JOURNAL OF LITERACY RESEARCH, 39(2), 145–162
Copyright © 2007, Lawrence Erlbaum Associates, Inc.
An Endangered Species Act for
Literacy Education
P. David Pearson
Graduate School of Education
University of California, Berkeley
In our quest to move every child ahead, we have fallen behind, lost our way as
a profession. In our all-out national effort to improve the quality of teaching and
learning literacy, we have compromised some of our most enduring principles
and practices. The time has come to reverse this trend and to restore the ecology
of literacy instruction to a healthier state.
In appropriating this well-worn metaphor of endangered species from the
world of ecology, I mean to single out a set of long-standing values and practices of teaching literacy that are in jeopardy of disappearing. They are casualties
of the presumably well-meaning cycle of reform sweeping our country. If we
give the most generous reading possible to the intentions of the current reforms,
we can label these casualties not as absurdities or conspiracies but as unintended
consequences and conspiracies of good intentions; they are principles and practices we have compromised even though we never meant to. As I examine what
is happening in our federal and state governments, in both the legislative and
executive branches, I see a system in which (at least) three of these fundamental values and practices—insistence on transfer of learning, faith in teacher
prerogative, and regard for individual differences as the hallmark of learning
and assessment—have all but disappeared from the educational landscape; we
seldom hear them in our public pedagogical conversations or see them in our
curricular practices. And when we do see them, they often appear as perversions
of their original constructs.
I call these three constructs endangered because they are. On the assessment
front, in our rush to accountability, we have created such a tight link between
Correspondence should be directed to P. David Pearson, Graduate School of Education, 1501
Tolman Hall, No. 1670, University of California, Berkeley, Berkeley, CA 94720-1670. E-mail:
[email protected]
145
146
PEARSON
instruction and assessment that we have made a mockery of the whole notion of
transfer of learning as the gold standard to which all students and teachers should
aspire. If an assessment does not look just like the instruction that prepared
kids to take it, we question its validity, and, even more pernicious, we operate
in exactly the other direction by adjusting our instruction to mimic the highstakes accountability assessments. Regarding teacher prerogative, in our quest to
achieve a consistent standard of instruction for all students in all classrooms in all
schools, we have reduced teachers’ professional choices to decisions about when
to begin and end the reading block—and even that is predetermined in many
schools and districts. Finally, in the world of No Child Left Behind (NCLB,
2002) and Reading First, we have decreed that individual differences will not
exist, either in the outcomes of education or in the means by which we attempt
to achieve those outcomes.
In this essay I address each of these endangered phenomena in turn—first
transfer as the gold standard of learning and assessment, then teacher prerogative,
and finally individual differences. I close each section with some suggestions
about the kind of research and professional advocacy that we should undertake
to fix what has become quite broken in our field.
TRANSFER
I first learned, in any formal way, about the construct of transfer when I was
in graduate school in the late 1960s. But I had at least a tacit knowledge of
it even as a high school and college student in the 1950s and 1960s. What I
learned was this: It was one thing to demonstrate the acquisition of a skill or a
body of knowledge in class—with support from the guiding hand of a thoughtful
teacher or the strong pull of classroom cultural practices—but it is quite another
to show that you can use that knowledge outside of class—in another class, on
a consequential exam, or, to be excessively bold, in everyday life. The question
of transfer, at least for me (and for lots of other educational scholars—see,
for example the National Academies of Science book, How Students Learn;
Committee on How People Learn, 2005) translates into answering the questions,
How far will my knowledge travel? How robust is it? How far (as indexed by
new context and new tasks) beyond that initial instructional context can I still
apply what I acquired in that classroom?
Why has it become hard to adhere to something as reasonable and highminded as transfer? Who would not want their students or children to be held
accountable for applying their knowledge? The answer to the second question is
captured by the infamous line from Walt Kelly’s 1953 Pogo cartoon strip, “We
have met the enemy, and the enemy is us!” We, collectively, have built and/or
accepted the accountability and assessment systems we have.
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
147
The answer to the first question is more complicated, but it turns on a very
understandable conspiracy of good intentions that we, as a field, have fallen
victim to. The problem arises, the conspiracy gains its force, from the tight link
that we have created between instruction and assessment. The first premise in
this conspiratorial conundrum is that we should be suspicious of assessments
that do not match our instruction. Many of us, in fact, point to standardized
assessments as the culprits here, and we characterize them as “instructionally
insensitive” if they do not pass the face validity test; we complain when they
do not look just like what we taught. Think back to all of the explanations
you have read about the failure to achieve a significant effect in a research
study and remind yourself of how often the “insensitive outcome measure” card
was played. In essence, what we are saying is that if there had been a tighter,
a more transparent, link between instruction and assessment, we would have
fared better.
The second premise in the conspiracy works in the opposite direction—from
tests to teaching—and it is propelled by high-stakes assessment—assessment
in which teachers’ and students’ futures are on the line if a failing score is
achieved. In an effort to meet these consequential outcomes, we scurry to find
materials and activities that we think will help students do better on the test (even
though we may suspect that it will not help them develop more of the cognitive
attribute the test is supposed to measure). Such is the fate of desperate people
when confronted with absurd but consequential barriers to their well-being.
Even the most well-meaning of us enact this absurdity when we fork over
$500 for a son or daughter to take a course or buy a computer program designed
to help him or her achieve a higher score on the SAT or the GRE. Do we really
believe that our offspring know more or are better problem solvers when they
finish such a course? Probably not, but we take solace when his or her test
score goes up. When we engage in this practice, we run the risk of promoting
what Haladyna, Nolan, and Haas (1991) have labeled test score pollution, a
phenomenon that occurs when an individual earns a score on a test that is out
of whack with his or her real level of cognitive capacity on the phenomenon
that the test is designed to measure. In a developmental scenario, it might mean
that a student’s test score rises without the student actually being able to read
any better than earlier.
My favorite example comes from a long-time colleague at Illinois who grew
up in New York in the heyday of the infamous Regents’ Exam. He tested well on
the Regent’s chemistry exam because his high school chemistry class consisted
almost entirely of completing old Regents exams. So when he took the highstakes test for real, he scored high enough to be placed in the third chemistry
course at Columbia. At the end of the first week of freshman classes, he realized
that he was in way over his head and backed into a more basic course that was
more commensurate with his real (not his assessed) knowledge and skill.
148
PEARSON
My colleague’s example is telling because it suggests that no form of assessment, even performance assessment, is immune to this phenomenon. Even
so, the impact is all the more pernicious with low level multiple-choice exams
that doom certain students, particularly low achievers, to school careers of basic
skills worksheets that look just like the exams they are required to pass. The
work of Herman and her colleagues (Dorr-Bremme & Herman, 1983; Herman
and Golan, 1991) documents exactly this differential effect for lower achievers, noting that they are much more likely to have their curricular activities
determined by a score on an exam than are higher achieving students, who, by
contrast, are more likely to have their activities determined by teacher creativity
and judgment. These examples illustrate an important feature of the assessment
dilemmas we face: If high-stakes promote test score pollution in teaching to the
test, then the worst situation we can have is high-stakes (grave consequences)
on tests that exhibit low challenge, for the combination will drive curriculum,
teaching, and learning to the lowest common denominator.
I suppose the lesson to be drawn from these examples is be careful what
you wish for. On the face of it, it seems highly desirable to have assessments
that match instruction to a T; is that not what we all hope for when building
“curriculum-embedded” assessments. But once you have them, and once they
become tools of accountability, then they take on a life of their own, becoming
implicit blueprints for curriculum. And it is when they adopt this blueprint role
that real damage can be done. Truth is that our assessments cannot stand the
stress of being a curriculum surrogate. I am reminded of a prophetic warning by
Robert Linn offered just as we crossed the threshold into high-stakes assessments
in the early year part of this century (2000):
I am led to conclude that in most cases the instruments and technology have not
been up to the demands that have been placed on them by high-stakes accountability. Assessment systems that are useful monitors lose much of dependability and
credibility for that purpose when high stakes are attached to them. The unintended
negative effects of high-stakes accountability uses often outweigh the intended
positive effects. (p. 14)
So what is to be done about this conundrum? Are we to abandon curriculumembedded assessments? Are we to limit our assessments only to applications that
represent far transfer? Neither, I think. Instead we should maintain and regularly
use both kinds of assessments because each tells us something the other cannot.
The curriculum-embedded assessment that looks just like the instruction (in
an ideal world the match is from the instruction to the assessment and not
the other way around) tells us whether students learned what we taught them
in pretty much the same context and format in which we taught it. (By the
way, these assessments can just as easily be turned on their ear to evaluate the
effectiveness of our teaching and give us some feedback about how to revamp
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
149
our instruction—but we seldom use them for this purpose.) The far transfer
assessments (ideally a set of increasingly distant transfer tasks) tell us how far
the learning will travel-into what new domains, contexts, texts, or tasks. That
information is useful, too, because it provides an index of the robustness of
student learning. As such, it is much closer to the gold standard to which our
students and we should aspire. Equally as important, we should not be regarded
as failures if we do not always achieve the far transfer standard. Instead, we
should celebrate our partial successes while redoubling our efforts to build more
robust learning.
Finally, this principle of transfer is even more important for evaluating research studies than it is for evaluating student learning or the everyday enactment
of curriculum. When the improvement of learning is the primary goal of a research study, the investigators need an explicit theory of transfer—from the most
local to the most distant index of knowledge and skill acquisition; it is a moral
obligation to inform research consumers of our answer to the question, How far
will the learning travel?
One footnote to this discussion of assessment and transfer: Let me point
out one current phenomenon that would be a less serious policy and pedagogy
problem if transfer had been a part of the assessment development process.
I refer to the astonishing growth of the DIBELS (Dynamic Indicators of Basic
Early Literacy Skills; Good & Kaminski, 2002) over the past several years.
Developed by the admission of its authors (DIBELS Data System, 2006), as
a “progress monitoring” assessment (an assessment that tells us whether a kid
is “on track”), it has achieved, by virtue of its actual use in schools around
the country, the status of a diagnostic assessment, as a blueprint for shaping
instruction. Students who read, name letters, segment phonemes, or retell words
in passages too slowly are sent off to remedial reading class to complete fluency,
or at least improved rate, lessons on all of those skills.
If the developers of DIBELS had incorporated a theory of transfer into their
assessment system, they would have provided both near and far assessments of
each of the key skills assessed. As a consequence, they might be assessing, and
perhaps promoting, the use of these component skills in real reading rather than
their narrow acquisition and fluency in a highly rarified format (rapid naming
or telling) and context (mostly none). In a far transfer mode, the proof of the
pudding for the acquisition of letter-sound knowledge would not be how fast
you can name the sounds of the letters but how nimbly you can use your consolidated knowledge of letter-sound correspondences to unlock puzzling words
when reading connected text. The DIBELS phenomenon, as astonishing as it is,
is a perfect example of test score pollution—of what happens when an indicator
becomes a goal. The current situation with DIBELS represents a fundamental violation the first rule of assessment uses: Never send a test out to do a
curriculum’s job.
150
PEARSON
PROFESSIONAL PREROGATIVE
Taking responsibility for one’s professional knowledge and ensuring that it is
used wisely in making difficult decisions in the face of uncertain evidence about
how to respond to the widely varying needs, interests, and circumstances of
individuals—or what to do next for the common good—have always been the
hallmarks of any profession, including teaching. But in the current policy context, I fear that the right of teachers to exercise professional knowledge, along
with the responsibility for possessing the most up-to-date knowledge possible, is
threatened seriously. I am alarmed by rhetoric and policies that betray such disregard for teachers and teaching. I am equally alarmed by the seeming reluctance
of the teaching profession, qua profession, to accept the baton of responsibility
for guaranteeing wide distribution of the knowledge that comes with the granting
of prerogative by a trusting society.
Within the new rhetoric, the text is teacher quality—indeed NCLB demands
that every state provide every child and parent with a “fully qualified teacher”
by 2007. And there is much talk about ensuring that teachers have access to
teacher education and professional development based on “scientifically based
reading research.” That is all well and good. Indeed, it is hard to argue with
full qualifications and evidence-based professional development. But the subtext is minimal professional standards. Only an assumption of minimal, not
maximal, standards will explain why the first Bush secretary of Education, Rod
Paige, characterized effective teaching as a simple combination of subject matter
knowledge and verbal fluency (U.S. Department of Education, 2002). Only by
an assumption of professional development as a simple matter of training rather
than education can one explain the increasing popularity of “scripted programs”
that dictate what teachers both do and say and the very day on which all teachers
in a district teach a certain lesson.
How else can one explain the flirtations with a marketplace model (Finn &
Kanstoroom, 2000) to replace a professional growth model (Darling-Hammond,
1996) for determining who stays in the profession? How else can one explain
the policy of championing scores on a multiple-choice test rather than program
completion, portfolios of accomplishment, or performance examinations as the
primary pathway to a credential? Followed to its logical conclusions, the policies
we are currently implementing will lead to a generation of teachers who pay
homage to externally imposed standards rather than to the needs of children and
their families as the primary criterion for determining what students do in their
classrooms.
I want to concede one point to those who control the current policy agenda:
As a profession, we have not met our responsibilities to ensure that all of us
as teachers, whether novice or veteran, possess the very best and most current knowledge available. We have been too ready to dismiss deep disciplinary
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
151
knowledge—linguistics (from phonology to text structure to pragmatics), language development, psychology of reading and learning, orthography, literature,
and culture—as too distant from the concerns of classroom teaching to merit
much emphasis in our preservice and in-service programs. I know there are
those among us in the teacher education community who complain that it is
the external mandates imposed by laws promoting basic skills or narrow views
of teacher knowledge that prevent us from holding ourselves and our teacher
candidates to high standards of knowledge. My only response to that is that it
was a political decision to set the bar low (or perhaps in the wrong domain);
and it must be a political effort that establishes high knowledge standards (and
in the right domains). In any event, and as a result of these efforts, we have
become vulnerable to criticisms that we privilege a narrow sort of vocationalism in our teacher education programs. Both the teaching profession and the
profession of teacher educators should redress this wrong by insisting on more
rigorous standards for teacher knowledge.
That point conceded, it fails to capture the real reason why we must always
press for knowledge deep and broad as the hallmark of our profession. No matter
how definitive our research, no matter how clear the findings from studies evaluating the relative efficacy for different interventions or approaches, no matter
how transparent our policies and mandates, our schools and our society need
teachers who can apply their craft with great flexibility. Why? Because of the
undeniable fact that children differ from one another. They differ from one another when they walk through the classroom door as kindergarteners—and the
longer they stay and the better we teach, the more they will differ from one
another. They cannot easily be threaded through the same needle’s eye.
Let me illustrate this point as vividly as I can, with the aid of some visual
displays. The point I want to make is a point we should understand about much
of human existence and social policy. It is this: What is best for the group is not
always best for each individual in the group. It applies not only to education but
to social policy, health, medicine, child rearing, and just about everything else.
Examine Figure 1. It illustrates the fantasy of a researcher talking about
having found a significant difference between two treatments. When we get a
significant result, you would think, by listening to the rhetoric, that the worst kid
FIGURE 1 A researcher’s fantasy about the distribution of scores between the experimental and control groups.
152
PEARSON
FIGURE 2 A more realistic depiction of the distribution of scores across the experimental
and the control groups.
in the experimental group outperformed the best student in the control group.
We NEVER get these results in education, and, by the way, not in medical
practice either. In education we get marginal improvements like the results in
Figure 2, where on average, one group makes greater gains than another. What
a statistically significant difference means is that, on average, for any given
child you might randomly select, you will obtain a slight advantage by choosing
the statistically superior approach over the other. But when we look at the
phenomenon, not on average, but for specific individuals, the picture is quite
different. For many children, it probably would not matter much which method
they receive; some might have been better off in the control group. When policies
mete out instructional methods across the board, for everyone, we will never
know what might have happened had teachers possessed the prerogative, the
knowledge, and the skill to make differential decisions for individual students.
We ask and expect no less of doctors. We want doctors who use the most upto-date knowledge of their field in concert with situated knowledge of patients’
histories and routines to determine optimal course of action—whether exercise,
diet, or drugs, or some combination of the three—is the most likely remedy for
a symptom or ailment such as high cholesterol. We want doctors who look for
reasons to prescribe a special treatment or combination of treatments for us, as
individual patients with unique needs and profiles. Moreover, we want doctors
who can use their inquiry skills to alter a treatment when the evidence tells them
it is not working. Frohlich (as cited in Allington, 2005), conveyed this tension
between the general case and the individual patient particularly well:
In selecting appropriate therapy, choose a drug or a combination of drugs for which
there is strong evidence of effectiveness in persons with the type of problem
found in the patient: : : : In choosing between a diuretic and an ACE inhibitor,
the physician can make a reasonable selection by reviewing the patient’s history
and course: : : : We must remember that trials describe population averages for the
purposes of developing guidelines, whereas physicians must focus on the individual
patient’s clinical responses. (pp. 640–641)
Allington (2005) argues compellingly that teachers require the same prerogative to engage in “contingent” application of professional practices—matching
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
153
a particular instructional regimen with both generic research-based knowledge
and knowledge of individual learners in particular contexts. I agree. We want
teachers who use their deep knowledge of subject matter—along with knowledge of individual children’s histories, routines, and dispositions—to create just
the right curricular mix for each student and for all students. Moreover we want
them to use their inquiry skills to alter those approaches when the evidence that
passes before their eyes says the interventions they are using are not working
(just as a physician changes a treatment when an individual responds negatively.
And like physicians, we want teachers to adhere to the most fundamental of
professional ethical standards—do no harm.
We want doctors, in fact we require the medical profession, to admit the side
effects of various treatments up front so we can evaluate the risks associated
with alternative treatments (would that not be a novel addition to educational
policy and practice—to admit side effects of particular products or interventions
up front). By the way, the other caveat in the drug industry (you hear it in all
their ads) is the advice that no matter how good the research on the drug is, you
should always consult your personal physician. Would it not be nice if things
reached the point where the educational publishing industry, in the course of
advertising their products, felt compelled to say, “Oh and by the way before
purchasing this product, you should, as with all educational products, consult
your personal classroom teacher or neighborhood school.”
Even if we have “modal” or “default” programs, we will always need to
promote flexibility and versatility. Even in situations in which teacher knowledge
is uneven (as it is in so many of our schools, especially urban schools where
they cannot recruit enough certified teachers), we ought not to mandate methods
for those who have demonstrated that they possess the knowledge required to
make these sorts of differential decisions.
We need also to remind ourselves that professional knowledge, deep and
broad, is the only basis for being granted flexibility of this sort, as well as
the only guarantee to a rightfully uneasy public. But these degrees of freedom
come with strings attached. This is not a recipe for pedagogical relativism, for
“anything goes in my classroom.” Teachers who aspire to professional prerogative must accept the responsibility for keeping their knowledge current, and
they must be prepared to alter their practice on the basis of new knowledge—to
accept the possibility that new knowledge trumps old practice, no matter how
comfortably the old ways fit. In short they must possess a disposition for lifelong
learning and continual inquiry.
One caution as we move down this road to greater prerogative in shaping
programs for individual children. It is easy, oh so easy, to move from accommodating the needs of individuals to the enactment of differential curriculum
for students of different races, cultures, language backgrounds, or income levels.
And we can end up doing it under the guise of the noblest of reasons—“these
154
PEARSON
kids aren’t ready for that yet.” The choices we make, the prerogative we are
granted, in meeting the needs of individuals must always be enacted in the
name of helping all students meet the highest of standards. When we slip into
the application of different standards for different groups of students, we end up
amplifying rather than diminishing the differences in cultural capital and educational opportunity among us. The result may be an outcome that is equally as bad
as the other bad choice—a uniform curriculum for all. So prerogative yes, but
prerogative in the service of high standards and great opportunity for all students.
What is not to like in this account? Who would want anything less for our
profession? Unfortunately, there are some who appear to want much less, who
would reduce the profession to the technocratic enactment of a carefully scripted
program, one in which teachers do precisely as they are told. A few years ago I
gained an unusual and unexpected audience with one of the most influential of
policy makers in California. We agreed neither on reading policy nor on teacher
development, and we knew that we had little common ground. But we talked
anyway. I pressed the question of the highly prescriptive program adoptions
for California and the equally prescriptive state-sanctioned professional development programs designed to “train” teachers to use those programs. I made the
typical teacher educator comment, “but what will you do when the state adopts
a new program with new features and routines? Would it not be better to invest
in genuine teacher education—the kind that will travel from one program to
another—rather than narrow training that loses its utility when the current program is gone?” The response was loud, clear, and straightforward: We simply
retrain them to use the next program-step by step, skill by skill, and page by
page. The assumption in such a view is that teachers, like fast food workers,
can be trained to follow a routine that guarantees equal opportunity for all.
How can we address this dilemma? How can we retain teacher education and
a professional model in the face of great pressure, which might aptly be labeled
the McDonaldization of teaching? A good first step would be for the teaching
profession to accept, without equivocation, the bargain of greater accountability
for knowledge in return for greater prerogative in responding to the inherent
variability among students. A good second step would be to document the claim,
so central to my thesis, that teachers teach better and students learn more when
teachers (and students) have more choices. To fully meet this challenge, many
more elements would need to be in place (recapturing authenticity in assessment,
crafting quality texts for early readers, ensuring access to the tools of nonfiction
and critical literacy-to name but a few), but I would be thrilled to see these two
steps (accepting professional accountability and linking teacher prerogative to
effective teaching and learning) get off to a flying start. We do not have much
time to get it right; those who would eliminate a professional model in favor of
a marketplace model continue to press their case (Finn & Kanstoroom, 2000;
Walsh & Tracy, 2006).
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
155
INDIVIDUAL DIFFERENCES
Earlier I made the point that individual differences are a fact of life in schools
and classrooms. Kids arrive in all sizes, shapes, levels of skill and knowledge,
interests, habits of mind, and so forth. And other things being equal, good
teaching only increases that variability. Historically, teachers and schools have
accepted, even celebrated, these differences. In fact, responding to that range of
variability has become almost a “holy grail” among teachers, captured in the
age-old mantra, “There is no such thing as a best method of teaching, only a best
method for a particular child learning a particular body of content, skill, strategy, or disposition.” This value is captured in a common standard for National
Board Certification-the assertion that accomplished teachers know their students
and respond to their individuality (National Board for Professional Teaching
Standards, 2006). In fact, the very first proposition of accomplished teachers is,
“They treat students equitably. They recognize the individual differences that
distinguish their students from one another and they take account for these differences in their practice.” If one looks at the history of educational reform over
the last 100 years, it is clear that the quest for structural means of accommodating intellectual, physical, emotional, and even dispositional differences has been
uppermost in our minds as educators. Our track record on this score has been
spotty, with even the most well-meaning of efforts. And in the name of accommodating individual differences, we have created all manner of insidious side
effects: tracking, grouping, and other means of physically separating students
who differ from one another have reified and intensified differences.
At the dawn of the 20th century, the notion of individual variation was just
beginning to emerge in educational and psychological measurement, with seminal studies by Galton (1865) in England; Wundt (1896–1897) in Germany; and
Cattell (1906), who studied with Wundt, in America. But the idea that schools
were responsible for finding ways to accommodate to the fact that not all students in a given grade were ready for materials and instruction geared to that
grade level was a creation of the developmental perspective that would emerge
in post-World War I psychology and incorporated into reading pedagogy by popular writers of methods texts, articles, and basal readers such as Gates (1927,
1937) and Gray (1937). The prevailing philosophy before 1915 was that it was
the school’s responsibility to provide students with opportunities to learn, instantiated as the curriculum; but it was the student’s responsibility to make what
he or she could of that opportunity. Individual differences in achievement were
expected and accepted as a natural consequence of curriculum implementation.
But there was no underlying theory about responding to those differences to
“level” the playing field for students of different abilities and aptitudes, nor was
there any sense of moral or ethical commitment to do so. Equity—what was
156
PEARSON
“fair”—was operationally defined as providing everyone with the same opportunity to learn from the very same materials and instruction.
With the rise of developmental and differential psychology, the idea that
schools could and should accommodate to the individual differences students
bring to the classroom door gained prominence in the 1920s, particularly in
reading (see Gates, 1937). Since that time there has been a steady stream of
proposals for reorganizing schools to achieve this feat. Most notable in the
teaching of reading, at least until about a decade ago, has been the quest for
some structural mechanism to ensure that each and every child is asked to
spend the majority of reading time actually reading materials that he or she can
read on his or her own (what we have dubbed the independent reading level)
or with a little help from a knowledgeable and thoughtful teacher (what we
call the instructional level).1 These plans usually entail some sort of grouping
mechanism, such as within-class grouping, cross-class or cross-grade grouping,
or schemes designed to provide each student with an “individualized” program
of some sort. Nearly all plans of this sort, but especially the within- and crossclass approaches, try to ensure an appropriate level of reading materials for
students by creating greater homogeneity of ability or achievement within each
instructional group.
One approach, individualized libraries, takes as its goal finding the right level
of difficulty and topics of interest for each and every student in a class. Noble
in spirit, individualized library programs, where each child had his or her own
reading list and weekly one-on-one book conferences with the teacher, were in
widespread use in the 1960s/1970s (see Veatch, 1960), and they were a staple
of the whole language era that would gather strength in the 1980s.
The next major development in the individual differences movement was a
direct consequence of the “mastery learning” construct championed by Benjamin Bloom (1968) and implicit in the work of John Carroll (1963) and Robert
Gagné (1965). Mastery learning is built on a different model of equity than the
traditional notion of equity as equal opportunity as indexed by equal curriculum and equal instruction. Instead of holding instruction constant and allowing
achievement to vary, we should do just the opposite—hold achievement constant
and allow instruction to vary, thus establishing equity of outcomes rather than
opportunity. The bet was that if we could just be more precise about the essential elements involved in learning any particular domain, skill, or process, we
could bring most, if not all students to high levels of achievement, perhaps even
levels where we could state with confidence that they had mastered the domain,
skill, or process. The precision could be achieved, according to the champions
of mastery learning (see Otto, 1977; Otto & Chester, 1976), by decomposing the
1 See
Shanahan’s (1983) account of the evolution of these constructs stemming from the work
of Betts (1946) and Kilgallon (1942).
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
157
phenomenon to be learned into essential elements. Then one could teach (and
test) each of the elements to mastery, so that the learner could then assemble the
parts into a functioning “whole.” A popular variant of individualized approaches
driven by the logic of mastery learning was individually guided education, an
outgrowth of the “continuous progress schools” of the 1970s, where fixed gradelevel curriculum gave way to the goal of ensuring that each child was on the
right step of the right path toward success (Klausmeier, Rossmiller, & Saily,
1977) through mechanisms such as ungraded primaries and cross-grade clusters
in the intermediate grades.
Mastery learning was “trumped” for about a decade by constructivist pedagogies—whole language, literature-based reading, and process writing—in the
1980s and 1990s. And with the constructivist pedagogies came a radical transformation in our approach to individual differences. This radical proposal for
accommodating individual differences is implicit in the instructional logic underlying the whole language movement and is captured in Harste’s notion of the
child as curriculum informant (Harste, Burke, & Woodward, 1984). The idea is
that in the process of learning to read and write in authentic learning environments, every student will send signals to a teacher about his or her strengths
and needs; a responsive teacher will know how to read these signals (responses
to the instruction and the tasks he or she is asked to do) and provide just the
right book, assignment, scaffold, or form of feedback to move the student to the
next step of development.
But mastery learning, in repackaged form, reemerged in the middle 1990s
in the standards-based reading reforms that swept the country in the wake of
major changes in Title I (Clinton’s Improving America’s Schools Act, 1994,
and Bush’s NCLB, 2002). If one compares the logic of the standards-based
movement (specify what students should know and be able to do—measure
achievement of the standards—then hold everyone accountable to the goal of
100% of the students achieving the standard), it is wholly consistent with mastery
learning. Indeed, many of the state and district standards tests spawned by the
standards movement assess reading at an elemental or atomistic level, in much
the same spirit as the tests developed in the heyday of the mastery learning era
(see Johnson & Pearson, 1975). In the first iteration of the standards movement,
from the early 1990s through the early 2000s, educators essentially made a deal
with policy makers: “We’ll hold ourselves accountable to the goal of all students
achieving the standard. But in return, you’ll have to provide us with maximum
flexibility and prerogative in achieving that goal.”
In recent years, however, the terms of the bargain have changed, and it is
the new bargain that has all but destroyed any hope of accommodating individual differences. In this new world of reading instruction—with pacing guides
to tell us when to teach what lessons and how much time to spend on each
activity, with scripts to tell us what to say when, with monitors looking over our
158
PEARSON
shoulders to check on our fidelity to the intervention, with the expectation that
ALL students will be exposed to the very same grade level selections whether
they can actually read them or not and to grade level activities and assignments
whether they can do them or not—we have created a new instructional reality.
In the oldest equation—where equity meant equal opportunity—we fixed the
means of instruction and allowed the ends (the outcomes) to vary. In the mastery learning and the pre-NCLB version of the standards based approach, we
fixed the ends and allowed the means to vary. What we seem to be doing in the
reified curricular world we have created since the enactment of NCLB is fixing
both the ends (through our standards) and the means (through tightly monitored
curricula), allowing nothing to vary. It is students and their teachers who must
accommodate to the curriculum in this model, not the other way around. And
this logic of dual fixation gets played out for older struggling readers as well
as for younger readers because so many of the middle and high school interventions simply retrace the steps (re-teaching phonemic awareness and phonics)
that students presumably stumbled on the first or second time around.
Given my earlier arguments about the inevitability of individual differences
(kids come to school with them, and if we are any good at instruction we
only increase them), it should be no surprise to learn that I find our current
situation hopelessly misguided and morally indefensible. It defies both logic and
experience to believe that the learning of all will be enhanced by a curriculum
that meets the individual needs of few if any. It strains professional judgment
to think that we can maximize the growth of low achievers by dragging them
through a curriculum in which they can read little of the material for themselves
and perform few of the assignments without extensive scaffolding. And what
little evidence we do have on this matter suggests that meeting kids where they
are is more effective; for example, O’Connor et al. (2002) found that tutoring
that matched the reading level of intermediate grade struggling readers promoted
greater achievement growth than did tutoring that matched the grade level of
the students. Perhaps even more serious is that research is not improving the
matter; “research-based” practice of the sort emphasized in the surge of materials
developed in the wake of the National Reading Panel (NICHD, 2000) and NCLB
offers little provision for meeting individual differences. As McGill-Franzen,
Zmach, Solic, and Zeig (2006) have documented for the state of Florida, the
materials developed for and sanctioned by Florida’s “research-based” standards
promote whole class instruction and grade-level materials for everyone. But, still,
approximately one third of the students fail to achieve even the minimum levels
of reading proficiency needed to meet the Florida standard for promotion to
the next grade. In short the scientific programs are no panacea for the enormous
problems we face as a profession when it comes to closing the achievement gap.
So what is to be done to restore some respect for individual differences
and some curricular moves that genuinely assist student growth? Research is
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
159
not a great help in deciding this question because we just have not conducted
the sort of research that would give us uncontested answers. One would think,
for example, that with the long tradition of IEPs (Individualized Educational
Plans) in special education, we would actually know whether students taught
in a manner consistent with their IEPs make greater progress than those whose
instruction is guided by some random or convenient approach. We do not. By the
way, nor do we have much research validating the tiered approach to intervention
(moving from regular to remedial to targeted education settings) that seems to
be rapidly sweeping the country as the latest solution to meeting individual
needs. One would think that we would know whether it was better for students
to spend most of their time reading materials that are in their zone of proximal
development-somewhere between independent and instructional level depending
on context and expectations. We have some hints in that direction but we do
not have anything approaching conclusive evidence. We surely need evidence
to test the validity of both claims, for they are cornerstone assumptions behind
the approach to coping with individual differences that I would champion. We
owe it to ourselves as educators and to our students and their parents to find
conclusive evidence to evaluate these claims. In the interim, until that evidence
is forthcoming, we should operate on the assumption that both claims are valid.
Logic, experience, and moral responsibility dictate adherence to these principles
of individualization in the face of no or inconclusive evidence.
AND SO : : :
These, then, are my three candidates for a reading education endangered species
act—transfer, professional prerogative, and individual differences. Each candidate is a less prominent feature of the reading policy and practice landscape
than it was a decade ago. Each has been bullied off the policy landscape by
unfortunate alternatives.
I believe that some species of scholarship and some classes of educational
principles deserve to be endangered, overturned, and forgotten—when they have
outlived their usefulness and outstripped the evidence supporting their influence
on practice. As near as I can tell, the endangerment of these three constructs
does not stem from the careful analysis of research findings—findings that would
call into question their efficacy or relevance to curriculum and teaching. They
have been called into question, relegated to a minor role, by curricular and policy
developments that lay claim to a strong research base but, on closer examination,
are as driven by ideology as any movement in recent decades.
It is time to take transfer, professional prerogative, and individual differences off the endangered species list. This will not be easy. Their alternatives
(curriculum-embedded assessment, pedagogical consistency through scripting
160
PEARSON
and pacing, and equal curricular opportunity) have attracted a great deal of political capital and credibility (not to mention the force of mandate). Nothing short
of a major policy initiative to reposition these three constructs in the evolving
policy conversation, combined with a focused research program to validate their
efficacy, will do. For starters, I think we should return them to a prominent position in the policy conversation and the policy agenda. And they should remain
there until and unless compelling evidence of their ineffectiveness is presented.
How we ever let them become endangered is mystifying to me. But our past
negligence should only increase our current resolve to restore them to health.
ACKNOWLEDGMENTS
Special thanks to Richard Allington, Barbara Taylor, Susie Goodin, and Helen Maniates for careful reviews of various drafts and sensible suggestions for
improvement.
REFERENCES
Allington, R. L. (2005). Federal intrusion in research and teaching and the medical model myth. In
J. Carlson & J. Levin (Eds.), The case of No Child Left Behind legislation: Educational research
and federal funding (pp. 37–48). Greenwich, CT: Information Age.
Betts, E. A. (1946). Foundations of reading instruction. New York: American Book.
Bloom, B. S. (1968). Learning for mastery. Evaluation Comment, 1.
Carroll, J. (1963). A model of school learning. Teachers College Record, 64, 723–732.
Cattell, J. M. (1906). Conceptions and methods of psychology. In H. J. Rogers (Ed.), Congress of
arts and science, Universal Exposition, St. Louis, 1904: Vol. 5 (pp. 593–604). Boston: Houghton
Mifflin.
Committee on How People Learn, A Targeted Report for Teachers. (2005). In M. S. Donovan &
J. D. Bransford (Eds.), How students learn: History, mathematics, and science in the classroom.
Washington, DC: National Academies Press.
Darling-Hammond, L. (1996). What matters most: Teaching for America’s future. New York: National Commission on Teaching & America’s Future.
DIBELS data system. (2006). Retrieved November 10, 2006, from https://dibels.uoregon.edu/data/
index.php
Dorr-Bremme, D., & Herman, J. (1983). Assessing student achievement: A profile of classroom
practices. Los Angeles: University of California, Los Angeles Center for the Study of Evaluation.
Finn, C. E., & Kanstoroom, M. (2000). Improving, empowering, dismantling. Washington, DC:
The Fordham Foundation. Retrieved November 6, 2006, from http://www.fordhamfoundation.org/
institute/publication/publicatio n.cfm?idD101
Gagné, R. M. (1965). The conditions of learning. New York: Holt, Rinehart, & Winston.
Galton, F. (1865). Hereditary talent and character. Macmillan’s Magazine, 12, 157–166, 318–327.
Gates, A. I. (1927). The improvement of reading. New York: Macmillan.
Gates, A. I. (1937). The necessary mental age for beginning reading. Elementary School Journal,
38, 497–498.
ENDANGERED SPECIES ACT FOR LITERACY EDUCATION
161
Good, R. H., & Kaminski, R. A. (Eds.). (2002). Dynamic indicators of basic early literacy skills
(6th ed.). Eugene, OR: Institute for the Development of Educational Achievement. Available from
http://dibels.uoregon.edu
Gray, W. S. (1937). A review of C. R. Stone, Better primary reading: How to adapt reading
instruction to the varying needs of the children. The Elementary School Journal, 38, 73–74.
Haladyna, T. M., Nolan, S. B., & Haas, N. S. (1991). Raising standardized achievement test scores
and the origins of test score pollution. Educational Researcher, 20, 2–7.
Harste, J., Burke, C., & Woodward, V. (1984). Language stories and literacy lessons. Portsmouth,
NH: Heinemann.
Herman, J., & Golan, S. (1991). Effects of standardized testing on teachers and learning-another
look (CSE Tech. Rep. No. 334). Los Angeles: Center for the Study of Evaluation.
Improving America’s Schools Act of 1994, Pub. L. No. 103-382, 108 Stat. 3518 et. seq. (1994).
Retrieved December 12, 2006, from http://www.ed.gov/legislation/ESEA/index.html
Johnson, D. D., & Pearson, P. D. (1975). Skills management systems: A critique. The Reading
Teacher, 28, 757–764.
Kelly, W. (1953). The Pogo papers. New York: Simon & Schuster.
Kilgallon, P. A. (1942). A study of the relationships among certain pupil adjustments in reading
situations. Unpublished doctoral dissertation, Pennsylvania State College.
Klausmeier, R. A., Rossmiller, R. A., & Saily, M. (Eds.). (1977). Individually guided elementary
education: Concepts and practices. New York: Academic.
Linn, R. L. (2000). Assessments and accountability. Educational Researcher, 29(2), 4–16.
McGill-Franzen, A., Zmach, C., Solic, K., & Zeig, J. L. (2006). The confluence of two policy
mandates: Core reading programs and third-grade retention in Florida. Elementary School Journal,
107(1), 67–91.
National Board for Professional Teaching Standards. (2006). The five core propositions. Retrieved
November 10, 2006, from http://www.nbpts.org/the_standards/the_five_core_propositio
National Institute of Child Health and Human Development. (2000). Report of the National Reading Panel: Teaching children to read: An evidence-based assessment of the scientific research
literature on reading and its implications for reading instruction: Reports of the subgroups.
Washington, DC: Author.
No Child Left Behind Act of 2001, Pub. L. No. 107–110, 115 Stat. 1425. (2002). Retrieved April
10, 2003, from http://www.ed.gov/offices/OESE/esea
O’Connor, R. E., Bell, K. M., Harty, K. R., Larkin, L. K., Sackor, S. M., & Zigmond, N. (2002).
Teaching reading to poor readers in the intermediate grades: A comparison of text difficulty.
Journal of Educational Psychology, 94(3), 474–485).
Otto, W. (1977). The Wisconsin design; A reading program for individually guided elementary education. In R. A. Klausmeier, R. A. Rossmiller, & M. Saily (Eds.), Individually guided elementary
education: Concepts and practices (pp. 216–237). New York: Academic.
Otto, W., & Chester, R. D. (1976). Objective-based reading. Reading, MA: Addison-Wesley.
Shanahan, T. (Ed.). (1983). Edited version of P. A. Killgallon, A study of relationships among
certain pupil adjustments in reading situations. In L. Gentile, M. L. Kamil, & J. Blanchard
(Eds.), Reading research revisited (pp. 553–556). Columbus, OH: Merrill.
U.S. Department of Education, Office of Post-Secondary Education, Office of Planning and Innovation. (2002). Meeting the highly qualified teachers challenge: The secretary’s annual report on
teacher quality. Washington, DC: Author.
Veatch, J. (1960). In defense of individualized reading. Elementary English, 37, 227–234.
Walsh, K., & Tracy, C. O. (2006). Increasing the odds: How good policies can yield better teachers.
Washington, DC: National Council on Teacher Quality.
Wundt, W. M. (1896–1897). Outlines of psychology (C. H. Judd., Trans.). Retrieved December 10,
2006, from http://www.hyorku.ca/dept/psych/classics/Wundt/Outlines/
162
PEARSON
P. David Pearson serves as Dean of the Graduate School of Education at the
University of California, Berkeley, and as a faculty member in the Language,
Literacy, and Culture program. His current research focuses on issues of reading
instruction and reading assessment policies and practices. With degrees from UC
Berkeley (BA in history) and Minnesota (PhD in education), Pearson has served
the reading and literacy education profession in a range of roles: as editor of
Reading Research Quarterly and the National Reading Conference Yearbook,
as president of NRC and member of the IRA board of directors, and as the
founding editor of the Handbook of Reading Research. Those contributions have
earned him several awards: IRA’s William S. Gray Citation of Merit (1990) and
Albert Harris Award (2005), NRC’s Oscar Causey Award (1989), and NCTE’s
Alan Purves Award (2003). Before coming to UC Berkeley, Pearson served
on the reading education faculties at Minnesota, Illinois, and Michigan State.
He can be contacted at the Graduate School of Education, 1501 Tolman Hall,
No. 1670, University of California, Berkeley, Berkeley CA 94720-1670. E-mail:
[email protected]