Tuesday, June 26, 2007

Reading First: FEW Gains

The Bush Regime and many educational pundits, including Peter Schrag of the Sacramento Bee regularly claim that reading scores have improved in the last few years as a result of the demands of NCLB and accountability.
Here is a letter with evidence to the contrary;

Some Useful Data from NCLB Testing
Sent to Rethinking Schools, June 23, 2007

In his insightful paper “Exit Strategy,” (Summer,
2007), Stan Karp argues that that tests mandated by
NCLB have not provided useful data that will result in
better instruction. Actually, they have: They have
shown that NCLB and Reading First have not resulted in
improved reading.

Reading First provides an extra 100 minutes a week of
instruction, approximately an extra semester every two
years. If Reading First is at all effective, the
impact should obvious. It isn’t.

On the basis of data released in April, The Department
of Education claimed that between 2004 and 2006, the
percentage of third graders in Reading First meeting
or exceeding the proficient level increased 12% on
tests of reading comprehension and 15% on tests of
fluency. Re-analysis revealed, however, that the gain
in reading comprehension was only 6% and in fluency it
was 9%. Also, these gains mask the fact that some
states did poorly: Pennsylvania declined 10% in
reading comprehension.

It also needs to be pointed out that the test used in
most states to measure fluency, Dibels, has serious
problems and is also easily available on the internet,
which means any teacher or parent can drill their
children on the actual test items.

In addition, the Department of Education violated a
fundamental scientific principle: There was no
comparison group. Any increases could have been due to
factors other than Reading First.

The most recent report, from the Center on Education
Policy, was also interpreted by the administration as
showing that NCLB-related programs have been a
success. Again, no comparison group was included, but
the Center’s report included data on elementary school
gains for the two years before and two years after
NCLB was implemented in 12 states. Before NCLB, the
yearly rate of improvement in these states was 1.93
percent, that is, 1.93 percent more students were
classified as proficient. After NCLB, it was 2.25
percent, a difference of less than one-third of one

In other words, reading scores were going up before
NCLB and NCLB did little or nothing to improve the
rate of improvement.

Before these two reports, the Department of Education
had claimed that NCLB had improved fourth grade NAEP
reading scores. Several analyses showed, however,
that the gains came before NAEP was implemented.

There is, so far, no evidence that this expensive and
time-consuming experiment has improved the reading
ability of American children.

Stephen Krashen

For those interested in the details:

“Reading First: ‘Impressive’ Gains?” by Stephen
Krashen (see also posted comments)

Problems with Dibels:
“A critical review of Dibels.” by Kenneth Goodman. In
K.Goodman (Ed.) The Truth about Dibels. Portsmouth:

Gains on national tests?
1. “The 16th Bracey Report on the Condition of Public
Education,” by Gerald W. Bracey, published in the
October 2006 Phi Delta Kappan.
2. “Selling NCLB: Would You Buy a Used Law From This
Woman?,” by James Crawford, available at
3. “Is the No Child Left Behind Act Working? The
Reliability of How States Track Achievement,” by Bruce
Fuller, Kathryn Gesicki, Erin Kang, and Joseph Wright,
published in 2006 by Policy Analysis for California
Education, at the University of California, Berkeley.
4. “Did Reading First Work?,” by Stephen Krashen,
5. “Tracking Achievement Gaps and Assessing the Impact
of NCLB on the Gaps: An In-Depth Look Into National
and State Reading and Math Outcome Trends,” by
Jaekyung Lee, published in 2006 by the Civil Rights
Project at Harvard University.

Also see: Collateral Damage: How High Stakes Testing Corrupts America's Schools. (2007)
by Nichols and Berliner
Post a Comment
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License.