Ta-Dah!! For Dave or anyone else-- here it is hot off the press:
Ok—Here is the information from Stahl’s chapter on Fluency in The Voice
of Evidence in Reading Research. This was sanctioned by the National
Institute of Child Health and Human Development and the various
sections were written by members or contributors of the Report of the
National Reading Panel. Its purpose is to put the panel’s research into
practice so this has the federal stamp of approval.
Stahl was a contributor to the NRP report. What he’s done is try to
pull together the various studies and the findings of the NRP and
discuss their implications for classroom teaching. I’m going to outline
the chapter section by section since so that the big picture isn’t
lost. If I have a personal comment, I’ll note it as such. Otherwise,
this is the flow of undiluted, uncherrypicked federally- sanctioned
Stahl. I think what he says is well-balanced, fair and makes a lot of
sense.
This is a draft—it is accurate in terms of content-- but no doubt has
spelling or grammatical errors but I’m not cleaning it up now. 1. Ch
starts with how fluency has been a neglected topic but is destined to
move to the forefront because of NCLB. (p. 187)
2. Then Stahl discusses models of reading development that emphasize
fluency—Chall, Ehri—and how the belief is that if kids can recognize
words automatically, they can devote energy to comprehension. This is
the same argument and the same researchers who support heavy phonics
instruction for the same reason. Note: He describes this in terms of
models and does not agree or disagree. He then moves on to the
definition of fluency (p.187-188)
3. Stahl states that conventionally, fluency is defined as 1) a
reasonable rate; 2) “accurate without too many miscues” NOTE: Stahl
uses the term “miscues” instead of “errors” thus appearing to validate
the concept—I would note that later he cites Marie Clay and others
connected to Reading Recovery to promote the use of context, rather
than decoding alone to help kids identify words. 3) prosodic—read with
expression to sound like language.
4. THEN—and this is important—Stahl (p. 188) says that these
definitions of reading (those 3 components) are over simplified and
that “Reading, however, is more complex”
5. He then qualifies and expands on that oversimplified definition of
fluency by saying, (p. 188) “Teachers assume that those who… are
struggling with the text, making many miscues, hesitating and repeating
words are struggling readers and that those who read the text
comfortably are comprehending accurately. He says, for the most part
this is true. But then he gets into the complexities and the
qualifications:
7. “Reading however is more complex. Sometimes children can be reading
accurately but do not understand what they read” He cites (Carpenter an
Paris and also Pinnell et al (1995)
8. He then goes on for 2 pages saying how the 3 components of fluency
(rate, accuracy and prosody) are not enough. He cites Pinnell’s study
for NAEP—who found that 4th graders tested in this special study
“showed that overall, oral reading accuracy was not significantly
related to comprehension “(p. 188)
9. Pinnell et all DID find however that “significant miscues were
strongly related to comprehension. ([pp.188)
10. Stahl then further supports the Pinnell findings that “oral
reading accuracy was only related to comprehension In first and second
grades with correlations in third grade and beyond dropping to near
zero.” This quote is from studies by Carpenter and Paris—also a study
by Schwnaenflugel, Kuhn , Meidnhrt, Bradley and Stahl.
11. Stahl then concludes. “Thus, oral reading accuracy may be
important only in early grades, with other factors such as vocabulary
and comprhension strategy use becoming important later.” NOTE: This
does not mean that training kids to read faster influences
comprehension since the results are correlational and correlation is
not causation. In fact, Stahl says this later on in the chapter.
11. He then distinguishes rate from accuracy and says that oral
reading rate remains important through elementary years (p.189). So
this answers Dave’s question. Rate then was associated with
comprehension. He then gets into the implications for reading/fluency
instruction:
12. His thesis is that fluency should be taught through contectual
reading, not through isolated word practice or isolated passage
practice: “Teaching children to say isolated words faster does not seem
to improve comprhension. A number of studies have examined teaching
children to say words that they know faster”—Then he lists a series of
studies. “Although all of these studies found that children’s passage
reading fluency improved, none found differences between the study
group and the control group.
13. Studies of repeated and assisted reading of connected text, not
isolated words do show strong effects of measures of comprehension as
well as on meas;ures of fluency. “Competent reading requires skills
that extend beyond the single word level to contextual reading and that
this skill can best be acquired by practicing reading in which the
words are in a meaningful context” (p. 189)
NOTE: What does this say about DIBELS practice in which some of the
practice/assessment isn’t even at the word level but has young children
call out nonsense words? This also has implications for ssr and wide
reading since the correlation can mean that reading/comprehension can
influence rate even as rate can influence comprehension.
14. He then cites criticisms of round robin reading and appropriate
reading feedback. He here cites Clay and suggests cueing children to
use their knowledge of words and meaning to decode unknown words in
context and says that these methods are more effective than round robin
reading.
15. Then he gets into “repeated readings, neurological impress and
similar techniques” as ways of promoting fluency (p.191). He says that
the problem was that most of these studies measured these techniques on
fluency of particular PASSAGES—in other words, the kids read passages
repeatedly and then their improvement or change was measured on a
posttest of the same passages. This is important—Stahl states that
these studies did not measure whether the improvement translated to
general reading.
NOTE: These are almost the exact words used by Michael Pressley in his
independent research on DIBELS—He states, DIBELS training makes kids
better in DIBELS and that’s it.
16. In other words, as with isolated word training, kids got better
and faster on calling out isolated words – but they did not improve in
general reading and comprehension. So the fact that the studies showed
improvement in passage fluency does not mean that the training results
in transfer to general reading achievement. However, when repeated
readings are combined with other measures, such as previewing a text or
listening, seems to be effective for first graders and struggling fifth
graders (NOTE: That is a really narrow population of effectiveness!)
17. Stahl then cites the NRP on repeated readings as a means of
training fluency as connected to achievement: “ But the panel reported
[as a qualification to the positive results of expanded repeated
readings on first graders and fifth graders]—
“It certainly cannot be inferred that repeated reading or other guided
repeated oral reading procedures would be effective in raising reading
achievement on the basis of these studies alone. (in Stahl, pl 191
citing NRP, p. 3-16)
NOTE: Melanie Kuhn’s study (that I cited and offered to send to you
all) was done after this chapter by Stahl was written and her
conclusions support those of Stahl’s here. Her repeated reading/fluency
trained group improved in fluency but not comprehension. On the other
hand the group that did lots of reading improved in BOTH fluency and
comprehension.
18. Stahl states that fluency training, including repeated readings,
has more of an influence on fluency than it does on comprehension:
“That the effects on measures of reading comprehension are lower than
those for fluency measures are not surprising.. Reading comprehension
is less directly related to fluency training than are more direct
measures of fluency. Transfer is always more difficult to find. However
the findings seem encouraging.” (p.192).
19. Stahl then gets into a huge section on how important authentic
practice is in reading. He discusses why the NRP did not find enough
experimental studies to support SSR—they left out the Elley Book Flood
studies and they used fluency rather than comprehension and vocabulary
for the outcomes of SSR. He then recommends SSR as part of every school
day where kids read books of their choice. I posted a lot on SSR and
it’s in my book so I’m not reposting all of that.
I20. n summary in “putting this all together” Stahl states, “Although
many successful approaches used repeated readings of a single text,
repetition does not seem to be necessary. Instead, it seems to be
necessary to increase the amount of reading that children do at an
appropriate level” (p.207)—THAT is a vital quote that somehow I left
out of my book, darn it.
“Although fuency—accuracy, rate, and prosody—is an important component
of effective reading, it is not sufficient to make a child a reader. ..
Our studies show that fluency is most important in first and second
grades, with other aspects of reading gaining importance in third grade
and higher.” (p. 208). “Disfluent reading can limit a child’s
comprehension but more than fluency is needed to make a child a good
reader” (p. 208)
You can agree or disagree with any of this, but this is an accurate
summary of what Stahl says the NRP says about fluency. So other than
the capitalized NOTES—where I expand or comment on some aspect of the
research (I think I’ve earned the right)—this is not me saying
this—this is the NRP
I would note that the National Literacy Panel on Minority Children and
Youth reports many of the same findings as the NRP—too much training
and focus on surface skills so kids read accurately and with
intonation—but do not comprehend. SAME with the phonics section of the
NRP_- training in phonics improves phonics skills on isolated word
lists, but does NOT transfer to comprehension.
Thanks . Elaine
_______________________________________________
Mosaic mailing list
[email protected]
To unsubscribe or modify your membership please go to
http://literacyworkshop.org/mailman/options/mosaic_literacyworkshop.org.
Search the MOSAIC archives at http://snipurl.com/MosaicArchive.
