This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 238 Research in the Teaching of English Volume 40 November 2005 At Last
The Focus on Form vs. Content in Teaching Writing George Hillocks, Jr.
University of Chicago For many years the teaching of writing has focused almost exclusively and to the point of obsession on teaching the forms of writing—the parts of paragraphs, the parts of essays, the structure of sentences, the elements of style, and so forth. Except for the fact that the models of writing used to illustrate such forms have some level of substance in and of themselves, teachers of and textbooks on writing have treated substance as though it were of little or no importance. The underlying assumption is that writing can be taught with little or, at best, sporadic reference to content: that once students learn the various forms, they are then prepared to write real prose—for instance, expository paragraphs or evaluative themes. The problem, of course, is that writers do not decide to write an expository paragraph or an evaluative theme. If they decide to write prose at all, they decide to write about specific subject matter. Moreover, knowledge of form does not translate into the strategies and skills necessary to wrest from the subject matter the ideas that make up a piece of writing. Indeed, we have research suggesting that other foci of instruction are more powerful than instruction on form, yet there are forces that appear to be influential in maintaining form as central to teaching writing. In this essay, I review some of the research indicating the obsession with form over the past fifty years or more; discuss my own research, which tells us our focus should be elsewhere; discuss those forces that keep us from heeding this broader understanding about effective writing instruction; and, finally, offer some predictions about the possibility of shifting policy and practice away from its present obsessions. Obsession with Form
Lynch and Evans (1963) published a study of high school English textbooks that was famous in the ’60s. The authors examined textbooks for grades 9 to 12 by every major publisher at the time, with copyright dates from 1949 to 1961. Lynch and
238 Research in the Teaching of English Volume 40, Number 2, November 2005 Copyright © 2005 by the National Council of Teachers of English. All rights reserved. f238-248_Nov05RTE 238 10/15/05, 11:01 AM HILLOCKS At Last 239 Evans devote only about 60 pages of their 500-page book to the textbooks’ concerns with rhetoric and composition because, in the case of the grammar and composition books, the space devoted to grammar, mechanics, and usage far exceeds the pages devoted to composition and rhetoric. For example, the grade 12 volume that we know as Warriner (1977) includes 351 pages on grammar, usage, and mechanics but only 133 pages on composition and rhetoric. According to Lynch and Evans, the material presented on composition and rhetoric is “virtually the same in a majority of texts” (p. 312), not only within a series but also across different series by different publishers. In their summary of paragraph instruction from a text by American Book Co. indicating that this instruction is representative of paragraph instruction in all series, they present a number of rules for writing paragraphs and topic sentences. The rules are followed by an “exercise in identifying the topic sentence in each of three brief paragraphs,” an exercise in identifying “topic sentences in five paragraphs” from the student’s history or science book, exercises in “selecting and writing appropriate topic sentences,” and a list of rules about what constitutes a good paragraph (p. 313).We know from Richard Braddock’s research (1974) that such rules provide an inappropriate analysis of the reality of paragraphs. In the corpus of 25 essays that Braddock examined, only 37% of the 889 paragraphs contained the kind of sentences that the textbooks present as topic sentences. (See p. 297 of Braddock.) The textbooks assume that students will have the information to develop such paragraphs at their disposal. Moreover, Lynch and Evans complain that the textbooks ignore a very “basic principle of composition . . . : the necessity of idea as a direct force in the selection and organization of developmental detail and in the shaping of the whole composition whether it is paragraph, essay, or research paper” (p. 329). They continue,
The textbooks exhort students to use the devices of effective composition in word and sentence, and they set exercises requiring students to use these devices. Thus composition becomes an exercise in using devices for effectiveness. But, in truth, “real” writers do not write in order to use the devices of effectiveness; they write to express ideas and in the process of expressing these ideas they come upon the need for specific words, all kinds of figures of speech, for sentence patterns that accurately reflect meaning and emphasis, etc. (p. 333). Some two decades later, Applebee (1981, 1984) studied writing in American secondary schools across subject matters: English, foreign language, science, math, social science, business education, and special education. He and his colleagues observed classrooms in two Midwestern high schools; conducted case studies of students in these classrooms; and also surveyed a large sample of teachers from 196 schools. Although writing was a major presence in all subject matters, taking up f238-248_Nov05RTE 239 10/15/05, 11:01 AM 240 Research in the Teaching of English Volume 40 November 2005 “an average of 44% of the observed lesson time” (1981, p. 30), the researchers found that students were spending only about 3% of their school time—in class or for homework—on writing of paragraph length or longer. On the other hand, students were engaged in a variety of related activities that involved writing but not composing: fill-in-the-blank exercises, worksheets requiring only short responses, translation from one language to another, and the like. The teaching of writing in most cases studied by Applebee appeared to be little more than the making of assignments. “In the observational studies, the amount of time devoted to prewriting activities amounted to just over three minutes. That included everything from the time the teacher started introducing the topic until the first student began to write” (1981, p. 74). When writing was taught more explicitly, it was through model pieces of writing used regularly in 32.9% of the English classes as a means of “introducing new forms of writing” (p. 78). However, according to Applebee, while “the major vehicle for writing instruction, in all subject matter areas, was the teacher’s comments [on] and corrections of completed work, errors in writing mechanics were the most common focus of these responses; comments concerned with the ideas the student was expressing were the least frequently reported” (pp. 90-91). Unfortunately for these results, we know from research like that of Sperling and Freedman (1987) that even the most promising students misunderstand and/or misconstrue the written comments that teachers write on their papers even when they are accompanied by conferences, peer group response, and whole class discussion of responses. In short, a quarter of a century ago, when Applebee conducted the study, writing was widely taught in very superficial ways. Teachers appeared to assume that very general knowledge of writing would suffice for most purposes. There is no indication that teachers taught their students any strategies for developing their ideas. The most concrete instruction appeared to be based on the use of models of the forms that students were expected to use. Now, more than two decades after Applebee conducted this study, while we have seen some changes, teachers of writing in the schools still appear to rely heavily on teaching the forms and devices of writing while neglecting how to work with the content. As part of my own study of state assessments (Hillocks, 2002), I found much the same thing. I interviewed teachers at every tested level and at levels not tested, as well as administrators and supervisors in six school districts in each of five states. Most teachers and supervisors described practice in considerable detail, and it appears that they were quite honest and straightforward about their practices. A major question of the interview was, “How do you prepare your students for the writing assessments and for other kinds of writing they have to do?” While Applebee indicates that only 3% of the time spent on writing was devoted to work on pieces of a paragraph or more, nearly all teachers I interviewed f238-248_Nov05RTE 240 10/15/05, 11:01 AM HILLOCKS At Last 241 talked only about the writing of multi-paragraph compositions, even at the elementary level. But in many districts, such writing focused on five-paragraph themes (5P), thus imposing a form as well as a limit of sorts. Even so, teachers indicated that they spend far more time in preparation for writing than the three minutes for the teachers in the Applebee study. Indeed, several teachers described strings of activities that precede writing. Yet, despite these and other apparent changes from 25 years ago, we found form to dominate instruction. We categorized all teacher interviews as to the teachers’ primary and secondary instructional foci. In the five states examined, an average of over 71% of teachers interviewed focused on model pieces of writing as the primary means of teaching writing, and for nearly all of the remainder, the use of models was a secondary focus. These teachers “see their task as teaching knowledge about standard forms of writing that have been in the American curriculum for 100 years. They tend not to deal with strategies for generating content beyond simple prewriting. They do not teach strategies for generating the specifics” (Hillocks, 2002, p. 190). In sum, all three of these studies across the fifty years found the primary focus of instruction to be on form from the presentation of model pieces of writing to the teacher comments, which are notoriously focused on form at the level of mechanics. Where the Instructional Focus Should Be
Our research provides considerable insight into the effectiveness of focusing instruction on writing forms. In 1986, I reviewed nearly 500 quasi-experimental studies of writing instruction conducted between 1963 and 1983, selecting those that met criteria for strong research design principles (Campbell & Stanley, 1966) in order to conduct a meta-analysis or research synthesis, which permits the comparison of results across studies (Cooper & Hedges, 1994). We coded the salient features of all experimental and control treatments along several dimensions, including focus of instruction, which refers to the dominant content of instruction, for example, sentence combining, grammar, or the study of model pieces of writing (Hillocks, 1986). For each study in each focus, I computed an effect size (which is the gain for the experimental groups minus the gain for the control groups, divided by the pooled standard deviation of the post test scores for all groups). Six foci were the subject of five or more studies each: grammar, study of model pieces of writing, sentence combining, the use of scales for judging and revising writing, inquiry, and free writing. Inquiry was operationally defined as focusing on sets of data and “activities designed to help students develop skills or strategies for dealing with the data in order to say or write something about it” (Hillocks, 1986, p. 211). f238-248_Nov05RTE 241 10/15/05, 11:01 AM 242 Research in the Teaching of English Volume 40 November 2005 The treatments with the largest gains were sentence combining, the use of scales, and inquiry, all of which, importantly, focus on teaching procedural knowledge, knowledge of how to do things. This contrasts with both grammar and models, which focus on learning what I call declarative knowledge, knowledge of what, that is, knowledge that teachers may hope will result in procedures, even though teachers do not engage students in the procedures as a means of teaching what is to be learned. Further, grammar, sentence combining, and scales all deal with form, while inquiry and free writing may be said to focus primarily on the content of writing. The effect size of inquiry is significantly greater than any other focus. But the difference between inquiry and free writing is highly significant (p < .0001). What accounts for that difference? Both focus on content. We can borrow from Vygotskian theory (1978) to help explain the difference. Freewriting, by definition, is unencumbered by instruction. It is in the zone of what the student can do without instruction. Inquiry, when it is thoughtfully organized, is in students’ zone of proximal development. It challenges students to do more than they can on their own but provides the scaffolding to allow them to push beyond what they can f238-248_Nov05RTE 242 10/15/05, 11:01 AM HILLOCKS At Last 243 already do. Most importantly, it gives students the power to work with ideas, far more important than encouraging them to toss random thoughts into a foreordained framework such as the 5P. Why Form Over Content?
We are left with the question of why, despite evidence to the contrary, form has been so overwhelming an instructional focus. There are probably dozens of reasons, but I would like to examine three: the assumption that effective writing requires knowledge of only a few basic principles; the pressure on schools and teachers engendered in an age of testing and accountability; and the methods of teaching writing promulgated by state writing exams. Assuming General Principles to be Sufficient for Effective Writing
Smagorinsky and Smith (1992) examined the kinds of knowledge thought to be involved in the teaching of writing. The advocates of general knowledge hold that a few general strategies suffice for any sort of writing. Smagorinsky and Smith cite Warriner and Griffith’s (1977) English Grammar and Composition, which states that “no matter what you are writing about, the basic steps involved in writing are almost always the same” (p. 282). According to Smagorinsky and Smith, the steps outlined in the book “include selecting and limiting a topic, assembling materials, organizing and outlining ideas, writing a draft that follows a particular form (usually including five paragraphs), revising, and preparing a final draft” (p. 282). Smagorinsky and Smith also make the case that “faith in the sufficiency of general knowledge of text structure is rare among the professoriate and has been replaced by a belief in general procedural knowledge that has begun to transform teaching and textbooks,” namely, “general procedures for producing texts that rely on nonlinear thinking such as brainstorming, clustering, and free writing” (p. 283). They claim that the idea of “non-linear thinking” has been most earnestly advanced as general knowledge of the writing process by Murray (1980, 1987) and Elbow (1973, 1981). They cite Murray (1987) as referring to the process approach to writing “as consisting of five steps: collecting, focusing, ordering, developing, and clarifying. Writers can apply this general process to any composing problem and couple the five steps with general strategies such as free writing, brainstorming, and mapping” (p. 283). Despite such guidelines invoking general knowledge, Smagorinsky and Smith indicate that the kind of knowledge involved in inquiry is not general, but rather task specific. The kind of knowledge required for effective narrative writing is quite different from that required for effective argument writing, for example. The prevalent belief in the efficacy of general knowledge for any writing task precludes the use of inquiry. f238-248_Nov05RTE 243 10/15/05, 11:01 AM 244 Research in the Teaching of English Volume 40 November 2005 The Pressure for Accountability
For more than the last two decades, the demand for accountability of schools and teachers has increased. Now teachers in nearly all states are faced with state tests of composition at three or more grade levels. Recently the SAT has added a composition section to the banks of multiple-choice items. The pressure on pushing scores up, even in affluent communities, has become paramount. In impoverished communities, the problems are enormous. Scores for individual schools are published in local papers along with rankings in the state. Individual teachers have told me that they knew that they would be identified as the teacher responsible for poor results especially in small schools where they may be the only teacher of writing at the grade level tested. While some states have been identified as high-stakes and others as low-stakes, the pressure on administrators and teachers for high test scores appears not to vary greatly between the two. Texas, a high-stakes-test state, can close failing schools, but so can cities like Chicago even though the Illinois tests are low-stakes. In my experience, administrators badgered by low test scores turn to any quick fix they can find. Too often, the quick fix in writing is some formula or other. Methods of Teaching Writing Promoted by States’ Exams
By and large, the rubrics that states have adopted to evaluate writing and the instructional materials they offer as suggestions for better teaching indicate that state officials support the idea that general knowledge suffices for teaching writing. For example, the criteria indicating what counts as a successful piece of writing are vague. For persuasive writing, most states call for elaboration and support; however, the state rubrics present no criteria for what counts as solid support. A careful analysis of the benchmark papers in Texas and Illinois indicates that, in reality, no evidence is necessary to achieve the highest scores. In fact, few of the benchmark papers at the highest levels contain any evidence at all. In Texas, for example, teachers told us in many interviews, that elaboration was the single most important criterion for passing the test at all grade levels (Hillocks, 2002). One supervisor told me that she commonly walked about in a classroom while students were writing during the exam and encouraged students to write more, an illegal practice in Texas. The most interesting thing is that she did not call for high quality additions, just additions. The Texas scoring guides devote the most space to discussions of elaboration. They tell us:
An essential part of successful writing is the effective use of support or elaboration, which requires the student to develop clearly, logically, and completely those ideas that lead the reader toward an understanding of the writer’s purpose in the response. The degree to which support or elaboration is achieved, therefore, is dependent not only on f238-248_Nov05RTE 244 10/15/05, 11:01 AM HILLOCKS At Last 245 the student’s ability to generate ideas, but also upon the extent to which the student provides the reader with a detailed explanation of those ideas . . . The more thoroughly and precisely each idea is developed with specific details, the stronger and more complete the support or elaboration will likely be. (Texas Education Agency, 1993, p. 3) While we can certainly agree with this statement, we need to see what actually counts for the “detailed explanation of . . . ideas.” The guide presents examples of reasons given at different levels of elaboration for responses to a prompt asking students to write a letter to an official concerning a proposal that all teens be required to do community service for a year following graduation from high school. The guide provides a series of reasons from “bare reason” to what the guide calls “fully elaborated.” The following is what the Texas guide calls the “fully elaborated reason”:
I am against the idea of making all high school graduates serve the community since it will disrupt and interfere with our plans to go to college or work. Students who plan to go to college will forget most of what they have learned if they spend a year away from school. Their chances to achieve a high GPA will be hamstrung. As far as plans to work go, some graduates, like me, plan to work for necessities. I have to pay $350.00 a month car and insurance payments on my Nissan 280 ZX. Making your payments on your first loan is important so you can establish good credit. Also, some people have to help their parents with rent and groceries, and medical bills (Texas Education Agency, 1993, p. G6). In an argument of policy, that something should or should not be done, we may have a major claim about what to do and several reasons for doing it or not doing it. The reasons may be called sub-claims which themselves require some kind of support. When we provide support for the sub-claims in an argument, we usually provide evidence, hard data about the “facts of a case” that are purported to be true. Sometimes such an argument also includes statements that serve to explain why data support a claim. These are general rules that parties in a discussion can accept as true or at least are willing to agree to provisionally. I recently purchased a used Avalon. One indirect piece of evidence that it was in good condition was that the previous owner had been a 73-year-old woman. What allowed me to adopt this evidence was my belief that older women tend to drive carefully and take good care of their cars. Such a statement is called a warrant because it explains why data are appropriate in support of a claim. That is, it warrants the data as support for the claim. If we use this Toulmin (1958) framework of major claims, sub-claims, evidence, and warrants, we can analyze what Texas calls a “fully elaborated reason” as follows: f238-248_Nov05RTE 245 10/15/05, 11:01 AM 246 Research in the Teaching of English Volume 40 November 2005 Major Claim: I am against the idea of making all high school graduates serve the community since it will disrupt and interfere with our plans to go to college or work. Sub-claim 1: Students who plan to go to college will forget most of what they have learned if they spend a year away from school. (Therefore, the writer implies,) Sub-claim 2: Their chances to achieve a high GPA will be hamstrung. Sub-claim 3: As far as plans to work go, some graduates, like me, plan to work for necessities. Evidence: I have to pay $350.00 a month car and insurance payments on my Nissan 280 ZX. Sub-claim 4: Making your payments on your first loan is important so you can establish good credit. (If this statement tied the evidence to sub-claim 3, about having to pay for necessities, it would function as a warrant. However, sub-claim 3 does not deal with establishing credit.) Sub-claim 5: Also, some people have to help their parents with rent and groceries, and medical bills. This analysis indicates that the sub-claims remain unsupported, which in turn leaves the major claim open to serious question. The strongest part of the argument is sub-claim 3, which is followed by evidence of a sort. (It is, of course, problematic that the Nissan 280 ZX is a necessity.) Sub-claims 1, 2, and 5 remain totally without support, and sub-claim 4 is irrelevant. What is disturbing about this example, provided by the state of Texas so that teachers will understand what is meant by elaboration, is that it involves no real evidence and leaves the major claim nearly totally unsupported. The example tells Texas teachers that any sort of elaboration will do. Preparation for the writing tests in Texas and many other states is largely a matter of learning a form for organization and filling the form with stuffing. The same appears to be true in other states that use comparable writing-on-demand prompts and rubrics. In New Jersey, Florida, Georgia, and Illinois we find similar unsubstantiated writing receiving high scores. It may be that the teaching of writing in the schools will continue to focus on form for the future. Most teachers interviewed for the study of testing (Hillocks, 2002) felt that the state provided an adequate rationale and framework for the teaching of writing. Testing conditions, prompts, and rubrics in most states, with the exception of Kentucky, almost demand the focus on form and stuffing that is so prevalent. New teachers know so little about teaching writing (Kennedy, 1998; Smagorinsky & Whiting, 1995) that they are unable to resist the pressure to teach the prevailing formulas and forms in the schools (Johnson, Smagorinsky, Thompson, & Fry, 2003). f238-248_Nov05RTE 246 10/15/05, 11:01 AM HILLOCKS At Last 247 Conclusion
There may be three possibilities for changing the obsession with form. One involves thorough grounding of new teachers in a different kind of teaching of writing than most universities and colleges now offer. But most teachers of writing and English education appear to be staunch advocates of the principle that a little general knowledge of writing suffices for most purposes. I see no hope of change coming from that direction in the foreseeable future. A second possibility for change lies with teachers in the field already. Change there is even more unlikely. A third possibility lies with the state tests. If states changed the format of their testing to include data (passages, pictures, information, etc.) for students to write about as do the New York Regents Examinations, then perhaps teachers would see a need to help their students learn to deal with content in creative and critical ways. What are the chances of engineering changes in state-established testing procedures? Perhaps, perhaps, perhaps. Making the changes required to improve the learning of writing will involve a tremendous effort on the part of many people. All we can do is to make changes here and there with teachers who are interested. Not much, but when one considers the odds, every change is a victory. REFERENCES
APPLEBEE, A. N. (1981). Writing in the secondary school: English in the content areas. (Research Report, No. 21). Urbana, IL: National Council of Teachers of English. APPLEBEE, A. N. (1984). Contexts for learning to write. Norwood, NJ: Ablex. BRADDOCK, R. (1974). The frequency and placement of topic sentences in expository prose. Research in the Teaching of English, 8, 287-302. CAMPBELL, D. T., & STANLEY, J. C. (1966). Experimental and quasi-experimental designs for research. Boston: Houghton Mifflin. COOPER, H., & HEDGES, L. V. (EDS.) (1994). The handbook of research synthesis. Washington, D.C.: Brookings Institution. ELBOW, P. (1973). Writing without teachers. New York: Oxford University Press. ELBOW, P. (1981).Writing with power: Techniques for mastering the writing process. New York: Oxford University Press. HILLOCKS, G., JR. (1986). Research on written composition: New directions for teaching. Urbana, IL: National Conference on Research in English/ERIC Clearinghouse on Reading and Communications Skills. HILLOCKS, G., JR. (2002). The testing trap: How state assessments of writing control learning. New York: Teachers College Press. JOHNSON, T. S., SMAGORINSKY, P., THOMPSON, L., & FRY, P. (2003). Learning to teach the fiveparagraph theme. Research in the Teaching of English, 38, 2, 136-176. KENNEDY, M. M. (1998). Learning to teach writing: Does teacher education make a difference? New York: Teachers College Press. LYNCH, J. J., & EVANS, B. (1963). High school English textbooks: A critical examination. Boston: Little, Brown. MURRAY, D. (1980). Writing as process. In T.R. Donovan & V. W. McClelland (Eds.), Eight approaches to teaching composition (pp. 3-20). Urbana, IL: National Council of Teachers of English. MURRAY, D. (1987). Write to learn. New York: Holt, Rinehart, and Winston. f238-248_Nov05RTE 247 10/15/05, 11:01 AM 248 Research in the Teaching of English Volume 40 November 2005 SMAGORINSKY, P., & SMITH, M. W. (1992). The nature of knowledge in composition and literary understanding: The question of specificity. Review of Educational Research, 62, 279-305. SMAGORINSKY, P., &WHITING, M. E. (1995). How English teachers get taught: Methods of teaching the methods class. Urbana, IL: National Council of Teachers of English. SPERLING, M., & FREEDMAN, S. W. (1987). A good girl writes like a good girl: Written responses to student writing. Written Communication, 9, 343-369. TEXAS EDUCATION AGENCY (1993). Exit level 1993-1994 writing collection scoring guide for persuasive writing. Austin, TX: Author. TOULMIN, S. (1958). The uses of argument. New York: Cambridge University Press. VYGOTSKY, L. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. WARRINER, J. E., & GRIFFITH, F. (1977). English grammar and composition (heritage edition): Complete course. New York: Harcourt Brace Jovanovich. Call for Submissions
The NCTE-LEA Research Series in Literacy and Composition publishes groundbreaking work on literacy, composition, and the intersections between the two. The series is coedited by David Bloome (The Ohio State University), Malea Powell (Michigan State University), Duane Roen (Arizona State University-East), and Arlette Ingram Willis (University of Illinois at Urbana-Champaign). Volumes in this series are primarily original, authored or co-authored works that are theoretically significant and hold broad relevance to literacy studies, composition, and rhetoric. The series may also include occasional landmark compendiums of research. The scope of the series includes qualitative and quantitative methodologies; a range of perspectives and approaches (e.g., sociocultural, cognitive, feminist, psycholinguistic, pedagogical, critical, historical); and research on diverse populations, contexts (e.g., classrooms, school systems, families, communities), and forms of literacy (e.g., print, electronic, popular media). The intended audience includes scholars, professionals, and students in a range of fields in English studies, including literacy education, language arts, composition, and rhetoric. For detailed submission guidelines, please visit the NCTE Web site at: http:// www.ncte.org/pubs/books/series/107588.htm or send a stamped, self-addressed envelope to NCTE-LEA Research Series, Books Program Assistant, 1111 W. Kenyon Road, Urbana, IL 61801-1096. f238-248_Nov05RTE 248 10/15/05, 11:01 AM ...
View Full Document
This note was uploaded on 03/29/2011 for the course ED 101 taught by Professor Propas during the Spring '11 term at Grand Valley State University.
- Spring '11