No subject

was just as rigorous and well designed as that for the "hard sciences"
and medical science, then we would not still be facing the seemingly
endless and fruitless arguments about what is the best way to teach
reading. 

However, she does not address the critical difference between physical,
medical, and educational sciences,  and research on the best ways to
teach children to grow up and become good physicists, physicians, or, in
the case of literacy, good readers. 

Contrary to the idea promulgated by the National Reading Panel’s report
that there are few good scientific studies of reading, there are in fact
a very large number of excellent, well-designed, scientifically
rigorous, evidence-based studies of reading. In fact, just as medical
science is comprised of anatomy, chemistry, physiology, etc. and other
sciences, the educational science of reading is comprised of studies of
sensory processes, perception, linguistics, cognition, neurology, etc. 

What is not noted by most commentators on the current move toward
scientifically based approaches to reading instruction is that, contrary
to what has been claimed about the wonders of physical and medical
sciences, there is scant evidence that today’s methods of teaching
physics or medicine are founded on scientific, evidence-based methods of
teaching these knowledge and skill domains. For instance, there is a
very small base of research on what makes a good experimental physicist
or a good medical clinician and how colleges and universities can teach
people to become expert practitioners in these fields. While physics and
medical sciences have developed "hard" evidence for their theoretical
and empirical knowledge, this does not indicate that they know how to
teach this knowledge to students any better today than they did half a
century ago. 

Similarly, even though we have volumes of solid, scientifically sound
research on reading as a socio-cultural and psychological process, this
does not automatically translate into better ways to teach teachers how
to be great teachers of reading nor how to succeed with the vast
diversity in biological, chemical, physiological, psychological,
socio-cultural factors comprised of  millions of individual children. 
The issue of diversity is much more profound for teaching reading to
young children than the teaching of physics or medicine to college
students because the latter have been screened to form a much more
homogenous group as far as their reading and other language and
cognitive differences are concerned. So college teachers have a much
less difficult teaching task in these disciplines than the first, second
or third grade teachers in the public schools teaching reading or adult
educators teaching adult literacy. 

As far as the call for "evidence-based" approaches to teaching reading
using methods of medical science are concerned, this becomes almost 
quixotical. Medical science has made some of its most profound
advancements in clinical practice using "double blind" experiments in
which randomly assigned physicians administering treatments do not know
what they are administering and randomly assigned research subjects
(patients) do not know what treatment they are getting, an experimental
drug or a placebo. A third party examines data not knowing if it is the
treatment or placebo group data and looks for reactions that are
presumably based on physiological, chemical reactions that occur free of
the will of the patient and of which the patient is unaware, in the case
of the treatment, and perhaps psychological factors of which the patient
is unaware in the case of placebo effects. 

Now try this in an experimental study of different "treatments" in
teaching reading. Students and teachers are randomly assigned to
different treatment groups. But now comes a problem. Can the teacher not
know what he or she is teaching? Is it possible to teach something
without knowing what it is you are teaching? Is it possible that the
student does not know what he or she is learning? How can this be?
Disinterested third parties may look at some kind of learning outcome
data to determine if one or the other treatment is associated with
greater outcomes in learning. But what outcomes? If teachers are
teaching different things then why should the same or even similar
outcomes be expected? 

It seems that the application of the hallmark of medical science
research methodologies in clinical trials is next to impossible to apply
to the study of the teaching of reading -  or teaching any thing else
for that matter. Still, that does not mean that good, scientifically
valid, evidence-based studies of the teaching and learning of different
knowledge and skill domains cannot be designed and conducted. But they
will have to be designed and conducted in a manner that is suitable for
the contexts in which the research is being pursued. That means, in the
case of the Adult  Education and Literacy System ( AELS) of the United
States, with its many contexts for adult learning, e.g., the work place,
prisons, homeless shelters, etc.,  that  research questions and their
methodologies may have to be adjusted to be used in the contexts of both
programs and adult’s lives, and may well have to be different from those
used in the Pre-school-K-12 system. 

Further, in most cases, unlike medical clinical trials, where the
efficacy of a particular pill (drug) is being tested, in educational
research, rather than finding the most effective specific curriculum or
method of teaching, what is generally identified are principles to be
followed in teaching. Even the National Reading Panel’s report does not
prescribe a particular curriculum, like Wilson’s, Lindamood’s, etc. in
teaching phonemic awareness or phonics for decoding written words.
Rather, the research typically leads to the recommendation of the
general principle that beginning reading instruction should follow a
code-emphasis with phonemic awareness training and phonics education
provided for   teaching word recognition,  rather than a
meaning-emphasis approach. But then there are always qualifying
statements like this will be more important for students from lower
socio-economic backgrounds, less deterministic for higher SES students
and so forth. 

The implementation of general principles in education is much more
difficult than prescribing a pill in medicine. There is room for much
variability in motivation, humor, warmth, patience, etc. between
teachers and their students and this may greatly affect outcomes, as the
National Reading Panel notes. This is what makes educational PRACTICE in
uncontrolled environmental and psychological contexts much more
difficult than testing hypotheses in the much more highly controlled
contexts typical of the educational SCIENCES.