THE PROFESSIONAL JOURNAL FOR EDUCATION
PHI DELTA
THE PROFESSIONAL JOURNAL FOR EDUCATION
ONLINE ARTICLE | ||
 |
T. S. Eliot, Collaboration,
|
In a corporate world that has all but abandoned the individual performance appraisal for the "work team review," schools - and perhaps remote fire observation stations - are the last holdouts. Continuing on this path, Mr. Wineburg maintains, will ensure that the chasm between schools and the real world continues to widen.
EZRA POUND called it a masterpiece, "one of the most important 19 pages in English." Conrad Aiken heralded it as "one of the most moving and original poems of our time." Even the trenchant I. A. Richards said that it expressed "the plight of a whole generation."1 "The Waste Land," T. S. Eliot's brilliant and infuriating critique of modernity, is known by anyone who has ever taken a college literature course. Its jarring juxtaposition of classic and modern, its magisterial allusions and disquieting meter have occupied - and mystified - literary critics for nearly a century, not to mention the baffled freshmen who have spent untold hours trying to decipher its meaning. Published in 1922, the poem immediately thrust Eliot into the limelight, but the story behind "his" poem remained shrouded in mystery - until 1968.
In that year, the original manuscript of "The Waste Land" was discovered, and a facsimile edition appeared three years later.2 This edition showed a typewritten version of "The Waste Land" with whole stanzas crossed out and marginal comments such as these: "too loose" (p. 32), "inversions not warranted by any real exigence of meter" (p. 52), "dogmatic deduction and wobbly as well" (p. 23). When Eliot wrote "cautious critics" in the section called "The Fire Sermon," the marginal note focused on the word "cautious," pointing out that, when speaking of London, this "adjective is tauto[logical]" (p. 27). In a handwritten section, the critic scrawled a large X across the page and in a marginal squib issued this judgment: "Bad, but can't attack until I get typescript" (p. 55).
The man behind the X was the esteemed poet Ezra Pound. Eliot had corresponded with Pound since 1914, and the elder poet remained a benefactor throughout Eliot's lifetime. The final form of "The Waste Land," whole sections elided by Pound's trenchant pen, is impossible to conceptualize apart from this literary collaboration. What we had imagined was Eliot's work turns out to be a curious mixture of Eliot and Pound.
"The Waste Land" is a literary achievement. But what if we were to consider it from another angle - not as a piece of high art, but instead as an entry in one of the new assessments springing up all over the country? Imagine that "The Waste Land" were submitted as part of a writing portfolio used to certify "accomplished teachers of creative writing." How should we regard this entry? The composition of the poem flies in the face of traditional notions of testing and measurement, particularly the notion of "psychometric authenticity," the assurance that a test reflects a person's own efforts and accomplishments, not those of others. Indeed, when considered from the vantage point of traditional educational theory, the status of "The Waste Land" becomes murkier the closer we look.
At least three issues immediately present themselves. First, what is the "psychometric authenticity" of this work? How can we even begin to separate Eliot's contribution from Pound's? Second, when evaluating Eliot's literary potential, how do we resolve the issue of equity? How do we compensate or, more modestly, take into account the fact that, in completing his portfolio entry, Eliot could draw on a mind like Ezra Pound's, while other poets toiled in isolation, or worse, had colleagues who wrote doggerel? Third, at what point does it become morally questionable to put Eliot's name on the entry labeled "The Waste Land"? At what point would Pound's help become "too much help"? At what point would Eliot cross some invisible line between a poem produced with aid and one produced with illegitimate aid? What if Pound not only suggested new stanzas but wrote them himself? How much rewriting would Pound have to do - a word, a line, a stanza, a canto - before we would become uneasy about calling it a solo-authored poem?
I use this hypothetical scenario as a starting point for considering the rapidly changing context of assessment. We used to have fewer things to worry about: before taking a test, we sharpened our pencils, removed everything from our desks, and kept our eyes on our own papers. But today's assessment landscape, far from the straight highway of yesterday, features craggy rocks, steep inclines, and blind curves. Portfolios, demonstrations, exhibitions, displays, and video attestations are just a sample of the dazzling array of assessments being used to evaluate teachers and their students. Before, if we talked to our neighbor, it was considered cheating. But some of today's assessments not only tolerate collaboration but invite it. If considering the status of "The Waste Land" makes our heads spin, what happens when we consider a site-based portfolio, compiled over a nine-month school year in a teacher's classroom, that draws on and benefits from the ongoing feedback of colleagues, administrators, friends, and loved ones?
We need to clear a path through these thorny questions, and our old road maps won't get us very far. Although my focus here is on teacher assessment, much of what I say applies to student assessment as well - particularly portfolios, displays, and exhibitions completed over extended periods. To understand how and why these new assessments differ from their predecessors, it is useful to begin with a brief look back at the history of modern testing.
The Test and the Individual
The history of testing is inseparable from the history of psychology more generally, and psychology has always looked to the individual as the appropriate unit of analysis. In Freud's day, the focus was on the unconscious motives of the individual; in Skinner's, on the environmental reinforcers acting on the individual; and in today's information-processing psychology, on how the individual deploys heuristics and "metacognitive" strategies to solve problems. Although the methods, research techniques, and theoretical stances have varied greatly over the past century, there has been an unwavering consistency in conceptualizing the proper unit of analysis as the individual.
Deeply embedded in traditional approaches to measurement is the notion that assessments tap constructs, or traits, that individuals carry with them from setting to setting. From a traditional perspective, when we administer a personality inventory and determine that Mary is an "extrovert," we presume that this judgment generalizes beyond the testing room. In other words, the designers of this test would hope that Mary is extroverted not only in the company of Ellen, her best friend, but in school, at the bowling alley, and in the supermarket checkout line. Although psychologists have long recognized situational variations in people's responses, they have, under the assumptions of classical test theory, viewed these variations as "error," or unwanted and unsystematic variation that accompanies an individual's "true score." According to this logic, we assume that deep within Mary lies a "quantity" of extroversion, her "true score" on the trait of extroversion. In any one context we will have a single "observation" of Mary's extroversion. If we were to compile hundreds of such observations, we could then plot the range of Mary's observed extroversion. The average of these observations would be the pure psychological quantity of extroversion that we would see if variations in context didn't interfere.
Given testing's focus on the individual, it is easy to see why collaboration threatens the validity of traditional assessment. If traits are inherent characteristics of individuals, and if individuals receive different amounts of aid in completing assessments, then how can we determine an individual's "true score" or even begin to compare the tests of two people? What can we say about Mary's "extroversion," "intelligence," or "writing ability" if the assessments she completes reflect to some unspecified degree the thinking, feeling, and reacting of others? If John receives extensive aid and Mary little, then what is the meaning of each of their individual scores? If we had a writing assessment, T. S. Eliot's entry would defy our ability to make a judgment about his innate "writing capacity." Collaboration might produce better poetry, to be sure. But if our goal were to understand the psychological ability to write poetry, we would need to appeal to different conditions. Paul Diederich's statement in a 1974 book for the National Council of Teachers of English hit the nail on the head: "As a test of writing ability, no test is as convincing to English teachers, to teachers in other departments, and to the public as actual samples of each student's writing, especially if the writing is done under test conditions in which one can be sure that each sample is the student's own unaided work" (emphasis added).3
An Alternative Approach
In recent years, there have been challenges to the traditional assumptions of classical test theory. Many of these challenges have come from scholars who have looked anew at the role of culture, social arrangement, and context in human performance. This shift can be broadly summarized as a change in focus from studying individuals in isolation, often in artificial environments, to studying individuals in concrete settings, where they can draw on features of the environment, including the people who surround them, in executing their response. Indeed, this shift is captured in the pithy response of Sarah Freeman, a leading researcher on writing assessment, to the above quotation by Diederich. "Diederich's words," wrote Freedman in the first issue of the new journal Educational Assessment, "are now dated."4
They are dated because many scholars in the writing community believe that conditions under which such writing samples are obtained - individuals isolated from one another, each composing single drafts under timed conditions - generalize only to other testing situations, not to the broad universe of situations in which individuals produce text. Good writers, according to the literature on composition, are closer to T. S. Eliot in their practice than to the student in the test hall who furiously drafts an essay, reads it over, and hands it in.5 These critics would argue that knowledge construction is a profoundly social process, and from their vantage point a focus on individual achievement actually distorts what individuals can do.
Many of these ideas derive from a theoretical position that has swept over the North American educational community like a tornado. Prior to 1960, few had heard of the Russian psychologist Lev Vygotsky. But today, due to the shifting winds in psychological and educational research, there are few who haven't heard of Vygotsky, and many hold him responsible for inciting a revolution in psychological assessment. Vygotsky's bold claim was that higher mental functioning was social in origin. For him, the principal unit of psychological analysis was not the individual child but the child in social relation - the child pointing at an object and the mother interpreting the gesture, fetching the object, and completing the interaction. This communicative act, first transacted between individuals, becomes, with time and social support, part of the child's repertoire of understanding. In Vygotsky's formal language, understanding moves from the interpsychic plane to the intrapsychic plane. In plain English, what we are first able to do with others we are eventually able to do by ourselves.
A concrete example of Vygotsky's influence is in the area of intelligence testing. The traditional I.Q. test assumes that what the child can do in isolation is the best measure of the child's intelligence. Vygotsky disagreed. He gave children traditional I.Q. tests, in which the examiner maintained a flat affect and said little, and then tested them a second time and had the examiner offer hints and clues. For a man whose writing is often impenetrable, Vygotsky is relatively clear about what happened:
Having found that the mental age of two children was, let us say, eight, we gave each of them harder problems than he could manage on his own and provided some slight assistance; the first step in a solution, a leading question, or some other form of help. We discovered that one child could, in cooperation, solve problems designed for twelve-year-olds, while the other could not go beyond problems intended for nine-year-olds. The discrepancy between a child's actual mental age and the level he reaches in solving problems with assistance indicates the zone of his proximal development. . . . Can we really say that their mental development is the same? Experience has shown that the child with the larger zone of proximal development will do much better in school. This measure gives a more helpful clue than mental age does to the dynamics of intellectual progress.6
Reasoning that humans exist in social worlds, where often the epitome of ignorance is the reluctance to seek help, Vygotsky experimented with testing and assessment situations unheard of in his day. Indeed, Vygotsky's work on interactive I.Q. testing has spawned the most revolutionary movement in intelligence testing since Binet invented the I.Q. test in 1897. Today the field of "dynamic assessment" is characterized by an impressive range of approaches both in the United States and abroad.
Further, the Vygotskian approach differs from its traditional counterparts in its attention to context, the concrete environments in which individuals respond. In contrast to traditional psychometric approaches, which seek to minimize variations in context to create uniform testing conditions, Vygotsky argued that human beings draw heavily on the specific features of their environment to structure and support mental activity. In other words, understanding how people think requires serious attention to the context in which their thought occurs.
An example of what Vygotsky meant is found in the work of Michael Cole and Kenneth Traupmann.7 These researchers sought to understand children's thinking in the overall context of the school environment as well as within variations they introduced into that environment. They focused their analysis on a private school in New York City and on one particular child, Archie, an 8-year-old whose teacher defined him as possessing "rather severe learning disabilities." The school psychologist corroborated this description based on Archie's performance on the WISC-R, an individually administered I.Q. test.
Cole and Traupmann wondered how Archie might fare in a context different from the psychologist's testing room. The researchers organized a weekly cooking club and engaged children in baking breads and cakes. Archie's partner was Ricky, whose test scores on the WISC were above average and who did well in class. From the very start, Archie took charge, reining in Ricky when his attention wandered to the girls at the next table, focusing Ricky on what ingredient to add to the mixture, and physically directing Ricky (sometimes with gentle touches on the arm, other times by thrusting the recipe in front of his face) to the task at hand. When faced with the highly complex (to an 8-year-old as well as to many adults) task of baking bread, which involves measuring, remembering, mixing, combining, and timing, it was Ricky, not Archie, who seemed disabled. As Cole and Traupmann pointed out, the statement that Archie was "learning disabled" was a statement about one of the contexts in which Archie found himself, the context of standardized testing confined largely to school settings. But if we were to look at the cooking club as a closer approximation of a real-world intellectual and social challenge, then we would be more likely to label Archie as "gifted" than "disabled."
This perspective on assessment turns the traditional notion of "psychometric authenticity" on its head. In traditional terms, the "authenticity" of a response rested on whether the person who executed it was indeed its author. According to this definition, the opposite of "authentic" would be "stolen," "plagiarized," or "misappropriated." "Authentic," however, has wholly different meanings when used by Grant Wiggins, Fred Newmann, and others associated with the "authentic assessment" movement. To be sure, these scholars would be just as concerned about plagiarized work as would members of the psychometric community. However, just because a piece of work was done entirely by the student tested does not, under their definitions, ensure authenticity.
The statement that Archie was "learning disabled" was a statement about one of the contexts in which Archie found himself. |
Authenticity, for Wiggins, is the extent to which a test, performance, or product used in an assessment bears a relationship to its real-world referent.8 For example, research on writing has underscored the important role of revision in the composition process. Skilled writers produce multiple drafts of written work, often separating the composition process from the editing process in the ongoing production of text. Therefore, a test like the New Jersey State Writing Assessment, which asks students to produce a piece of writing under conditions that allow scant time for revision, would be considered inauthentic by Wiggins, even though that writing sample would satisfy psychometric standards of authenticity. By contrast, the very features that would make writing authentic in Wiggins' view - e.g., sharing an essay with colleagues, having a friend copy-edit it - would threaten its psychometric authenticity.
Changes in Teacher Assessment
In 1988 anthropologist Jean Lave set out to understand how adults solve math problems.9 The location for her fieldwork was not an exotic country but the ordinary American supermarket. Lave followed 25 adults up and down supermarket aisles, trying to understand how they figured out whether to purchase a 24-ounce package of spaghetti for $1.87 or a 16-ounce package for $1.49. She also had people complete a paper-and-pencil test of math problems. To her surprise, she discovered that people who faltered in solving test problems could, in the context of a supermarket aisle, perform the same calculations flawlessly. These results led Lave to assert that problem solving in a natural context is profoundly different from that done between the pages of a test booklet.
A similar logic guided the design of site-based portfolios for teachers. Educational reformers reasoned that a portfolio consisting of videos of actual teaching, reflective essays, and annotated examples of student work - gathered over a year in a real classroom - would be a better measure of accomplished teaching than a score on a multiple-choice test. But in addition to seeking changes in how teachers were assessed, groups like the National Board for Professional Teaching Standards (NBPTS) had an overt social agenda: to promote an image of teachers as accomplished professionals who reflected on their practice, updated their knowledge, and drew from a variety of sources in making decisions. Central to the image of teacher as professional was the notion of teacher collaboration.
According to policy statements of the NBPTS, accomplished teachers display a "readiness to work collaboratively," participate in "collaborative efforts to improve the effectiveness of the school," and "cultivate a critical spirit in appraising . . . schooling."10 Teachers were urged to collaborate on site councils, curriculum revisions, and school and district policy boards. But what about collaborating in gathering and revising the materials that go into a site-based portfolio? This question, of course, returns us to the dilemma raised at the beginning of this article. How do education reformers balance a commitment to teacher collaboration with the need to create assessments that are valid, professionally credible, and acceptable to the public?
There is obviously a range of responses to this question, each rooted in different assumptions about teachers, teaching, and the purposes of assessment. To flesh out these issues, let me use an actual example of a portfolio exercise used by the NBPTS for certification in early adolescence/English language arts and lay out the way this exercise would look if guided by different policies on collaboration.
Option 1: Limiting Collaboration to Planning the Response
The "Post-Reading Interpretative Discussion" Exercise (PRIDE) asks the teacher to videotape a lesson in which students discuss what a particular text means. The goals of the lesson are to help students develop their abilities to interpret text, to use background knowledge, and to deal with sophisticated literature. The teacher selects a 15- to 20-minute segment of the videotape that shows students engaged in the kind of discussion that promotes critical analysis of literature. The candidate then prepares a written commentary that evaluates the discussion and explains how it promotes literacy goals in the classroom.
In an approach that restricted collaboration to the planning stage, the candidate would be free to consult with colleagues about how to conduct a post-reading interpretative discussion and even to solicit their help in setting up such a lesson. Collaboration during the planning stage might take the form of a candidate's turning to a more experienced colleague for advice about the types of questions to ask, the piece of literature to use for the lesson, general strategies for probing student understanding, and so on. But once the lesson had been videotaped, the candidate would not be allowed to discuss what happened. If candidates videotaped several lessons, they would not be allowed to consult their colleagues about which of these lessons to select for their portfolio entry.
The advantage of this approach is that it makes a gesture to safeguard the psychometric authenticity of the response. Even though the candidate may have received aid in planning the taped discussion, the selection of the final tape to include in the portfolio and the analysis of that tape are still, in large measure, the sole intellectual work of the candidate. In contrast to a policy that forbids collaboration, this setup allows the possibility of collegial exchange. The disadvantage of this arrangement is that it does so in an extremely circumscribed and artificial way.
Option 2: Limiting Collaboration to Planning and Selection
In this arrangement, collaboration is extended from the planning of materials to their selection for inclusion in the portfolio. Once candidates had assembled materials (e.g., examples of student work, videotapes of instruction) they could ask their colleagues' advice on which of these materials to include in the portfolio. Collaborators could help candidates select final materials, but they would not be allowed to engage in an analysis of what those materials mean.
While this approach extends the bounds of collaboration, it can be argued that it would be counterproductive since it is hard to imagine two colleagues having an insightful discussion about which tape to select without that discussion spilling over to issues about why a particular tape should be selected. Restricting collaboration to aspects of selection and planning makes a gesture toward collaborative work, but it is hard to see how any genuine benefits could accrue from such a limited interpretation of collaboration.
Option 3: Collaboration on Analysis but Not Production
Here the kinds of discussions and collegial exchanges that were restricted above would be allowed. Not only would help-seeking be sanctioned in the selection of materials but it would be approved for discussions about what these materials mean. Under this arrangement, we could easily imagine a candidate viewing a tape of a discussion with a colleague and engaging in a collaborative analysis of what they saw. In the best of all worlds, we would want the tape to be a stimulus for joint reflection about the goals of literacy instruction, the subtleties of classroom discussion, and the effects of teacher actions on student learning. There would be no limit to the range of topics that could be broached. Everything could be discussed until the point when the candidate sat down to write the commentary accompanying the tape. At this point, clear written guidelines would instruct candidates to compose their commentaries by themselves.
This kind of collaborative discussion certainly fits the image of professional exchange better than the previous scenarios. It allows for dialogue and idea building and brings adults together over issues of substance. But from the traditional psychometric vantage point, there are problems. If portfolios are to be scored on the keenness of the insights they contain, we are back to the problem of trying to disentangle a candidate's contribution from the "value-added" contribution of the collaborator. Equity issues crop up in trying to distinguish between entries that reflect "collaborator-rich" versus "collaborator-poor" environments.
Even if we were to bracket such concerns, the technical issues in implementing this form of collaboration are daunting. Consider the seemingly simple issue of taking notes. During these collegial discussions, can the teacher candidate take notes? Can the discussion be audiotaped? If the answer is yes, then how do we know whether the words in the final portfolio essay are the candidate's own or a direct transcription of something said by the collaborator? If we were still devoted to the notion of limiting collaboration up to and including analysis but not production, written instructions would have to specify that the candidate could not take notes during conversations or tape-record them. Otherwise, we would create a slew of problems in determining the real author of the final product.
The problems with this approach should be apparent. First, it has the same contrived quality as the other scenarios. To tell adults that they are forbidden to jot down notes during a conversation lest phrases from that conversation make their way into the candidate's final commentary hardly promotes an atmosphere of professionalism. Indeed, "lifting words" is a much smaller threat - if our goal is to find out what the candidate can do on his or her own - than lifting ideas the candidate may have received from a collaborator. Such prohibitions would contribute to a cloak-and-dagger atmosphere that thwarts collegiality.
A Different Way
All these approaches, while different on the surface, draw from a common source. All view collaboration with suspicion. Even if we embrace the most liberal assumptions offered by traditional psychometrics, we will view minimal collaboration as a threat to the integrity of a portfolio. If, on the other hand, we view teaching through the lens of Vygotsky and other sociocultural theorists, we will see collaboration in a different light. Instead of worrying that collaboration wreaks havoc on the meaning of the overall score, we may view the lack of collaboration as a more serious defect than its inclusion. But before we adopt this perspective, we had better be clear about our own assumptions.
An approach to assessment that asks teachers to sit together to analyze student work, to show one another videotapes of instruction, and to consult with one another about how to interpret evidence of student progress cannot simultaneously embrace the notion that the purest measure of teaching ability is what the solo teacher can do toiling in isolation. Instead, an assumption similar to the one guiding dynamic assessments would have to be adopted, an assumption that says, in effect, that what a teacher can do with social support is a better measure of effective teaching than what that teacher can do alone. A portfolio entry built on this assumption would look quite different from existing exercises.
For example, PRIDE, as used by the NBPTS, asks a candidate to videotape a literature discussion and write a commentary on teaching. In an exercise that granted collaboration a key role, we could imagine a feature called "Consultation with Colleagues." The instructions accompanying this feature might look like this:
To complete this exercise, you should arrange to show your videotape to at least three colleagues in your school. You should assemble these colleagues and explain to them what they are about to see. If you had a lesson plan or other materials you used as a guide for this lesson, you should distribute copies before you show the tape. Prior to showing the videotape, you should do two more things: first, you should arrange to audiotape this discussion (you will submit a copy of this tape in your portfolio), and second, you should explain to your colleagues what your goals were, what your students' prior experience with these ideas was, and what you hoped to accomplish in this lesson. After the tape is over, you should engage your colleagues in a discussion of these questions:
The point of engaging in this process would be to produce new learning and reflection. Therefore, the next section of this exercise would ask candidates to write an essay reflecting on this collegial discussion. Candidates would be asked to address what they learned as a result of this discussion and to select examples of comments they agreed and disagreed with, explaining the basis for their differing reactions.
Legitimate and Illegitimate Aid
Where would such a stance toward collaboration leave us? Would there be no limit to the forms of aid we could receive from colleagues? Would "cheating" be eliminated from our vocabulary?
Hardly. It would still be possible to cheat in this assessment - to cross the line between work that bears the influence of colleagues and work that is conceptualized and executed by colleagues. The latter form of "help" constitutes cheating and misrepresentation in most commonsense definitions and would be considered illegitimate here as well.
Recall that when T. S. Eliot met Ezra Pound, he came with manuscript in hand. It would have been a different story had Eliot's "The Waste Land" been an adaptation of something Pound had originally written. In this spirit, assessment designers would need to convey clear expectations about collaborative work. At a minimum, these would include the expectation that candidates be the originators of all written work produced for the portfolio - that they, like Eliot, should go to their collaborators with draft in hand. It is useful to distinguish among three different types of collaborative aid a candidate might receive, only one of which would be viewed approvingly:
From written materials alone, how would evaluators determine the type of aid a candidate received in assembling a portfolio? Although there are clues for determining illegitimate aid (sudden shifts in style, stilted vocabulary, unnatural tone), what about determining ineffective aid? In short, we are at a great disadvantage when we rely exclusively on written materials to make this sort of judgment.
One of the best ways to judge the kind of aid represented in the portfolio is to combine the portfolio with other forms of evidence. An "oral defense" of the portfolio, similar to the oral defense of a written thesis in universities, would be indispensable. One reason the oral defense has endured is the age-old fear that the thesis represents more of the advisor's thinking than the student's. If a doctoral candidate cannot give a reasoned justification for the logic, analysis, and argument contained in the written thesis, it is assumed that the person has not understood, in a deep and enduring way, the thinking embodied in it.
We could imagine a similar situation in the context of teacher assessment. If, in writing about literacy instruction, a teacher made reference to the work on "reciprocal teaching" by Annmarie Palincsar and Ann Brown and drew heavily on terms such as "scaffolding questions" and "scaffolding discussion" but then, in the assessment center, could not distinguish between scaffolded and direct instruction, we would have strong indication that the help received was ineffective. The candidate may have understood "scaffolded instruction" in the presence of the collaborator but had not internalized that understanding. This aid would be deemed "ineffective."
Toward a New Way of Thinking
The praises of teacher collaboration are widely sung in the literature of education reform. Teachers are urged to collaborate with their colleagues on school councils, on site-based management groups, and on multidisciplinary teams with social workers, school psychologists, and parents. All these activities are worthwhile; all should be encouraged. But none touches the heart of teaching: the direct improvement of classroom instruction. It is in this realm that a culture of isolation dominates. Most of public school teaching remains private, conducted behind closed doors, out of the earshot of other adults. In an age of $500 video cameras, it's tough to find a school without one. But in most places, this equipment is used for everything except videotaping classroom instruction so that it can be reviewed by other teachers.
The hallmark of achievement in many professions is peer review, the submission of one's work to a body of peers for scrutiny and critique. But a culture of critique is foreign to teaching. It is a bitter irony that at the same time that teachers urge students to collaborate on projects and to submit their work to peers for feedback, they, as teachers, can cite no analogous process among their own peers. How long can teachers sustain a community of learners among students when they have no learning community to nourish themselves? New assessments can help to foster learning communities among teachers. Site-based portfolios that require significant collaboration would provide the profession with a structured forum for discussing the improvement of teaching. In the long run, it may turn out that portfolios are more valuable as professional development tools than as devices for making distinctions among teachers.
To view portfolios, exhibitions, and video critiques as technological improvements in assessment surely misses the point. These innovations are not "better mousetraps" - devices that allow us to do what we've always done more efficiently. Indeed, if efficiency, cost, and ease of scoring are our prime concerns, we should leave well enough alone.
These assessments - from site-based portfolios for teachers to elaborate exhibitions and demonstrations by students - ask us to think about human capabilities in different ways. A portfolio is a measure not of what an individual can do in isolation but of what that individual can do in the midst of social community. In a corporate world that has all but abandoned the individual performance appraisal for the "work team review," schools - and perhaps remote fire observation stations - are the last holdouts. Continuing on this path will ensure that the chasm between schools and the real world continues to widen.
1. For the source of these quotations, see Nancy K. Gish, The Waste Land: A Poem of Memory and Desire (Boston: Twayne, 1988), pp. 10-13.
2. T. S. Eliot, The Waste Land: A Facsimile and Transcript of the Original Drafts Including the Annotations of Ezra Pound (New York: Harcourt Brace Jovanovich, 1971).
3. Paul Diederich, Measuring Growth in English (Urbana, Ill.: National Council of Teachers of English, 1974), p. 1.
4. Sarah W. Freeman, "Linking Large-Scale Testing and Classroom Portfolio Assessments of Student Writing," Educational Assessment, vol. 1, 1993, pp. 27-52.
5. For an overview of the literature on skilled writers, see Carl Bereiter and Marlene Scardamalia, "Literary Expertise," in K. Anders Ericsson and Jacqui Smith, eds., Toward a General Theory of Expertise (New York: Cambridge University Press, 1991).
6. Lev Vygotsky, Thought and Language (Cambridge, Mass.: MIT Press, 1986), p. 187.
7. Michael Cole and Kenneth Traupmann, "Comparative Cognitive Research: Learning from a Learning Disabled Child," in W. A. Collins, ed., Aspects of the Development of Competence: The Minnesota Symposium on Child Psychology, vol. 14 (Hillsdale, N.J.: Erlbaum, 1981), pp. 125-53.
8. Grant Wiggins, "The Case of Authentic Assessment," ERIC ED 328 611, 1990.
9. See Jean Lave, Cognition in Practice: Mind, Mathematics, and Culture in Everyday Life (New York: Cambridge University Press, 1988).
10. National Board for Professional Teaching Standards, "Creating a Profession of Teaching: The Role of National Certification," American Educator, Summer 1990, p. 42.