Re: Fw: SF Chronical vs tests and the VA version

["Art Burke" <aburke@vansd.org>]

NAEP is conducted on random samples of enrolled students.  The samples
are drawn by the contractor for NAEP from lists of enrolled students
submitted by districts selected for testing.  

Art

>>> wcala@rochester.rr.com 01/02/03 01:50PM >>>

----- Original Message ----- 
From: William Cala 
To: Assessment Reform Network Mailing List 
Sent: Thursday, January 02, 2003 4:46 PM
Subject: Re: SF Chronical vs tests and the VA version


The Virgina Pilot apparently just read the charts and not the text of
the study.  Now its been several months since I have read it, but I
recall that positive gains on NAEPs were negated by state policies that
limited who takes the NAEP in each successive test administration.  In
other words, states have been culling out those who will not perfom well
on the NAEP.  Of course, the same logic applies to the state tests as
well.  SATs and ACTs are another story.  Citing a gain in one year, as
did the Virgina Pilot, is next to meaningless.  

I am sure Mickey can refute the alleged gains in VA. much better than
I.

BC
  ----- Original Message ----- 
  From: PAVURSOL@AOL.COM 
  To: ARN-L@LISTS.CUA.EDU 
  Sent: Thursday, January 02, 2003 12:20 PM
  Subject: Re: SF Chronical vs tests and the VA version


  Here's one VA response to the study.
  Mickey

  State's SOLs pass another test
  The Virginian-Pilot
  © January 2, 2003 

  Every reform has its backlash. High-stakes testing in public
education is no exception. Nor should it be. The consequences are too
great for states across the nation to turn to radical new models of
accountability and expect that every version in every state will turn
out right the first time. Assessment and revision should be constants in
any reform. 

  To that end, a broad new study performed by researchers at Arizona
State University and financed by teachers unions must be considered
carefully, as must other studies that are sure to follow in the years
ahead. Billed as the largest study to date, the report suggests that
rigorous testing linked to tough consequences does little to improve
achievement and may even worsen academic performance and dropout rates.
The study reviewed student performance on SAT and ACT college entrance
exams and the National Assessment of Educational Progress (NAEP), which
is administered nationally every four years. 

  While scores were going up, sometimes dramatically, on the
high-stakes state tests, the study found that more states were slipping
against the national average on the independent exams than were gaining.
The conclusion of the study's authors is that ``teachers are focusing so
intently on the high-stakes tests that they are neglecting other things
that are ultimately more important.'' 

  Those are sobering results, but before they bring education reform to
a screeching halt, several factors need weighing. 

  First, some states are bucking the trend detected by the study, and
Virginia is one of them. In Virginia, improvement on other test measures
has accompanied the SOLs or Standards of Learning tests. For instance,
the average math score for Virginia seniors on the SAT in 2002 was five
points higher than in 2001 and nine points higher than in 1997, the year
before the first SOL tests were administered. 

  Reading gains of four points between 1997 and 2002 were smaller, but
steady. 

  As for the NAEP, the national study placed Virginia in the negative
category on 4th grade reading scores. But that's deceptive. The Virginia
scores were down from 221 to 218 between 1992 and 1998. But the major
drop was between 1992 and 1994 when the score fell eight points to 213.
That dismal showing was part of the impetus for education reform in
Virginia, not part of the result. 

  Revised SOLs were adopted in 1995 and the first statewide SOL tests
were administered in 1998. Meanwhile, the NAEP 4th grade reading scores
went up five points between 1994 and 1998. 

  The larger problem with usin
g the NAEP reading scores as a measure of
SOL reform in Virginia is that the most recent test results are from
1998. High-stakes testing was too new then in Virginia to make much of a
judgment. A far better measure will come later this year when results
from the 2002 tests are released. 

  Meanwhile, on the NAEP math tests, which follow a different
timetable, the trends are positive. Average 4th grade math scores in
Virginia went from 221 in 1992 to 223 in 1996 and 230 in 2000. Eighth
grade gains went from 268 in 1992 to 270 in 1996 and 277 in 2000. 

  A second major concern of the national study is that rising dropout
rates accompany test-driven graduation requirements. In Virginia, the
Class of 2004 is the first that will have to pass a prescribed number of
tests before graduating, so its probably too soon to judge here. 

  But headed toward 2004, the Virginia dropout trend continues in the
right direction. The percentage of dropouts has declined every year
since 1996-97. 

  None of this is to say that there are not legitimate concerns about
teachers focusing too narrowly on prescribed instructional material. But
if tests accurately reflect seasoned judgment about the body of
knowledge that students ought to obtain, then there is value in driving
that material home. 

  Critics of the SOLs need to remember the climate in which reform was
born. For years, Virginia administered a middle-school Literacy Passport
Test and year after year, about one-third of students failed its
relatively simple requirements. Meanwhile, a high percentage of Virginia
high school graduates had to take remedial courses in reading and math
when they arrived at college, and many employers complained about the
declining skills of high school graduates. 

  Today, scores are rising across the state on tests that are far more
rigorous than the Literacy Passport. Enormous energy is being focused on
children and school districts that were often ignored in the past. Both
Democratic and Republican politicians are invested in the promise to
Leave No Child Behind. The focus on accountability reassures a
cost-conscious public that investments in education are linked to
results. 

  Yes, Virginia must discourage teaching narrowly to a single test.
Yes, dropouts rates must be closely monitored as tests equate more and
more with graduation. But all the trends are in the right direction in
Virginia. It is no time to turn back now. 





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