[Also co-authored by Reinhold Hill, and also from the early 2000s]
Introduction
Any Writing Program Administrator occasionally has the frustrating experience of failing to get administrators, colleagues, parents, and even students to understand the bases of our decisions–why classes must remain small, why instructors need training in rhetoric and composition, and so on. This kind of experience is frustrating because we often find ourselves talking to someone with different assumptions about teaching, writing, and research. For such an audience, position statements are often not helpful, as our interlocutors do not even know the organizations whose position statements we’re likely to cite.
This is not to say that such discussions are necessarily an impasse, nor that the different assumptions interlocutors have are incommensurable. It is simply that, often being from different disciplines, we all bring different assumptions that seem transparently obvious to each of us. We may know from experience that one gets better writing from students if they are required to revise, but an administrator with a different disciplinary background may be sincerely concerned that we are not assessing our classes and students on the extent to which students have retained the information we have given them in lectures and readings. For many people, that is learning. As far as they are concerned, if we are not lecturing and assigning reading, then we are not teaching; if we are not testing our students, then we are not assessing students objectively.
In addition, the articles and books that we are likely to cite to explain our practices look very strange to some people–it’s just argument, a colleague once complained. We can rely on argument because we teach argument, and we are comfortable assessing arguments. We can rely on anecdote and personal experience more than people in many fields because we share an experience–the teaching of writing. Thus, if an author narrates a specific incident, we are likely to find it a reasonable form of proof, if the incident is typical of our own experience. In some other fields, however, quantified, empirical evidence is the only credible sort of proof, or an assertion must be supported by a large number of studies (regardless of how problematic any individual study might be). This is particularly an issue with class size, as minor change in enrollment (from an administrator’s perspective) are strongly resisted by Writing Program Administrators. Our intention in this article is to try to help Writing Program Administrators argue for responsible and ethical class sizes in writing courses.
There are few topics about which Writing Program Administrators and upper administrators are likely to disagree quite so unproductively as class size. While Writing Program Administrators typically argue for keeping first year writing courses as small as possible, upper administrators are often focussed on the considerable savings that could be effected by even a small change in enrollment. WPAs can cite position statements and recommendations from NCTE and ADE, but upper administrators cite such passages as the following from Pascarella and Terenzini who summarize the “substantial amount of research over the last sixty years” on class size in college teaching:
The consensus of these reviews–and of our own synthesis of the existing evidence–is that class size is not a particularly important factor when the goal of instruction is the acquisition of of subject matter knowledge and academic skills. (87).
With the backing of such an authority, upper administrators are likely to be mystified at WPA’s resistance to first year writing classes of twenty-five to thirty.
This is not to say that WPAs have no research on the side of smaller classes. Despite what Pascarella and Terenzini say, there is considerable research which identifies benefits in smaller classes. The meta-analysis of Glass and Smith (not mentioned by Pascarella and Terenzini) concludes that reduced class size is beneficial at all grade levels; Slavin found small positive short-term benefit; and several studies found benefit if (and only if) teachers engaged in teaching strategies that took advantage of the smaller size (Chatman, Tomlinson). On the other hand, there is at least one study too recent to be cited by Pascarella and Terenzini that concludes no demonstrable benefit to reducing class size (e.g., David Williams). Thus, it may seem to be a case of warring research.
On the contrary, we will argue that the apparently disparate results of research can be explained, in a comment made by Pascarella and Terenzini. After the passage quoted above, they say “It is probably the case, however, that smaller classes are somewhat more effective than larger ones when the goals of instruction are motivational, attitudinal, or higher-level cognitive processes” (87).
There are two points which we wish to make about Pascarella and Terenzini’s negative conclusion regarding class size. First, it is striking how dated the research is–although Pascarella and Terenzini’s book came out in 1991, the most recent study they cite is 1985. Of the eighteen studies they mention, three are from the twenties, one from 1945, two from the fifties, two from the sixties, seven from the seventies, and three from the eighties. This is particularly important for the teaching of writing, as there was a major reversal in the sixties in pedagogy, returning from the lecture-based presentation of models which students were expected to imitate to the classical method which put greater emphasis on the process of inventing and arranging an effective argument.
This issue of teaching model is crucial. The impact that varying class size has on the outcome in terms of student writing depends heavily on the goal and method of the writing courses in question. If the courses are lecture courses, in which only the teacher is expected to read the students’ writing, then the only limit on class size comes from the amount of time one expects the teacher to spend grading. While that is not a model we endorse (and we will discuss the reasons below), it can still be the basis of a useful discussion.
At a “Research I” institution, faculty members are usually assessed on the assumption that they spend forty per cent of their time teaching two courses, or, one day out of a five day work week (eight hours). At schools with more teaching responsibilities, the math works out in similar ways (with a fairly ugly exception for universities with Research I publishing expectations and a three or four course teaching load). Graduate students are usually assumed to have teaching responsibilities that account for half of their half-time appointment, or ten hours per course. With three hours per week in the classroom, and three hours of office hours, graduate students instructors are left with four hours per week of grading and course preparation.
One reason that administrators and WPAs often disagree about the amount of work involved in teaching writing courses is that administrators’ experience is with what Hillocks calls the “presentational” mode of teaching. The first year is hellish, but then the instructor has prepared the presentations, and future years involve tinkering with prepared lectures. Hence, course preparation is presumed to be minimal. But, of course, most WPAs are not imagining instructors’ spending class time lecturing because the presentational mode has been demonstrated, conclusively, to be the least effective method of teaching writing.
Still and all, if one assumes that a course is supposed to take 150 hours of an instructor’s time over the course of a semester (not including pre-semester course preparation), and 45 hours of that time is spend in class, and another 45 hours is spent in office hours, there are 60 hours left for individual conferences, grading, and course preparation. If there are twenty students per class, then meeting twice with each student for a half hour conference uses up 20 hours. Even assuming an efficient teacher who is dusting off lecture notes for course preparation, one should expect an hour per week of course preparation (15), leaving 25 hours for grading. Advocates of minimal marking (a problematic issue to be discussed below) describe a process that takes only twenty minutes per paper. Obviously, then, the amount of time an instructor spends on grading depends upon the number of papers, but a course with only three papers would use up almost all of the time left. Since most programs require more than three papers (and most instructors spend more than twenty minutes per paper), more than twenty students per course puts instructors into unethical working conditions.[1]
But, as we said, the deeper issue concerns just what happens in a writing class. The issue is whether one sees writing instruction as the inculcation of subject matter knowledge or as the development of higher-level cognitive processes. To the extent that it is the latter, classes should be small; to the extent that it is the former, class size is limited only by instructor workload Or, in other words, what do we teach when we teach college writing?
Interestingly enough, this is one of those questions that is not a question for people outside of the field. It seems obvious enough to people unfamiliar with research in Linguistics, Rhetoric, and English Education who tend to give what appears a straightforward answer: we teach the rules of good writing. Behind that apparent consensus is an interesting disagreement. For some people, the “rules of good writing” describe formal characteristics in writing that all educated readers acknowledge is high quality (e.g., the thesis in the first paragraph, an interest-catching first sentence). For others, those rules describe procedures that all good writers follow when writing (e.g., keep notes on three by five cards, write a formal outline with at least two sub-points). As qualitative and quantitative research has shown, however, both of those perceptions regarding the rules of good writing–regardless of how widespread they are–are false.
In the first place, there is less consensus about what constitutes “good” writing than many people think. Writers have fallen in and out of fashion, so that there is not any author who has not had his or her detractors–critical reception of Henry David Thoreau’s Walden was so hostile that it was nearly turned into pulp; Addison and Steele, always included in composition textbooks until the 1960s, are now considered nearly unreadable; even Shakespeare has been severely criticized for his mixed metaphors, complicated language, and drops into purple prose. As research in reader response criticism demonstrated as long ago as the early part of this century (see I.A. Richards), students (and readers) do not immediately recognize the merits of canonical literature; there is considerable disagreement as to just what the best writing is, so that the “canon” of accepted great writing has constantly been in flux (see Ohmann, Fish, Graff).
To a large degree, this disagreement is disciplinary; that is, different disciplines have different requirements for writing. This divergence is most obvious in regard to format–such as citation methods and order of elements. It is equally present and more important in regard to style: in the experimental and social sciences, for instance, “good” writing uses passive voice, nominalization, long clusters of noun phrases, and various other qualities which are considered “bad” writing in journalism, literature, and various humanistic disciplines. Even the notion of what constitutes error varies–social science writing is rife with what most usage handbooks identify as mixed metaphors, predication errors, reference errors, non parallel structure, split infinitives, dangling modifiers, and agreement errors. Lab reports, resumes, and much business writing permit, if not require, fragment sentences. Meanwhile, people from some disciplines recoil at the use of first person in ethnographic writing, literary criticism, some journalism, and other humanistic courses.
Disciplines also disagree as to what constitutes good evidence (for more on this issue, see Miller, Bazerman). Some disciplines accept personal observation (e.g., cultural anthropology), while some do not (e.g., economics). There are similarly profound disagreements regarding the validity of textual analysis, quantitative experimentation, qualitative research, interviews, argument from authority, and so on. There is a tendency for people to be so convinced of the epistemological superiority of their form of research that, when confronted with the fact of differing opinions on what constitutes good writing, they dismiss the standards of some other discipline (thus, for instance, Richard Lanham’s popular textbook Revising Prose condemns all use of the passive voice, and Joseph Williams’ even more popular Style: Ten Lessons in Clarity and Grace prohibits clusters of nouns). Our goal is not to take a side in the issue of which discipline promotes the best writing, but to insist upon the important point out that there is disagreement. Thus, a writing course cannot teach “the” rules of good writing that will be accepted in all disciplines because no such rules exist (unless the rules are extremely abstract, as discussed below).
In the second place, as several studies have shown, the ‘”rules” of good writing which we give students for student writing do not describe published writing. For instance, students are generally told to end their introductions with their ‘thesis statements,” to begin each paragraph with their topic sentences (which is assumed to be the main claim of the paragraph), and to focus on the use of correct grammar. Published writing, however, does not have those qualities. Thesis statements are usually in conclusions (Trail), and introductions most often end with a clear statement of the problem (Swale), or what classical rhetoricians called the “hypothesis” (meaning a statement that points toward the thesis). Textbook advice regarding topic sentences is simply false (Braddock), and, readers are much more oblivious about errors in published writing than much writing instruction would suggest (Williams). In fact, error may not have quite the role that many teachers think–while college instructors say that correctness is an important quality of good writing (Hairston), studies in which they rank actual papers shows a privileging of what compositionists call “higher order concerns”–appropriateness to assignment, quality of reasoning, and organization over format and correctness (see Huot’s review of research on this issue).
Finally, several meta-analyses of research conclude that teaching writing as rules has a harmful effect on student writing (see especially Knoblauch and Brannon, Hillocks, Rose). The common sense assumption is that students prone to writing blocks lack the knowledge of rules for writing that effective writers have; on the contrary, students prone to writing block may know too many rules. In contrast to more fluid writers, who tend to focus on what is called “the rhetorical situation” (explained below), student writers prone to writing blocks focus on rules they have been told (Flower and Hayes, Rose). Students taught these rules of writing try to produce an error-free first draft which they minimally revise (Emig, Sommers). Effective and accomplished writers, in contrast, have rich and recursive writing processes that depends heavily upon revision (Emig, Flower and Hayes, Berkenkotter, Faigley and Witte).
For many people unaware of research in linguistics and English education, the assumption is that the “rules” of good writing are the rules regarding usage (usually described as “grammar rules,” which is itself an instance of an error in usage). It is assumed that there is agreement regarding these rules, and they are to be found in any usage handbook. Further, it is assumed that one can improve students’ “grammar” (another interesting usage error–what people mean is “reduce usage error” or “improve correctness”) by getting them to memorize those universally agreed-upon usage rules. These assumptions are wrong in almost every way.
Research in linguistics demonstrates that language has considerable variation over time and region. To put it simply, at any given moment, there are numerous dialects within a language which are each “correct” within their community of discourse (e.g., “impact” for “influence,” “thinking outside the box”). Some dialects are more privileged than others, and the uninformed often assume that facility with the more privileged dialect signifies greater intelligence; this is patently false (Chomsky, Labov and Smitherman, Baron). All dialects have a grammar, so students (and colleagues) who use a different dialect are not ignorant of “grammar;” they know the grammar of a dialect not considered appropriate in academic discourse, the dialect which linguists sometimes call “standard edited English.” It is easy to overstate agreement regarding “standard edited English,” as that dialect has varied substantially over time; the “shall” versus “will” distinction used to be considered extraordinarily important, “correct” comma usage differs in British and American English and even more from the nineteenth century to now, and usage handbooks disagree on numerous issues (such as agreement). The notion of a correct dialect upon which there is universal agreement is simply a fantasy.
In our experience, people respond to this research by objecting to the pedagogy they assume it necessarily implies. People assume that to note the reality–considerable regional and historical disagreement regarding linguistic correctness–necessarily implies a complete abandonment of attention to error. That is not the necessary conclusion, nor is it our point. Our point here is simply that one central assumption in this view of writing instruction is wrong–there is not universal agreement as to rules regarding “correct” language use.
In addition, this research does not necessarily imply a “whatever goes” pedagogy. While some have drawn that conclusion, others have used this research to argue for teaching grammar and usage as a community of discourse issue (e.g., Labov and Smitherman); that is, rather than denigrate some dialects, teachers should present “standard edited English” as a useful dialect which students should use under some circumstances and with some audiences (see, for instance, “Students’ Right to Their Own Language”). Others have argued that grammar and usage should be taught as a rhetorical issue, as a question of clarity and rhetorical effect (Williams, Kolln, Dawkins).
And this leads us to the second point–the assumption that one can reduce errors in student writing through making students learn the rules of standard edited English. On the contrary, in the nearly one hundred years that this issue has been studied, there has not been a single study which showed improvement in student writing resulting from formal instruction in the rules of grammar, while there are several studies which showed a mark deterioration (see Knoblauch and Brannon, Hartwell, Hillocks for more on the history of this research). That deterioration may be the consequence of increased anxiety leading to students’ mistrusting their implicit knowledge (Hartwell), or that the time taken for grammar instruction was time away from more productive forms of writing instruction (Knoblauch and Brannnon).
In our experience, this point too is misunderstood. We are not saying that instruction in grammar and usage is pointless, but that certain approaches to it demonstrably are. And those are precisely the pedagogies into which one is forced in large classes–lecturing, drilling, assigning worksheets, and testing students on usage rules.
Indeed, research suggests that there is probably not a pedagogy which can be applied to all students in the same way. Issues of linguistic correctness result from different causes, depending upon the students. Hence, the solution varies. For students whose native dialect is fairly close to standard edited English, for instance, errors in usage sometimes result from lack of clarity about their own argument; students make more usage errors, for instance, when they are writing about something they do not fully understand. For such students, clarifying the concepts will enable the students to correct the errors.
For other students, usage errors are a time management issue–they did not leave themselves time to proofread. What Haswell has somewhat misleadingly called “minimal marking” is generally the best strategy under those circumstances (it is misleading in that it depends upon students’ resubmitting their corrected papers, so it can be fairly time-consuming for the instructor, albeit far less time-consuming and more effective than copy-editing). What he advocates, however, is not a kind of marking that takes minimal time on the part of the instructor.
For students whose dialect is markedly different from standard edited English, there is the possibility of what linguists call “dialect interference”–instances of using their (academically inappropriate) dialect, engaging in hypercorrectness (“between you and I”), or simply being unsure how to apply the rules. There are also students whose experience with written English is minimal, and who may have a tendency toward what are called “errors of transciption” (e.g., errors regarding the placement of commas and periods). For these students, “minimal marking” is ineffective, but neither do they benefit from lectures and quizzes on grammar rules. Instead, they seem to benefit most from individual instruction. Several studies show strong short-term improvement from sentence embedding (Hillocks), but many instructors moved away from it due to its inherently time-consuming nature.
In short, as Mina Shaughnessy pointed out long ago, improving students’ usage is not something one can do in the same way with all students. One must know exactly what specific problems exist with each student, why that student is having that problem, and what method will best work with that problem and that student. In other words, effective instruction in grammar and usage necessitates classes small enough that the teacher can know students well enough to know the cause of the problem. If the students have major problems, as from dialect interference, then the classes have to be small enough for the teacher to be able to engage in the extremely time-consuming methods necessary for such students.
One might wonder, if writing teachers are not teaching rules of writing, what are we teaching? And the answer seems to be that we are teaching rhetoric. That is, while one cannot present students with rules that apply to all circumstances–never use I, always begin with a personal anecdote, your thesis should have three reasons–there are principles which do seem effective in most circumstances. Those principles are encapsulated in the concept of the “rhetorical situation”–that the quality of a piece of discourse is determined by the extent to which its strategies are appropriate for effecting the author(s) particular intention on the specific audience. Thus, were one to examine prize-winning articles in philosophy, economics, literary criticism, engineering, behavioral psychology, and theoretical physics, one would see wide variation in terms of format, style, organization, and nature of evidence, one would see that each piece was appropriate for its audience.
One advantage of this approach to the teaching of writing is that it is more effective. Lecturing and drilling are, as several studies have shown, ineffective methods of writing instruction (Hillocks). This method remains tremendously popular, however, especially among teachers whose own instruction followed that method, who are cynical regarding student achievement, and who are generally convinced that the teaching of writing is the transmission of information (Hillocks). This point is important, as it showed up in our own experiment with reducing class size–students in classes with teachers who relied heavily on lecture did not show any benefit from a smaller class. The fact is that lectures are ineffective in writing classes; reducing the class size does not suddenly make lecturing an effective teaching strategy.
When we had the opportunity to look closely at class size at our previous institution, we made some surprising discoveries. One of the major motivations for undertaking the experiment was a sense of frustration, among faculty and graduate students, with graduate student instructors’ progress toward their degrees. Prior to the change in program emphasis, a large number of our instructors used class time to present advice on writing papers as well as to present writing products which students used as models (what Hillocks calls the “presentational” mode, and which he identifies as the least effective method of writing instruction). Perhaps because this method of instruction did not work particularly well for so many students, instructors also relied heavily on individual conferences with students–conferences which took so much time that they necessitated long blocks of time outside normal office hours. The dominance of this mixing of presentational and individualized modes of instruction had fairly predictable consequences.
The accretion of assignments and expectations for the course meant that it was actually impossible to teach the course in the ten hours per week a graduate student was supposed to spend on it. While such a situation is far from uncommon–many programs pay writing teachers a salary that presumes that the course takes much less time than it actually does–it is unethical. It also means that instructors, especially ones with multiple commitments (e.g., graduate students who are also taking courses, part-time instructors with obligations at several campuses, tenure-track teachers facing publication pressures), are encouraged to adopt pedagogies which feel more efficient but which research strongly indicates are less effective (i.e., the presentational mode of teaching, discussed previously).
Graduate student instructors responded to this situation in various ways. According to a survey, as well as faculty observation, many let their own coursework suffer in favor of their teaching. Others simplified assignments, so that the papers were short and simple enough that they could be graded in ten to fifteen minutes a piece. Several instructors essentially abandoned assessing student work, and graded students purely on attendance. Many instructors reported spending long hours on teaching, something that, not surprisingly, resulted in frustration–the first year composition course was openly discussed as the least desirable teaching assignment. In this context, it should be clear why we were looking for a method that would reduce the amount of time that instructors spent on their first year composition courses, without simply shifting them to quick, but ineffective, methods such as lecturing, drilling, and superficial grading.
When we reduced class size to fifteen for many of the instructors, we found that those instructors generally spent less time on the course (instructors in control groups reported spending an average of ten to fourteen hours per week on their courses, instructors in the sections with reduced class size reported averages of between twelve and fifteen). We also found that many instructors took advantage of the reduced class size to create new assignments, to take more time to comment on papers, to meet more often with students, or to add another project. Such a consequence–instructors taking the opportunity to increase the amount of work in the course–is echoed in at least one other study on class size. The San Juan Unified School District report on the results of the Morgan-Hart Class Size Reduction Act of 1989 concludes that
As a result of smaller classes, students were more actively involved in the instructional process. This was demonstrated by an increase in the number of student reading and writing assignments, more oral presentations and frequent classroom discussions. Students also received increased feedback on their English assignments and teachers had time to work with students individually.
One benefit of reducing class size, then, is that instructors appear more willing to experiment with and examine their teaching styles. Whether this is a bug or feature would depend on the program goals. Certainly, although they may not have spent less time on the courses, they reported much higher satisfaction. Teachers like smaller classes.
But, they did not always use the time well. We found that instructors heavily committed to the presentational mode did not effect much change in their students’ writing processes. Similarly, class size did not increase overall student satisfaction if the instructor engaged in the presentational mode.
In conclusion, our experience fits with Sheree Goettler-Sopko’s summary of research on class-size and reading achievement. She concludes that “The central theme which runs through the current research literature is that academic achievement does not necessarily improve with the reduction of student/teacher ratio unless appropriate learning styles and effective teaching styles are utilized” (5).
Class Size and Minimal Teaching
George Hillocks long ago showed the importance and superiority of constructivist approaches to the teaching of writing (Research in Written Composition, Teaching Writing as Reflective Practice, and more recently Ways of Thinking, Ways of Teaching). This means that effective teaching requires an approach which does not set the task of teaching writing as getting students to memorize and understand certain objects of knowledge (the objectivist approach), but as setting students tasks during which they will learn and giving them appropriate feedback along the way. The more that one engages in constructivist teaching, the more important is class size; the more that the goals and practices of a program are objectivist, the less class size matters. While reducing class size does not guarantee constructivist teaching, increasing class size does prevent it.
One can see this effect simply by thinking about the amount of time for which writing instructors are paid. The assumption at many universities is that each class is supposed to take 8-10 hours per week of instructor time. Instructors spend three hours each week in class, and it is optimistic, but not necessarily irrationally so, to assume that an efficient and highly experienced teacher can prepare for class on a one-to-one basis (that is, that it takes approximately one hour to prepare for one hour of class). A teacher therefore has two to four hours a week (almost precisely what is required by most universities for office hours). If an instructor has twenty students per class, s/he has, over the course of the semester 30-60 hours, which comes, at best, to three hours per student for conferences and grading. This situation necessitates cutting the students short on something–short papers which can be graded quickly, cursory grading of student work generally, discouraging students from using office hours. All in all, it means that one cannot do what Pascarelli and Terenzini say “effective teachers do” when “They signal their accessibility in and out of the classroom” (652). Simply put, if instructors have to use office hours to grade student work, they cannot signal accessibility. Pascarelli and Terenzini say, “They give students formal and informal feedback on their performance” (652), but, if instructors are restricted to three hours of grading per semester per student, they have to minimize the amount of feedback given. In other words, large classes force instructors away from what “we know” to be good practice.
The larger the class, the more the teacher is forced into lecturing. Yet, according to Pascarelli and Terenzini,
Our review indicates that individualized instructional approaches that accommodate variations in students’ learning styles and rates consistently appear to produce greater subject matter learning than do more conventional approaches, such as lecturing. These advantages are especially apparent with instructional approaches that rely on small, modularized content units, require a student to master one instructional unit before proceeding to the next, and elicit active student involvement in the learning process. Perhaps even more promising is the evidence suggesting that these learning advantages are the same for students of different aptitudes and different levels of subject area competence. Probably in no other realm is the evidence so clear and consistent. (646, emphasis added)
If we want instructors to be effective writing instructors, then we have to ensure that they are in a situation which will permit good practice. Reducing class size will not necessarily cause such practice, but it is a necessary condition thereof.
Works Cited
ADE. “ADE Guidelines for Class Size and Workload for College and University Teachers of English: A Statement of Policy.” Online. http://www.ade.org/policy/policy_guidelines.htm. 1998.
Baron, Dennis E. Grammar and Good Taste : Reforming the American Language. New Haven: Yale University Press, 1982.
Bazerman, Charles. Shaping Written Knowledge: The Genre and Activity of the Experimental Article in Science. Madison: University of Wisconsin Press, 1999.
Berkentotter, Carol. “Decisions and Revisions: The Planning Strategies of a Publishing Writer.” Landmark Essays on Writing Process. Sondra Perl, ed. Davis, CA: Hermagoras Press, 1994. 127-40.
Braddock, Richard. “The Frequency and Placement of Topic Sentences in Expository Prose.” On Writing Research: The Braddock Essays, 1975-1998. Ed. Lisa Ede. New York: Bedford, St. Martin’s, 1999. 29-42.
Chatman, Steve. “Lower Division Class Size at U.S. Postsecondary Institutions.” Paper presented at the Annual Forum of the Association for Institutional Research. Albuquerque: 1996.
Chomsky, Noam. N. Aspects of the Theory of Syntax. Cambridge:
MIT P, 1965.
Davis, Barbara Gross, Michael Scriven, and Susan Thomas. The Evaluation of Composition Instruction. 2nd. Ed. New York: Teachers College Press. 1987.
Dawkins, John. “Teaching Punctuation as a Rhetorical Tool.” CCC (Dec. 1995): 533-548.
Emig, Janet. The Composing Processes of Twelfth Graders. Urbana: NCTE, 1971.
Faigley, Lester, and Stephen Witte. Evaluating College Writing Programs. Carbondale: Southern Illinois UP, 1983.
Fish, Stanley. Is There a Text in this Class? Cambridge: Harvard UP, 1982.
Flower, Linda, and John R. Hayes. “The Cognition of Discovery: Defining a Rhetorical Problem.” Landmark Essays on Writing Process. Sondra Perl, ed. Davis, CA: Hermagoras Press, 1994. 63-74.
Glass, Gene V., and Mary Lee Smith. “Meta-Analysis of Research on the Relationship of Class-Size and Achievement. The Class Size and Instruction Project.” Washington D.C.: National Institute of Education, 1978.
Goettler-Sopko, Sheree. “The Effect of Class Size on Reading Achievement.” Washington D.C.: U.S. Department of Education, 1990.
Graff, Gerald. Beyond the Culture Wars: How Teaching the Conflicts Can Revitalize American Education. New York: WW Norton, 1993.
Hairston, Maxine. “Working with Advanced Writers.” CCC 35(1984): 196–208.
Hartwell, Patrick. “Grammar, Grammars, and the Teaching of Grammar.” College English 47 (February 1985): 105–27.
Haswell, Richard H. “Minimal Marking.” College English 45.6 (1983): 166-70.
Hillocks, George. Research in Written Composition: New Directions for Teaching. Urbana: NCTE, 1986.
– – -. Teaching Writing as Reflective Practice: Integrating Theories. New York: Teachers College P., 1995.
– – -. Ways of Thinking, Ways of Teaching. New York: Teachers College P., 1999.
Huot, Brian. “Toward a New Theory of Writing Assessment.” CCC 47.4 (1996): 549-66.
Knoblauch, C.H. and Lil Brannon. “On Students’ Rights to Their Own Texts: A Model of Teacher Response”, College Composition and Communication, 33 (1982): 157-66.
Kolln, Martha. Rhetorical Grammar: Grammatical Choices, Rhetorical Effects. 4th Ed. New York: Pearson, 2002.
Labov, William. The Logic of Non-Standard English. Champaign: National Council of Teachers of English, 1970.
Lanham, Richard. Revising Prose. 4th ed. New York: Pearson Longman, 1999.
Miller, Susan. Textual Carnivals: The Politics of Composition. Carbondale: Southern Illinois UP, 1991.
NCTE College Section Steering Committee. “Guidelines for the Workload of the College English Teacher.” Online. http://www.ncte.org/positions/workload-col.html. 1998.
Ohman, Richard. English in America: A Radical View of Profession. New York: Oxford UP, 1976.
Pascarella, E.T. And Terenzini, P.T. How College Affects Students: Findings and Insights from Twenty Years of Research. San Francisco: Jossey-Bass, 1991.
Richards, I.A. The Meaning of Meaning: A Study of the Influence of Language upon
Thought and of the Science of Symbolism. 8th ed. New York: Harcourt, Brace
& World, 1946.
Rose, Mike. Lives on the Boundary. New York: Penguin, 1990.
Sommers, Nancy. “Revision Strategies of Student Writers and Experienced Adult Writers.” CCC 31 (December 1980): 378–88.
San Juan Unified School District. “Class Size Reduction Evaluation: Freshman English, Spring 1991.” Washington D.C.: U.S. Department of Education, 1992.
Shaughnessy, Mina. Errors and Expectations. New York: Oxford UP, 1979.
Slavin, Robert, “Class Size and Student Achievement: Is Smaller Better?” Contemporary Education 62 (Fall 1990): 6-12.
Smitherman, Geneva. “‘Students’ Right to Their Own Language’: A Retrospective.” English Journal 84.1 (l995): 21-27.
Swales, John, and Hazem Najjar. “The Writing of Research Article Introductions” Writtten Communication 4.2 (April 1987): 175-91.
Tomlinson, T. M. “Class Size and Public Policy: Politics and Panaceas.” Educational Policy 3 (1989): 261-273.
Trail, George Y. Rhetorical Terms and Concepts: A Contemporary Glossary. New York: Harcourt, 2000.
Williams, David D., et al. “University Class Size: Is Smaller Better?” Research in Higher Education 23.3: 307-318.
Williams, Joseph. “The Phenomenology of Error.” CCC 32 (May 1981): 152-68.
—. Style: Ten Lessons in Clarity and Grace. Chicago: U. Chicago P., 1997.
[1] Unhappily, in our experience, the expectation is that instructors should spend more than forty hour per week on their jobs, or cut corners in various ways. For instance, it is often assumed that office hours can be used for course preparation or grading, but that amounts to an official policy that office hours are not times when students can expect the full attention of the instructor. Hence, when upper administrators say that office hours should not be counted separately from course preparation, the correct answer is, “Put that in writing.”