Thursday, April 19, 2007

Reading first and honest use of data

At least we should be able to agree to look at data honestly. The President and his staff can not make this claim.

The President and Secretary of Education constantly
maintain that Reading First has had a positive effect
on national tests, because of an increase on NAEP
trend test scores for fourth grade reading between
1999 and 2004, from 212 to 219. (Margaret Spellings
said this at the NCLB summit in April, 2006, and it
was repeated a year later, in April, 2007 on the White
House website, with many repetitions over the last

Trend tests, considered appropriate for comparison,
were not given between 1999 and 2004. Gerald Bracey
(2006a) has pointed out that Reading First did not go
into effect until 2002-2003, so it is not clear that
Reading First deserves the credit for the increase
(according to the recent GAO report, 25 states were
funded in 2002-2003 and 25 more in 2003-2004). In
fact, several analyses show that reading scores have
not improved since Reading First went into effect.
(Fuller et. al., 2006; Lee, 2006).

Bracey (2006b) also doubted that many children in
districts that did Reading First took the NAEP in
2004. The test is given to nine year olds, and Reading
First is aimed at grades three and lower.

Data from the recent GAO report confirms Bracey’s
suspicion. The report tells us that 1,200 districts in
the US were awarded Reading First grants. That’s only
seven percent of all districts in the US.

Let us give Reading First the benefit of the doubt and
ignore all the counterarguments presented above. Let
us give Reading First even more benefit of the doubt
and assume that even though only seven percent of
districts did Reading First, these districts were
gigantic, and that 21% of all children who took the
NAEP in 2004 had Reading First. If Reading First is
to take credit for the seven-point gain, those
children would have had to outscore their 1999
counterparts by a fantastic 33 points, a score that is
nearly exactly at the 75% percentile. If the Reading
First districts contributed only seven percent of
those who took the test, they would have had to score
312, outscoring their 1999 counterparts by 100 points
(The 90% percentile in 2005 was 263).

Again, several studies tell us that there was no gain
on national tests after Reading First went into
effect, and Bracey gives us reason to doubt that many
Reading First children were old enough to take the
test. But even ignoring these arguments, all estimates
of the number of children who took the test who were
in Reading First districts makes it highly unlikely
that Reading First deserves any credit for the
1999-2004 increase.
S. Krashen

Bracey, Gerald. 2006a. The 16th Bracey Report on the
Condition of Public Education. Phi Delta Kappan 18
(2): 151-166

Bracey, Gerald. 2006b. Letter to Congressperson George
Miller and Senator Edward Kennedy, September 25, 2006.

Fuller, Bruce, Gesicki, Kathryn, Kang, Erin, and
Wright, Joseph. 2006. Is the No Child Left Behind Act
Working? The Reliability of How States Track
Achievement. University of California, Berkeley:
Policy Analysis for California Education

GAO 2007. Reading First GAO 07-161 (States report
improvements in reading instruction, but additional
procedures would clarify Education’s role in ensuring
proper implementation by states.)

Lee, Jaekyung. 2006. Tracking achievement gaps and
assessing the impact of NCLB on the gaps: An in-depth
look into national and state reading and math outcome
Cambridge, MA: The Civil Rights Project at Harvard
Post a Comment
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License.