ELL Advocates

Sol Stern (2008) feels that the Department of Education report that found no difference between Reading First and comparison schools was flawed. First, Stern maintains that the sample was too small. But the study included 128 schools in the RF group. That’s quite large. Many studies are done with groups of 30 or 40 children in each group. Second, Stern feels there was contamination, that the comparisons also included aspects of Reading First pedagogy. He quotes Timothy Shanahan’s observation: “The comparisons were not Reading First with non–Reading First schools, but Reading First with less–Reading First.” If so, if Reading First were effective, the Reading First children should have done better. They didn’t.

[More] »

According to the “Reading First Impact Report,” children in Reading First did no better than children in comparison groups. In the media descriptions, as well as in the report itself, there have been several serious omissions.

Not the First Failure

This isn’t the first failure of Reading First. The current dismal results are consistent with all previous analyses.

A number of studies have shown that since Reading First was implemented, there has been no change in the rate of improvement on reading tests given by the states. Members of the administration have repeatedly claimed that national (NAEP) reading scores for fourth graders are at an all-time high, but a brief look at the data in NAEP publications, such as The Nation’s Report Card, shows that nearly all of the improvement took place before Reading First was implemented. Also, American fourth graders did not improve between 2001 and 2006 on the international PIRLS reading test. In addition, there has been no reduction of the gap between children from high and low-income families on any measure. Reading First has failed every time it has been put to the empirical test.

Reading First Less Effective than Comparison Methods

The Reading First Impact Report (and previous analyses) actually showed that Reading First was less effective than the comparison methods. Reading First students had an additional ten minutes per day of instruction on the elements of reading that Reading First assumed were crucial (derived from the National Reading Panel report). That’s about an extra six weeks of instruction every year. Even if Reading First were only mildly effective, the extra time should result in higher reading scores. It didn’t. This means that time was taken from other subjects and activities, as well as recess, and students got nothing in return. This must be one of the great failures of educational research.

Credit Where Credit is Due

The foundation of Reading First is the report of the National Reading Panel, which recommended phonemic awareness training, intensive systematic phonics, and de-emphasized the role of pleasure reading in school. There has been no mention of the many criticisms of this report (e.g. the work of Elaine Garan and Gerald Coles), and no mention of the vast research supporting a different view of how children learn to read. There has, in particular, been no mention of Frank Smith and Kenneth Goodman, world famous scholars, who proposed years ago that we learn to read by reading, by understanding what is on the page. Their view, and the considerable evidence supporting it, has been mischaracterized or ignored by advocates of intensive systematic phonics. The Reading First Impact Report has confirmed that Smith and Goodman and the critics of the National Reading Panel, were correct all along.

Amid Sen. Clinton’s militant rhetoric — “put an end to the unfunded mandate called No Child Left Behind” — one revealing statement has received little attention. She is on record supporting high-stakes testing, in the form of rewarding teachers based on standardized test results.

Here’s what she had to say in response to a question about merit pay for teachers in a debate last November:

I support school-based merit pay. … The school is a team, and I think it’s important that we reward that collaboration. You know, a child who moves from kindergarten to sixth grade, say, in the same school, every one of those teachers is going to affect that child.

While arguing that merit pay for individual teachers would be “demeaning and discouraging,” Clinton likes the idea of financial rewards for all teachers in a school where test scores go up. In other words, she supports the controversial deal negotiated last fall by Randi Weingarten, president of the United Federation of Teachers in New York City and heir apparent to the presidency of the American Federation of Teachers. By no coincidence, the AFT has endorsed Clinton for president.

Making a distinction between “school-based” and “individual” merit pay may be meaningful from a trade-unionist point of view. But how does it mitigate the perverse effects of high-stakes testing on kids? Answer: it doesn’t. Arguably, peer pressure to ensure that all of a school’s teachers get a year-end bonus could be an even more powerful incentive to teach to the test, neglect any kind of instruction to which high stakes are not attached, expend time and energy gaming the system, and encourage low-achievers to drop out of school.

So it would be naive to take Clinton’s promise to “eliminate” NCLB at face value. It has about as much credibility to me as her tales of dodging sniper fire in Bosnia or — as she claimed yesterday — of being a gun-toting Second Amendment enthusiast since childhood.

Not that Hillary or John do either. But here’s what Sen. Obama had to say about NCLB on Friday in a Q&A session with high-school students in Columbus, Indiana.

Q: I want to be a teacher when I get out of college and stuff. What are you going to do with No Child Left Behind? Are you going to let it die or are you going to change it?

Obama began by saying “the assessment situation [is] really the biggest problem in No Child Left Behind.” So far so good. But then he went on to offer his analysis of that “situation”:

Now, as I said, I have no problem with a test being applied to let you know where are students at any given point. And I believe in accountability in the schools.

But the way No Child Left Behind sets it up, the test is taken in the middle of the school year. There’s no coordination between that test and some of the state tests that are taken, so you’ve got all kinds of tests, and the subject matter that’s being tested is very narrow, it doesn’t take into account progress or growth during the course of the school year, so if a kid comes in three years behind grade level and ends up at the end of the year only one year behind grade level, that teacher’s done a good job.

But under No Child Left Behind that school is still labeled a failure. So we’ve got to have a better assessment system.

In other words, he promises to build a better mouse-trap — to perfect the current system of high-stakes testing — and, of course, to pour more federal funds into that enterprise. Sad to say, this remains the dominant mindset on Capitol Hill. Members of Congress still fear that “political realities” forbid any serious rethinking of the crime-and- punishment approach to schools.

[More] »

It’s encouraging to see that both Democratic candidates are increasingly critical of the No Child Left Behind Act, as illustrated by a comprehensive report in today’s Pittsburgh Post-Gazette. Obviously, they are responding to voter sentiment they encounter every day on the campaign trail. Attacking NCLB has become a sure-fire applause line, as illustrated last week in Montana. But their statements remain rather general and they have yet to propose any coherent alternative, leaving themselves plenty of wiggle room.

For example, when Sen. Clinton — who has taken semantic games to new heights in this campaign — says she would “put an end to the unfunded mandate called No Child Left Behind,” she could end up voting later to make it a “funded mandate” with only marginal changes. Indeed, that’s what many Democrats have advocated all along.

Sen. Obama points out that Clinton was an enthusiastic backer of NCLB in 2001, which she called “landmark legislation [that] sends a clear message that all American children deserve a world class education.” Perhaps, as with her vote to authorize the Iraq war, this was another case of Hillary being cruelly deceived by George Bush. She hasn’t explained, but Bill Clinton helpfully recalled in a recent speech that “Hillary was facing this impossible vote. She said ‘Bill this is not going to work, [NCLB] is going to be a disaster.’” Prophetic (if true), but apparently political calculations won out.

Meanwhile Obama has been, if anything, even more vague than Clinton on what changes he would favor in NCLB. While both candidates have criticized the overemphasis on standardized testing, neither has explicitly called for repealing the testing mandate or the high stakes attached. Nor have they promised to eliminate the absurd requirement of “100% proficiency by 2014.”

So it would be illuminating if Sen. Kennedy introduces his long-promised bill to reauthorize NCLB and holds a vote in the HELP Committee this spring. As committee members, Obama and Clinton would then have to commit themselves to specifics. Or reveal their unwillingness to do so.

This is the fifth in a series of posts commenting on Sol Stern’s “Too Good to Last: The True Story of Reading First.”

Claim: Jeanne Chall demonstrated that most children benefit from learning the “phonetic code” of the English Language and that learning to read was not natural (page 15).

A “small improvement”

What did Chall (1967) actually find? Chall reported that in general heavier and more intensive phonics (systematic phonics) was more effective than lighter phonics (intrinsic phonics), and both were more effective than zero phonics (Look-Say, a method based on the recognition of sight words). But this was not a huge victory for phonics instruction, as some claim; the phonics advantage was modest. Chall characterized it as a “small improvement” (p. 309).

Fades over time

Chall reported that the phonics advantage faded over time on tests of reading comprehension with few differences among the methods by the end of grade four (pp. 109-113).

Other factors stronger

A close look at some of the studies Chall cited shows the modest size of the phonics advantage, and also shows that factors other than phonics can have stronger effects than phonics. Table 1 presents data from Sexton and Herron (1928).
[More] »

This is the fourth in a series of posts commenting on Sol Stern’s “Too Good to Last: The True Story of Reading First.”

Claim: Reading First has worked. It has resulted in “dramatic gains” (page 5, from the Introduction by Finn and Petrilli) and has been successful in helping economically disadvantaged children (p. 32).

There are several ways of evaluating the effect of Reading First, the reading component of No Child Left Behind (NCLB). The most straight-forward is to examine overall scores on tests. Measures considered to be more sensitive to progress made by disadvantaged children include determining the percent of children who reach proficiency (assuming this can be defined, which has not been demonstrated), and success in closing the gap between high and low-income children.

Reading First has failed to produce overall improvement on state tests, national tests, and on tests that are part of international comparisons. It has also failed to improve the rate of improvement in the percentage of children reaching “proficiency” and in closing the gap.
[More] »

This is my third post commenting on Sol Stern’s “Too Good to Last: The True Story of Reading First.”

Claim: Referring to the report of the National Reading Panel, there is a “scientific consensus” that the elements of Reading First are valid. In fact, there are “mountains of scientific data.”

Stern does not mention that the conclusions and methodology of the National Reading Panel have been thoroughly dismantled by scholars. I mention here only a few examples:

1. The panel’s endorsement of phonemic awareness (PA) training: Coles (2003) has pointed out numerous methodological flaws in the studies supposedly supporting phonemic awareness training, noting, for example, that in some cases PA training was compared to doing nothing. Members of the panel agreed with Krashen (2001a) that the overall effect size for studies relating PA training and reading comprehension using English speakers was small and not statistically significant (Ehri, Shanahan, and Numes, 2002), but this has not affected policy derived from the panel’s report.
[More] »

This is a follow-up to my earlier post commenting on inaccurate claims made by Sol Stern in his paper “Too Good to Last: The True Story of Reading First” (Fordham).

Claim: Whole language caused a decline in reading in California. Because of a new whole language policy, California’s reading scores “plummeted to near the bottom in the nation, according to the National Assessment of Educational Progress (NAEP).”

It is true that California’s fourth graders did poorly on the NAEP reading test in 1992. But there is no evidence that “whole language” had anything to do with this.

Whole language, according to (urban) legend, was introduced by the 1987 Framework committee, which I was a member of. The 1987 Framework committee never mentioned whole language. We recommended that language arts be literature-based, hardly a revolutionary idea. Phonics was never forbidden.

Even if the committee had introduced whole language, it could not have had much influence. The NAEP is given to 4th graders. It is unlikely that many 4th graders in 1992 had experienced much a new approach conceived in 1987, designed for the lower grades.

Most important, 1992 was the first time NAEP scores were analyzed by state. There was no “pretest,” no evidence that things had been better in the past. McQuillan’s analysis of other test scores in California revealed that scores were low well before the 1987 Framework committee met (McQuillan, 1998).

Of great interest, and rarely noted, is that fact that California still ranks at the bottom of the US. NAEP scores released 2007 show that California is still in the basement, in a virtual tie for last place with Mississippi and Louisiana. Dumping whole language did not improve things.

[More] »

A few weeks ago I contributed to the general cynicism about our politics when I noted, in a post comparing Hillary Clinton and Barack Obama:

Complaining that politicians tend to shade the truth … is about as useful as complaining that wolves are cold-blooded killers. It’s the nature of the species.

Today’s remarks by Obama, confronting hardened attitudes on both sides of the racial divide, tempt me to retract that verdict. This speech wasn’t composed of poll-driven half-truths; it wasn’t sanctimonious, self-serving, or otherwise calculated for near-term advantage. It took risks. The candidate condemned divisive sermons by his former pastor, the Rev. Jeremiah Wright, yet refused to condemn the man, his friend and mentor. Almost unheard of in our political discourse, Obama spoke to Americans as adults, capable of reaching fair and intelligent judgments on an issue that has divided the nation since its origins. I have never heard this level of honesty — on race or any difficult subject — from another mainstream politician.

Obama was not my first choice in this election. I have been skeptical (and remain so) about the prospect of rising above partisanship and bridging the ideological chasms in Washington, largely created by the Right, without abandoning the principles of the Left. While it’s inspiring to hear Obama speak in broad strokes about “change” — including his determination to end inequities in American education — specifics from his campaign (e.g., on what he’d actually do about NCLB) have been rare.

Still, I must acknowledge: there’s something different about this guy.

According to a Russian proverb, “Power is like a high, steep cliff; only eagles and reptiles may ascend to it.” It’s no secret which category most of our politicians belong to. But Obama? I’m not so sure. He could be an eagle at a time we could really use one.