Re: Elementary Schools Post Lower Scores

["gerald bracey" <gbracey@erols.com>]

Yes, that was the piece.  Bob took a 50-year perspective on using tests to
improve education.  I use one of his closing paragraphs in speeches (I break
it into two for effect):

"As someone who has spent his entire carreer doing research, writing, and
thinking about educational testing and assessment issues, I would like to
conclude by summarizing a compelling case showing that the major uses of
tests for student and school accountability during the past 50 years have
improved education and student learning in dramatic ways.

Unfortunately, that is not my conclusion.  Instead I am led to conclude that
in most cases the instruments and technology have not been up to the demands
that have been placed on them by high-stakes accountability.  Assessment
systems that are useful monitors lose much of their dependibility and
credibility for that purpose when high stakes are attached to them.  The
unintended negative effects of high-stakes accountability uses often
outweight the intended positive effects."

JB


----- Original Message -----
From: "George Sheridan" <learn@jps.net>
To: <arn-l@interversity.org>
Sent: Thursday, August 19, 2004 1:31 AM
Subject: Re: [arn-l] Elementary Schools Post Lower Scores


Jerry Bracey wrote: "I think Bob did put the most salient graph in an
article somewhere."

Jerry may be referring to Robert Linn's article, "Assessments and
Accountability," in the March 2000 issue of Educational Researcher , which
reviewed factors that influence the trustworthiness of assessment and
accountability systems. I've cited this article on ARN before. After all
this time it's still a valuable reference. Linn warned against the
"overreliance" on high-stakes accountability uses of assessments that was
already prevalent at that time.

Linn wrote that improved test scores are not necessarily indicators of
actual gains in student learning. In his words, the evidence raises "doubts
about the generalizability or robustness of the putative gains that were
reported on standardized tests." He summarized research on the "lack of
generalizability of the apparent gains on a test that is reused for several
years," a problem partly due to "focused teaching to the test encouraged by
accountability uses of results." He cited, along with other evidence, the
"saw tooth" pattern of "a sharp drop in scores the first year a new test is
administered followed by gains on administrations in following years." This
point was illustrated with a graph which I believe is the one Jerry
describes.

Linn suggested that the higher the stakes, the more likely teachers and
administrators are to "teach narrowly to the test."  He questioned whether
there is either the know-how or the will to develop tests that are
"sufficiently immune to this type of corruption." "We should not expect,"
he concluded, "inexpensive tests designed for other low-stakes purposes to
withstand the pressures now being placed on them by high-stakes
accountability systems."


At 10:29 AM 8/18/2004 -0400, Jerry Bracey wrote:
>I agree with Kathy--"focus and enthusiasm" has nothing to do with it.
>Familiarity with the content and format does.  Of course, these results
have
>huge implications, not only for CA, but all states.  No state can afford a
>diminishing-returns curve under NCLB.  Some states' plans to attain 100%
>proficiency by 2014 call for linear increase while others, including CA,
>have accelerating curves--little improvement is needed now, much
improvement
>need in the last 5 years or so.  These plans are equivalent to taking out a
>mortgage with a huge baloon payment at the end and hoping that the bank has
>gone away when it finally comes due.
>
>Related is a study that Dan Koretz, Bob Linn and Lorrie Shepard did about
15
>years ago and, unfortunately, never published although I think Bob did put
>the most salient graph in an article somewhere.
>
>They studied districts which changed tests.  In the first year of the new
>test, scores were well below where they had been on the old test and
>then--surprise!--they rose over a period of three years to where they had
>been on the old test.  Then the Sneaky Three came in and administered the
>old test--scores fell to where they had been in the first year of the new
>test.
>
>JB
>
>----- Original Message -----
>From: "Kelley Messina" <dancinglight@sbcglobal.net>
>To: <arn-l@interversity.org>
>Sent: Wednesday, August 18, 2004 8:23 AM
>Subject: Re: [arn-l] Elementary Schools Post Lower Scores
>
>
> > Okay, so it's been 15 years or so since that
> > psychological measurement course I took that covered
> > standardized testing; my memory is a little fuzzy.
> >
> > Please feel free to clarify the following for me:
> >
> > "Experts also said the results followed a familiar
> > pattern found in other testing programs across the
> > country: Scores begin to trail off about three years
> > after new exams are introduced as schools' initial
> > focus and enthusiasm begin to fade."
> >
> > I seem to remember something that said standardized
> > scores always do this; when a new test is used, scores
> > drop, then increase each year until they reach a
> > plateau.  I don't think this is a new pattern that
> > experts are just noticing since high stakes testing
> > took over public education. I also don't believe that
> > lack of focus or enthusiasm is the reason for the
> > plateau.  Who are these "experts?"  Any statisticians
> > out there want to weigh in on this?
> >
> > Kelley
> >
> >
> > --- Peter Farruggio <pfarr@berkeley.edu> wrote:
> >
> > >
> > >
> > > August 17, 2004
> > >
> > > Elementary Schools Post Lower Scores
> > >   After several years of gains, state test results
> > > in math and English
> > > either declined or showed no improvement at a
> > > majority of campuses.
> > > August 17, 2004
> > > By Duke Helfand and Doug Smith, Times Staff Writers
> > >
> > >
> > > After several years of marked gains, the majority of
> > > public elementary
> > > schools in California posted lower scores or showed
> > > no improvement this
> > > year on standardized English and math tests,
> > > according to data released Monday.
> > >
> > > State officials tried to make the best of the
> > > disappointing results,
> > > pointing out that scores remained significantly
> > > higher than they did at the
> > > beginning of the current testing program four years
> > > ago.
> > >
> > > But fewer than half of the state's elementary
> > > schools showed any increases
> > > this year, compared with the nearly 90% that
> > > improved last year, according
> > > to a Los Angeles Times analysis that combined
> > > English and math scores.
> > >
> > > Middle schools showed little change this year on the
> > > tests that are tied to
> > > the state's academic content standards. High schools
> > > continued to struggle;
> > > only one-quarter did better this year, significantly
> > > worse than the
> > > previous year. The slowdown put a dent in the
> > > state's efforts to meet new
> > > demands under the federal No Child Left Behind
> > > education law. It requires
> > > all students to be proficient in English and math
> > > within a decade.
> > >
> > > Even after the increases of the last several years,
> > > far less than half of
> > > California students are proficient in
> > > English-language arts or math. For
> > > example, just 30% of third-graders were proficient
> > > in English, down 3
> > > points from last year. And only 35% of sixth-graders
> > > were proficient in
> > > math, up 1 point from last year.
> > >
> > > "This is not where we want to be. This is not where
> > > we hoped we would be,"
> > > said Jack O'Connell, state superintendent of public
> > > instruction. "These
> > > scores should be viewed as a wake-up call for us
> > > all."
> > >
> > > Educators cited several factors to explain the
> > > lackluster performance,
> > > including budget cuts that have eroded teacher
> > > training programs and
> > > increases in the numbers of students who are still
> > > learning English.
> > >
> > > Experts also said the results followed a familiar
> > > pattern found in other
> > > testing programs across the country: Scores begin to
> > > trail off about three
> > > years after new exams are introduced as schools'
> > > initial focus and
> > > enthusiasm begin to fade.
> > >
> > > "The first year or two it was brand new. People were
> > > really trying to make
> > > it work," said Eva Baker, co-director of UCLA's
> > > Center for Research on
> > > Evaluation, Standards and Student Testing. "Now the
> > > motivation is at a
> > > different place. No matter how hard we work, we're
> > > getting incremental" change.
> > >
> > > Nearly 4.8 million students in grades 2 through 11
> > > took the standards tests
> > > last spring in English, math, history and science.
> > > Individual student
> > > scores are being sent home; district and school
> > > scores are available on The
> > > Times website, latimes.com/schoolscores. The state's
> > > goal was for students
> > > to reach at least the "proficient" level in each
> > > subject.
> > >
> > > Forty-five percent of the state's elementary schools
> > > had a decline in the
> > > percentage of proficient students this year, 10% had
> > > no change, and the
> > > remainder showed an increase, according to The
> > > Times' analysis.
> > >
> > > On average, the state's elementary schools showed
> > > only slight gains this
> > > year in the percentage of students at the proficient
> > > or advanced levels on
> > > the English and math tests combined.
> > >
> > > Montebello Park Elementary, a year-round school
> > > where more than two-thirds
> > > of the 422 students are still learning English,
> > > struggled to keep its
> > > scores up. Overall, the percentage of students
> > > proficient in English and
> > > math at the East Los Angeles campus dropped by 1
> > > point from last year, to 22%.
> > >
> > > Principal Maria Nuñez said it's tough to
> > > consistently raise the scores of
> > > English learners at a school where the
> > > stop-and-start, year-round calendar
> > > gives teachers little chance to communicate.
> > >
> > > "It's always disappointing when there's a slight
> > > dip, but you look and say,
> > > 'Now I know what work needs to go on,' " Nuñez said.
> > >
> > > Students at Hoover Elementary in South Los Angeles
> > > showed no improvement
> > > this year on combined English and math scores.
> > > Teachers blamed the results
> > > on a combination of factors: They said that they
> > > lost classroom aides to
> > > budget cuts and that increases in testing and
> > > paperwork had eaten into time
> > > they would have otherwise spent teaching.
> > >
> > > "We're doing double work," said Queena Kim, a
> > > fourth-grade instructor.
> > > Hoover Elementary was among the schools that helped
> > > produce an overall
> > > mixed picture of achievement in the Los Angeles
> > > Unified School District,
> > > the state's largest school system.
> > >
> > > In English-language arts, for example, scores for
> > > L.A. Unified's elementary
> > > grades, except fifth, were down from last year.
> > > Middle schools showed
> > > improvement in that subject, while high schools once
> > > again declined.
> > >
> > > In math, L.A. Unified's elementary grades improved
> > > generally but at a
> > > slower rate than the year before. Scores dropped
> > > precipitously in
> > > eighth-grade algebra, and high school students
> > > taking algebra and geometry
> > > also lost ground.
> > >
> > > Los Angeles Schools Supt. Roy Romer touted the math
> > > improvement in the
> > > lower grades but acknowledged the slower pace of
> > > overall improvement.
> > >
> > > "We have our work cut out for us, but we have made
> > > tremendous gains," Romer
> > > said. "We should feel good that we're on the right
> > > track. We have a course
> > > of action . we need to stay with it. We need to do
> > > better."
> > >
> > > Steven Keller, assistant superintendent for
> > > curriculum and instruction for
> > > the Laguna Beach Unified School District, said
> > > administrators and teachers
> > > were surprised at the drop in scores in English for
> > > elementary school
> > > students. The districts' elementary scores fell from
> > > 70% of students
> > > testing proficient or advanced in English last year
> > > to 63% this year. The
> > > math scores increased from 70% to 71%.
> > >
> > > The district began a new math program that officials
> > > say led to a
> > > significant increase in 2003. But when they revamped
> > > the district's English
> > > program, officials were surprised to see the scores
> > > drop in grades 2 through 4.
> > >
> > > The Murrieta Valley Unified School District in
> > > Riverside County experienced
> > > a large dip in math scores. School district
> > > officials attributed the
> > > results to a new state policy that requires algebra
> > > to be taught in the
> > >
> > === message truncated ===
> >
> > -------------------------------------------------------
> > ARN-L archives:
> > http://interversity.org/lists/arn-l/archives.html
> >

George Sheridan
Northside School
Cool, California
95614-------------------------------------------------------
Subscribe ARN-L:
http://interversity.org/lists/arn-l/subscribe.html