start on that.
    Good old Finland,  near the top in scores and competitiveness, gives its teachers free education, high salaries, few standards, and supports 100% unionization. There are no vouchers, charters, or even private schools. Oddly, in some circles here, vouchers, charters,  private schools, and the elimination of teachers’ unions are the “gold standards” of “reform.” Go figure.
    You can see that there are a lot of values expressed here, and some costs too. I’m not so sure there is a direct line to teacher evaluation, except the 7-10% issue. A waste of money? Only if you find enhancing human potential and inculcating democratic values is a waste.

    Report this comment for abusive language, hate speech and profanity

  13. Paul Muench

    Mr. Ravani,
    I think these two statements are the ones that need more clarification  “Schools in affluent neighborhoods are as good as schools anywhere.” and “The solution for the poor and minority students is to bring those conditions in affluent neighborhoods that support educational achievement to their neighborhoods”.  How did you determine that affluent neighborhoods have good schools?  What conditions are needed to bring all neighborhoods to the level of affluent neighborhoods?  Are you willing to hold schools with the right conditions to standards of performance?  Are you willing to hold schools to a relative level of performance based on relative levels of meeting the conditions you are thinking of?

    Report this comment for abusive language, hate speech and profanity

  14. Gary Ravani

    Mr. Muench:
    Perhaps you are aware of schools in affluent areas of CA that do not have high test scores, high graduation rates, high AP class participation, high rates of acceptance to college. I am not, but then again I have close to 40 years in the business, what do I know?
    How about a little “project learning?” Most newspapers have a data site where you can find test/API information on local schools. I don’t know where you live, but in my area there are a couple of hundred schools listed.
    Take a “random” sample of schools, API scores and ED/EL percentages. I take about every 15th school just moving down a column (very much a presentation like sports box scores) which give me a list of about 20 something schools. Now rank order them by API score, I do high at the top. Then I do a little process developing what I modestly call the RAPI, that’s the Ravani Affluent Parent Index: I simply average the EL/ED numbers to come up with a single number.
    Now I rank order those schools using the RAPI, but this time in reverse, with low RAPIs at the top. Now what do you know, for the most part the lists are the same. High API=low RAPI.
    Hint: You’ll likely find that the task of averaging ED/EL pretty simple as they are often quite close. The exception, and I should have mentioned this, is for charter schools that often have incredibly low RAPIs (sometimes 0% EL/ED) or very high. This says two things: 1) charters, in violation of Ed Code, often have student populations that don’t reflect community demographics (charters often don’t have representative numbers of special education students either); and 2) “choice” has resulted in a re-segregation of the schools.Because of the above I skip over schools with “charter,” academy.” or “prep.” in their name when developing my list.
    None of this is new information and you’ll have to do your own homework to look up the citations.
    Anyway, the whole RAPI exercise is not very scientific, but in a crude way it can be illuminating. If you live in an area of economic/ethnic diversity that is. Take a look at the lists of schools getting School Improvement Grants (SIG). Find one in an affluent area.  Schools in affluent areas-those you note with the “right conditions” are already doing fine. Find some that aren’t.
    Here is a copy of the paragraph I wrote above re conditions needed to begin to give kids a level educational playing field:
    The solution for the poor and minority students is to bring those conditions in affluent neighborhoods that support educational achievement to their neighborhoods. Make sure, as much as possible that kids start school on an even playing field, which is not at all the case now. Then there’s “summer loss,” which is much more sever for poor and minorities. Make sure all kids have maximum access to the cultural advantages that middle class kids do. Then we have: parent’s access to living wage jobs building family stability: decent and affordable housing; access to health dental, and vision care; access to high quality child care and early childhood education.
    The “solution” noted above would probably only take an investment, primarily in neighborhoods with minority children,  about the size of the Marshall Plan in constant dollars. Should not be a problem to people of good will who have a genuine interest in see poor and minority children have an even break. I know you’ll be holding your breath.
    I am very interested in a discussion about accountability that takes into account all of the relevant issues and all of the responsible stakeholders. Recall ETS studies: Parsing the Achievement Gap & The Family: America’s Smallest Schools, They are as far away as Google. I’ll give you some hints: 1) of the various correlates with school achievement the majority of them are outside of schools’ control; and, 2) school related factors account for about 1/3 of test score variability (teachers for about a third of that).
    I am interested in accountability discussions that don’t propose the 33% tail wag the 66% dog. Nor am I interested in discussions where mentioning the 66% dog elicits knee-jerk sloganeering about “defending the status quo” or “being afraid of accountability.” In other words i am interested in reality based accountability discussions.
    Under CA’s previous testing regime, the CAP, we actually did compare schools only with schools of comparable demographics- called (may recollection tells me) “comparison bands,”
    You still need to take care here though. We would tend to  link schools with more or less common ED (economically disadvantaged) number together. That remains as crude as my RAPI. Say a school with 90% ED, based typically on free and reduced lunch numbers. A school with 30% free lunch and 60% reduced will be a very different kind of school than one with 60% free and 30% reduced. The University of Chicago has written on the subtleties of parsing differences at schools within the context of their communities.

    Report this comment for abusive language, hate speech and profanity

  15. Navigio

    Or you can use the NAPI. It doesn’t even require math.. ;-)
    School demographic pages on CDE include PEL. You can use that to sort API as well and get a similar phenomenon. Note however that PEL has the disadvantage of allowing one to opt out of reporting. The demographic page also has a percentage reported for PEL so you can use that as a kind of ‘confidence’ measure.
    And if you’re really feeling energetic you can rank STAR  proficiency rates by PEL as well. If the sample is large enough you can even figure out who declined to state by comparing their proficiency rates with the other PEL groups’ rates.
    It’s also handy that the demographics page is linked directly from the school’s API growth page.
    Granted, this takes a bit more navigation than just going to the local newspaper site, but anyone who is interested enough to read this blog should know how to do this. :-)

    Report this comment for abusive language, hate speech and profanity

  16. Paul Muench

    Mr. Ravani,
    I’d agree that all of these measures reflect something of value “high test scores, high graduation rates, high AP class participation, high rates of acceptance to college”.  But of this list only the test scores are frequent enough to be useful in making a difference in current students lives.  Using only end of schooling results maybe useful for future students, but I imagine establishing causality with only those indicators would be more difficult than using frequent indicators.
    In regards to conditions, if we could bring to pass all the things you mention as a prerequisite to schooling then it seems to me we wouldn’t really need schools.  Whatever brought those things about would make schools irrelevant.  I’ve always assumed schools would play a significant role in creating the conditions you mention.
    And thank you for this references, I will research them when I get some time.

    Report this comment for abusive language, hate speech and profanity

  17. el

    Paul, this is from the 2009 Bracey report:

    I said above that if there are to be more high-quality schools (or at least, “high-quality” schools in terms of high or rising test scores), they will have to be developed in low-income neighborhoods. Evidence for this contention comes from the 2001 Progress in International Reading Literacy Study (PIRLS).4 It is evidence that suggests the magnitude of the problem to be overcome.
    Table 1. PIRLS Performance and Poverty

    Percent of Students
    in the School in Poverty


    Percent of U. S. students attending schools in this category
















    Highest scoring nation: Sweden 561 (Finland did not participate)
    U. S. average: 542; International Average: 500; Number of countries: 27
    Several points from these data should be noted. First, the schools with, say, fewer than 10% of the students in poverty are not necessarily the nation’s wealthiest schools, although some, no doubt, are. The other 90% of students could be from white collar homes, blue collar homes, or some mix. From this statistic alone, we cannot tell. Second, the top two categories, which contain about 34% of all students, outscored the highest nation. Third, although it cannot be seen from these few figures, if the schools with poverty levels of 25-50% constituted a nation, their average score of 551 would rank them 4th among the 27 participating countries. In fact, then, the majority of American students did very well on this measure: the only American schools with scores well below the national average were those whose student population was composed almost entirely of students from poor homes.

    Report this comment for abusive language, hate speech and profanity

  18. Pingback: API has served its purpose (if it ever did) | Thoughts on Public Education

  19. Pingback: Educated Guess: API has served its purpose (if it ever did) : SCOE News Reader

  20. Gary Ravani

    Mr. Muench:
    Do I detect just a wee bit of the disingenuous in your comments about testing?
    If you give the responses, including mine, a close reading you can see what’s being pointed out; that is, that test scores are an indication of the SES status of parents, not the learning of students. International test scores, for the most part, are more indicators of child well being than anything else. It’s why countries like Finland don’t do much internal testing, they use something comparable to our NAEP, they know their kids live in households with high relative levels of SES. If you read the literature you can see that high test scores are just a “side-effect” of  an internal economic system that emphasizes equity.
    Finland: 3% childhood poverty–the US: 20%+ childhood poverty.
    That is why discussions of educational equity (the “achievement gap”) here degenerate so rapidly into sloganeering about “accountability.” And typically from people who are going to be held accountable for nothing related to education.
    At all costs a real discussion of the closing of the economic and social gaps that would result in closing the achievement gap must be avoided. You start taking about social and economic equity the next thing that comes up is the taxes necessary to support it. Can’t have that!
    Tests have value. In 35 years in the classroom I gave them almost weekly. It told me what the kids were accomplishing and gave me direction on where I thought we needed to go next. Worked.
    It is the abuse and obsessive nature of the test jihad that educators, and others who care about real learning, are opposed to.

    Report this comment for abusive language, hate speech and profanity

  21. Pingback: LAUSD Ruling Clear on the Law, Won’t Shake Up the State « InterACT

  22. Pingback: Stull Act | Access to Excellence

  23. Bill Jones

    I am flummoxed by your long reply.  Are your performance appraisals based upon taking one simple idea and expanding it into a long, turgid, treatise on how someone else should be superman?
    Let’s keep this very, very, concise.  IF the talent you ask for were available, it would be found.  Our labor markets are remarkable information gatherers. Many lecturers at our best universities in California are NOT bilingual.
    Secondly, Kahn academy has made the delivery of academic content nearly a zero opportunity cost TO PRODUCE.  You make NO mention of the quality.  Students will watch his videos, of questionable quality, derive NO new knowledge, and suffer lost study time.  THAT is a significant opportunity cost.
    If teachers were required to belabor simple ideas into long replies our public education system in this state would be dead.  What you do is a luxury this state can ill afford.  Do taxpayers NEED what you DO?  I think not.

    Report this comment for abusive language, hate speech and profanity

  24. Pingback: EdSource Today: Los Angeles Unified, union cut deal on test scores and evaluations : SCOE News Reader

  25. Pingback: draft imazeki | Thoughts on Public Education

  26. Pingback: February 25 & 26, 2013: The Annual California PTA Legislative Conference (Teacher Evaluations) | K-12 News Network

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>