Re: CTA President on state assessment system

[Karen Canty <kvscanty@PACBELL.NET>]

George,

It may be improbable, but not impossible.  In one grade level (7th, if I
remember correctly) in a small, upper middle class school district near
us,
100% of the students scored above average in the language arts section.
They have been trying to explain to their parents, since then, why they
have not been able to duplicate that feat!!!!

Karen

-----Original Message-----
From: Assessment Reform Network Mailing List
[mailto:ARN-L@LISTS.CUA.EDU] On Behalf Of George K Cunningham
Sent: Sunday, December 01, 2002 5:31 PM
To: ARN-L@LISTS.CUA.EDU
Subject: Re: CTA President on state assessment system

Karen,

It is possible for every student at a school or in a district to be
above the national average, but quite unlikely.  How likely would it be
that all 12 year olds in a school would be taller than the national
average for 12-year olds much less in a whole district?

George K. Cunningham
University of Louisville

>>> kac@KARENCOLE.COM 12/01/02 06:48PM >>>
As I understand it, a score on a norm-referenced test compares the test
taker to the norming sample group, not to his/her co-test-takers. Is
this right, and if so, does this mean the Woebegone (everyone above
average) effect is technically possible? Not that it matters if it's an
irrelevant test - I'm just curious.
-Karen
----- Original Message -----
  From: George Sheridan
  To: ARN-L@LISTS.CUA.EDU
  Sent: Sunday, December 01, 2002 5:02 PM
  Subject: CTA President on state assessment system


  This is the latest column on standardized testing by Wayne Johnson,
president of the California Teachers Association.

  Make no mistake about it
  California Educator
  November 2002

  Standardized testing has spun out of control. Large numbers of
children are not prepared to take these tests due to their
poverty-stricken backgrounds and limited English language skills.

  "Poor children are much more likely [than middle-class children] to
suffer developmental delay or damage," says Ruby Payne in her book A
Framework for Understanding Poverty.

  Policy Analysis for California Education agrees. In 1999 it reported,
"Poor children are two or three years behind their more affluent peers
on several measures long before their first year of school."

  In 2000, poverty was defined by Julian Palmer at Columbia University
as a family of four earning $17,524 a year. According to 1998 figures
from Columbia, the United States leads the industrialized world in child
poverty.

  Twenty-five percent of children under 18 and 33 percent of Latino
children live in poverty, and EdSource reports that 42 percent of
California's 6.4 million K-12 students are Latino.

  "Well-off white kids continue to outperform their disadvantaged or
minority peers, often by a sizable margin," says a January 2002 article
in U.S. News and World Report.

  California's Star program test scores reveal this sad reality and
little else. Scores reflect almost perfectly the socioeconomic status of
the children who are tested. And despite this knowledge, teachers are
being pushed to the limit to raise test scores. It has become the
political and administrative mantra in California: Teachers, raise those
test scores!

  We are given no assistance to help the 40 to 45 percent of our
children whose families are low-income or are living below the poverty
line.

  In California last year, we tested 4.5 million kids in grades 2
through 11. Their test results were published in every newspaper in the
state. The state then used the Academic Performance Index (API) to rank
every school from the bottom 10 percent to the top 10 percent. Guess who
was at the top and who was at the bottom?

  Two years ago, CTA had the API scores analyzed. We were shocked to
find that in the bottom 10 percent of API schools, 86 percent of the
students were poor while in the top 10 percent of schools, only 7
percent came from impoverished backgrounds. In the bottom 10 percent of
schools, 46 percent of the students were English language learners
whereas in the top 10 percent, only 2.6 percent had to overcome language
difficulties.

  In April 2001, the National Assessment of Educational Progress
reported that 60 percent of America's fourth-graders from poverty
families read "below basic" on its fourth-grade reading test. Simply
put, they can't read. Again, there's no special help for an identified
group of children who aren't making it despite the best efforts of their
underfunded schools and overworked teachers.

  Now let's take a look at the reality of testing and what it is doing
to our schools.

  The SAT-9 test, the major component of the STAR test, is a
norm-referenced test. That means no matter how the 4.5 million kids
score, there will be a top 50 percent and a bottom 50 percent. Half the
kids and half the teachers lose no matter what! Absurdly, this test is
not aligned to curriculum or textbooks. It is aligned with some of the
more than 400 academic standards. Experts on testing tell me that
setting 30 academic standards would be good, but 400 is a joke. One
referred to them as California's "wish list" of academic standards.

  The test is not aligned to what we do in the classroom. That's bad
enough, but then we make 25 percent of the kids take the test when they
don't understand English, and people are appalled that these kids score
poorly. We make another 10 percent of students - those with learning
disabilities - take the test with no accommodations. That's 35 percent
of the kids taking the test who are virtually assured they will not do
well. Guess who is going to be in the bottom 50 percent of test scores?

  A series of news articles by Sarah Tully Tapia, Keith Sharon and
Ronald Campbell in the Orange County Register, citing research by
Richard Hill, David Rogosa and others, reported that API scores have a
20-point margin of error. Despite this, schools have been put on the
list of underperforming schools on the basis of one point. You certainly
wouldn't trust an opinion poll with a margin of error of 20 points. Why
would you drive an entire education system on the basis of a test with
such a huge margin of error?

  The reporters also wrote, "Students who traditionally score lower,
African Americans and special education students, are excluded [from the
API results at their school] at a higher rate than white and Asian
students."

  James Fleming, superintendent of Capistrano Unified School District,
excluded 1,259 of the district's 3,201 special education students from
his district's API scores.

  The reporters also wrote that Thomas J. Kane of UCLA and Douglas
Staiger of Dartmouth studied the API bonus awards system and found it
had a "perverse effect on diverse schools." Fifty-eight percent of
schools, mostly those in affluent white areas, won bonus awards. Only 29
percent of schools with four or more ethnic groups won awards.

  One year a school in San Bernardino County raised its test scores by
102 points and won bonus awards. The next year its scores dropped by 105
points. This is not uncommon.

  As the Public Policy Institute of California revealed in 2000, "Much
of the variation in [STAR] test scores among urban, suburban, and rural
schools that appear in raw data can be accounted for by variation in
students' socioeconomic status and school resources."

  One of the major problems in California and the U.S. is that the
perception of public schools is based on these tests. A strong case can
be made that these STAR Test results are totally invalid, yet they are
driving public education in California. Teachers are forced to
administer them. Despite the fact that 50 percent of the students will
always score in the bottom 50 percent, teachers are threatened with
repercussions if they don't raise the test scores when it is virtually
impossible to do so. In this testing system, the rule is if someone goes
up, someone else must go down. We already know who will be at the
bottom.

  We must reform this testing nightmare in California. The educational
propaganda suggests that all children can score well on the STAR Test
when it is impossible. It reminds me of the carnival hustlers when I was
a kid. "Step up and win a prize." It was impossible to win! This system
of testing guarantees that 50 percent of our kids and teachers will be
losers.

  CTA and teachers must fight to change this terrible system. We must
have tests that are aligned with textbooks and curriculum. We must have
a criterion-referenced test. Then if 100 percent of the kids scored in
the top 10 percent, that is what would be reported. We must do away with
a system that requires 50 percent of kids to score in the bottom 50
percent of test scorers.

  Last year, CTA sponsored legislation for that exact reform and
Assembly Member Jackie Goldberg led our fight.

  Make no mistake about it. CTA will be back! We will continue to fight
to bring sanity and fairness to California's testing system. With your
support, 330,000 CTA members will win this battle.

  George Sheridan
  Northside School
  Cool, California 95614
------------------------------------------------------------------------
--
To unsubscribe from the ARN-L list, send command SIGNOFF ARN-L to
LISTSERV@LISTS.CUA.EDU.

------------------------------------------------------------------------
--
To unsubscribe from the ARN-L list, send command SIGNOFF ARN-L
to LISTSERV@LISTS.CUA.EDU.

------------------------------------------------------------------------
--
To unsubscribe from the ARN-L list, send command SIGNOFF ARN-L
to LISTSERV@LISTS.CUA.EDU.

--------------------------------------------------------------------------
To unsubscribe from the ARN-L list, send command SIGNOFF ARN-L
to LISTSERV@LISTS.CUA.EDU.