« Scientifically based? | Main | Reading First is not a mandate! »

Reading First and Unintended Consequences


Dear Diane,

We were both irritated by Chancellor Klein's effort to mandate that all teachers in NYC use the Lucy Calkins Writing Workshop Method. So it surprised me that you were sympathetic to the Federal government for doing the same re Reading First. I sometimes think it may stem from where we see ourselves in the pecking order of power—with me always imagining myself in the position of the receiver not deliverer of orders. But the many unexpected ways intelligent people—including 5 year olds—make sense of the same world is why I love being a teacher! So let's explore this difference.

Re: your argument that it's not a mandate, but strictly voluntary. Well…yes, the states can turn down ESEA funds. But at a fairly heavy price and one that would fall on the most vulnerable kids.

Secondly, we need a longer discussion about the Reading Panel report upon which Reading First's claims are based. Of course if kids are not instructed in certain pre-reading activities they will do worse on measures of these activities. But whether such prereading skills are necessary for reading comprehension is a different question. Yes, for some kids. No, for others. That's a fact. (Even the study was far more nuanced than Reading First's interpretation.)

Thirdly, the Panelists definition of successful readers and mine are not the same. By 4th grade US kids are right behind the top scoring Finns who don't teach reading at all until kids are 7 years old. What matters after 4th grade is not whether they can but whether they are "in the habit" of picking up books. I think certain forms of learning to read are obstacles to developing the habit and love of reading. See the minority report by panelist Elaine Garan, Resisting Reading Mandates, Heineman Press.

Fourth, one-size no more fits learning how to read than learning how to do anything else. Doctors know that sometimes only trial and error can tell which medicine will work. Ditto for reading. ETS researchers Chittenden, Amarel, and Bussis, ( Inquiry Into Meaning, Teachers College Press) followed fairly typical kids learning to read and concluded that teachers need to understand how to instruct in various ways if they were to create classrooms that served all kids.

Finally, there are those unintended consequences of different approaches to teaching reading. As a teacher I started with the easiest and most natural approach first because it saved a lot of time. It's all most kids needed. When we spend time on one thing, we have less for others. So I had time to devote to science, social studies, the arts, and the sheer love of the language, written and spoken. Lucky are the Finnish children who are allowed to "read the world", not just "the word" in their early years of schooling. Given the complexity of what we lately assume all children need to be explicitly taught it's no wonder that our elementary schools teach almost nothing but literacy any more.

My friend George Wood has taken to reading to audiences Story 117—A Girl in a Cave—from the approved Reading First system produced by SRA/Macmillan/McGraw Hill. It makes one long for Dick and Jane. That's the fix I'd be in today were I still teaching Kindergarten.

Sure, our judgment is fallible, but it's at the heart of the democratic ideal—which is also fallible.

As a friend wrote me: The phaenmenl pweor of the hmuan mind azmaes me. Aoccdrnig to a rsaeerch taem at Cmabrigde Uinervtisy it doenost mttaer in waht oredr the ltteers in a wrod are, the olnyu iprmoatnt tihng is taht the frist and lsat ltteer be in the rghit pclae. The rset can be a taotl mses and you can sitll raed it wouthit a porbelm, Amzanig huh?


p.s. I'm avoiding for the moment the Inspector General's report on how Reading First created its list of Approved vendors.


I would really just like to thank you both. This discussion is as rich a conversation as exists in education today, and I look forward to reading what you both have to say every day.

Thank you both for modeling how educators can tackle the difficult and pressing issues we face with care, grace and respect.

Deborah Meier writes,

"We were both irritated by Chancellor Klein's effort to mandate that all teachers in NYC use the Lucy Calkins Writing Workshop Method. So it surprised me that you were sympathetic to the Federal government for doing the same re Reading First."

This statement raises several questions.

At least it does for me, and that's what's important.

Is Ms. Meier irritated because of Klein's effort to mandate a less-than-satisfactory program, or because he mandated something, period?

If the latter, why? Medical boards mandate certain procedures, because these procedures work and because the medical board has a moral obligation to ensure that physicians do what works.

Would it be a good idea NOT to mandate effective reading programs? If so, good for whom?

If, however, Ms. Meier is irritated at an effort to mandate a specific program (Calkins), why? What exactly is wrong with it? Does it fail to teach all five reading skills? Does it fail to teach these skills in a systematic, conprehensivem, and direct fashion? Does it fail to assess students at entry, periodically, and at the end? Does it have little experimental research to support claims of effectiveness?

I believe that Ms. Meier cannot adequately criticize Calkins' program without using the tenets of Reading First, or so-called scientifically based reading research (above). In which case, her beef with Reading First collapses like (or as) a wet sock.

Little vim, zip, and zest for life, if you catch my drift.

Dear Deborah & Diane
If there are no common, proven, effective tools for teaching reading then please explain why www.headsprout.com an online software driven tool is so effective at taking nearly every user from early phonics to early reading in 80 lessons?

First: yes, I was irritated at the idea that someone Klein mandated the use of a specific method for teaching writing to each and every child. And yes, I was irritated despite the fact that I'm generally sympathetic to the particular programs he mandated. For the same reason, perhaps, that I wouldn't want to go to your doctor! In the end I expect my doctor to use his judgment, his knowledge of my history, and his careful observation of me--above all me--to recommend treatment, tell me about other possible treatments and their plusses and minuses, and then if we decide on a path to be prepared to change course if it isn't working.

John and Martin. Thousands of kids went to school like mine 70 years ago and got taught by Dick and Jane (a look-say approach). Years later I learned something about phenemes, phonics, etc--long after I finished graduate school. Like me, most of my grade school peers in the 1930s learned to read. Some didn't because their teachers didn't think of trying other approaches. It was Dick and Jane or nothing. I appreciate a "scientific mindsert"--the power of close observation, but every claim I've known for a full-proof prescription for learning even not so complex matters, like reading, turn oujt, upon close examination, to be .... well, exaggerated!

Re "headsprout" -- I shall indeed look it up. But it might be that "early reading" is the easier part,; and the thing we actually have done pretty darn well at in American schools--thus our #2 international ranking at 4th grade. But the next step requires an engagement with the author that turns out to be more complicated. Thanks, I always enjoy discovering another good technique--and the very fact that we disagree about this seems like added evidence of how unique we all are and the very reason why I want to keep decision-making close to the ground..


The phaenmenl pweor of the hmuan mind azmaes me. Aoccdrnig to a rsaeerch taem at Cmabrigde Uinervtisy it doenost mttaer in waht oredr the ltteers in a wrod are, the olnyu iprmoatnt tihng is taht the frist and lsat ltteer be in the rghit pclae. The rset can be a taotl mses and you can sitll raed it wouthit a porbelm, Amzanig huh?

Actually it isn't. Try reading this in less than a minute:

Annoye who is a ponfiricet docdeer sluhod be clapabe of cornscutting
pegasass that sorpupt this poonehmenn. Unisg salimir pinprelics
uncle-dadboe peassags can be concretstud as well. This “peenhoonmn” is haswogh.

Dear KDeRosa,

It did take a little longer--thanks for trying to play with this. I had trouble on the first quick read with two words, but it barely interfered with my getting your point! I use the word recoding on purpose, because I've not found it helps a lot to go from one code (written symbols) to another code (spoken sounds) unless I can then go the heart of the matter--comprehension. I discovered in reading Russian novels that it wasn't worth my time to figure out how all the characters names were pronounced. And when I read in Spanish, I've discovered that making pretty accurate renditions of the sounds doesn't help me with comprehension. I also discovered as a mother that I could read a night time story aloud and meanwhile be making lists in my head, and get caught suddenly realizing I was not attending to the meaning of what was tumbling out of my mouth. I hear a lot of this kind of reading in school.

Kids running into the equivalent of Russian place or people's names have many choices. And that's why I appreciate your playing with this idea.

But where I think we flatly disagree is in your seeing the ability to read and make sense of these scrambled words is merely hogwash. For me it's evidence of the wonder of the human mind--which can't resist trying to make sense of anything it confronts. Schools should surely exploit this wonder wherever they can--for each and every child.


It is true that some skilled decoders have comprehension problems. But, the comprehension problems have little to do with reading skill. These children will fail to comprehend the same passage whether they are reading it or whether it is spoekn to them. This points to a vocabulary and underlying knowledge deficiency which plagues low SES children and which is not an easy problem to remediate. This is because teaching vocabulary is not amenable to acceleration, which is waht these kids desperatekly need since they some into school so far behind their mioddle-class peers in language skills.

But here's one thing we do know about actual reading ability and that is that kids with poor decoding skills will always be poor readers regardless of their comprehension abilities. We also know that some reading methods produce signficantly more poor decoders than other methods, especially in low SES kids. In addition, there is no evidence that these programs that produce large numbers of poor decoders are effective in raising comprehension than the programs that produce good decoders.

I fail to see the advantage of using reading methods that produce large numbers of poor decoders and that are no better than other programs at increasing comprehension. Maybe you could enlighten me.

It'd be one thing if we could reliably identify the kids who aren't going to be proficient decoders beforehand and exclude them from these reading programs you seem to favor, but demonstrably we are unable to do this in a reliable manner.

In fact, educators had to come up with the label "specific learning disability" to account for all the seemingly normal kids who unexpicably fail to learn to read in a timely manner. Almost every one of these kids will have poor decoding skills regardless of their comprehension abilities.

The aim of Reading First was to serve these kids who find reading difficult by using reading programs that are more likely to be successful with them and avoiding programs that are less likely to be successful. That is what the scientific reading research tells us as opposed to the pseudo-science and opinion relied upon by the hawkers of these ineffective reading programs.

How could you in good conscience advocate reading programs that are less likely to be effective with our most vulnerable children?

First of all, the research that Reading First is based on did not examine reading, if by reading you mean making sense of print, which is my definition of reading. It tested the ability to decode phonics, and found that limited teaching of phonics helped primary grade students do better on tests of phonics. Not surprising. However, it said nothing about comprehension. Even on the score of phonics and phonemic awarenbess this research was not generalizable to problem readers, nor to second language learners, nor to older children, some of the main groups Reading First curriculum is being used on (at least here where I live). However, like all research, even then, these are averages--some students did worse and some better in each method. Not all students did better given any one model. Yet Reading First curriclum mandates one method for all students.

Reading First curriculum also relies on a lot more use pf phonics and phonemic awareness teaching than the research supports. The research only supported a limited amount of such instruction as being useful.

So, the research did not test what I mean by reading, was generalizable to only a small subset of the population it is being used on, and does not consider individual differences.

On the question of the "misspelled" paragraph. While I too had slightly more trouble reading KDeRoss's attempt, only slightly more so. Part of that was the use of generally longer words. But it does not change the point. I could still read it fairly easily, and certainly understood the meaning, even if I had to work at getting every individual word (and getting every individual word can be poor reading – fast readers certainly do not worry about that, and to survive college, you had better not read that carefully!)

On the point of poor decoders always being poor readers. Do we know that? I don't. People read Chinese fine without any decoding for instance, so I'm not convinced. Second, you are confusing correlation with causation. Good decoding may come from being a good reader, and not the other way around. My experience in fourteen years of elementary teaching found that those who were taught using primarily decoding methods, had the most trouble reading--they were so caught up trying to sound out words, they had trouble paying attention to the meaning. And by the end of such teaching methods, many students, even if they have learned to read, probably hate reading.

Reading First is based on the NRP meta-study which examined reading comprehension.

The research did examine the subgroups you list.

RF requires a balanced reading program with phonics as one element of five.

and getting every individual word can be poor reading – fast readers certainly do not worry about that, and to survive college, you had better not read that carefully!

Perhaps for ed school that may be true.

On the point of poor decoders always being poor readers. Do we know that?

Yes, see the eye movement research.

Also, the fact that it is possible to learn to read using whole word techniques does not mean that it is desirable for most kids to do so, especially the at-risk kids.

Your anecdotal experience is interesting, but it doesn't comport with the research.

Let's start with the fact that there was no such research at Cambridge, or anywhere else. The whole paragraph is based on fraud.

We understand it, with no problem, right? Those *specific* scramblings were carefully selected, and it's skilled readers who can get past the scrambling.

Comments are now closed for this post.


Most Viewed on Education Week



Recent Comments