Learning and Teaching Writing
Writing is a very demanding and complex task. Even a seemingly simple text, such as Cat in the Hat, can require considerable effort and expertise. It took Dr. Seuss well over a year to write the book, and he noted that “every word is a struggle—every sentence like a pang of birth.”
Writing is a goal directed and self-sustained activity requiring the skillful management of the writing environment; the constraints imposed by the writing topic; the intentions of the writer(s), and the processes, knowledge, and skills involved in composing (Zimmerman & Rei-semberg, 1997). It entails much more than this, however, as writing is a social activity involving either an implicit or explicit dialogue between writer(s) and reader(s). Writing is further shaped by the community of the writer. For example, written discourse differs considerably amongst a community of friends sharing ideas via email and texts written by biologists (Nystrand, 2006). Moreover, writing competence in one social community does not ensure competence in another. For instance, a good technical writer may not be a good novelist. What and how people write is also influenced by the cultural, societal, institutional, political, and historical background in which they are situated (Schultz & Fecho, 2000). To illustrate, students' concepts about writing are shaped, at least in part, by institutional decisions about pedagogy and curriculum. If a school's writing program places a heavy emphasis on correct form, students' revising efforts will most likely involve editing. A different approach to revising is likely, though, if form is deemphasized and meaning and process are stressed.
Given its complexity, it is not surprising that there is currently no model or theory of writing that fully or adequately captures it. One conceptual approach to studying writing focuses mostly on the individual writer and concentrates on understanding the cognitive and the motivational processes involved in composing (Graham, 2006). This cognitive or cognitive/motivational approach is exemplified in an influential model of writing developed by Hayes (1996). In his model, he takes into account, at least in part, the interaction between the task environment for writing and the internal capabilities of the writer. The task environment includes both a social component (e.g., the audience, other texts read while writing, and collaborators) as well as a physical component (e.g., text read so far and the writing medium, such as a word processor).
Internal factors include four main elements. First, cognitive processes: text interpretation, reflection, and text production. These processes allow the writer to form an internal representation of the writing task that can be acted upon; devise a plan to reach one or more writing goals; draw conclusions about the audience and possible writing content; use cues from the writing plan or text produced so far to retrieve semantic information that is then turned into written sentences; and evaluate plans and text and modify them as needed. Second, motivation, which includes the goals, predispositions, beliefs, and attitudes that influence the writing process. Third, long-term memory—knowledge of the writing topic and audience as well as linguistic and genre knowledge, including task schemas that specify how to carry out particular writing tasks. Fourth, working memory, which serves as an interface between cognitive processes, motivation, and memory, providing a space for holding information and ideas for writing as well as carrying out cognitive activities that require the writer's conscious attention.
In the model proposed by Hayes (1996) only limited attention is devoted to the social nature of writing. The influence of writing community, culture, society, institution, politics, and history are mostly ignored. One or more of these factors are captured in sociocultural theories of writing. For example, Russell (1997) developed a theory for explaining how macro-level social and political forces influence micro-level writing actions and vice versa. A basic unit in this model is an activity system, which examines how actors (an individual, dyad, or collective—perceived in social terms and taking into account the history of their involvement in the activity system) use concrete tools (e.g., writing) to accomplish some action with some outcome (this is accomplished in a problem space where subjects use tools in an ongoing interaction with others to shape an object over time in a shared direction).
Russell's theory also employs the concept of genre, “as typified ways of purposefully interacting in and among some activity system(s)” (p. 513). Genres are stabilized through regularized use of tools within and among individuals, creating a relatively predictable way of interacting with others, but they are only stabilized-for-now structures, as they are subject to change depending upon the context. Newcomers to an established activity system appropriate some of the routinized tools used by others (e.g., a particular structure for writing), but interactions between and among individuals and activity systems can change typified ways of acting
(i.e., genres), as they may be modified or abandoned in response to changing conditions. Activity theory provides a method for describing and analyzing activity systems for writing and how they interact with macro-level activity systems involving academic discipline, culture, institution, society, and so forth.
To illustrate activity theory in action an example is given below of how political, institutional, societal, community, cultural, and historical factors might influence what happens in a second-grade class focusing on story writing. In this particular instance, the teacher's decision to concentrate on story writing was shaped by the district curriculum guide and the state's high-stakes testing program (story writing was emphasized in both) as well as the teacher's and her students' interest in story telling. The way in which story writing was introduced and taught was influenced by the teachers' beliefs about how to teach (which was previously influenced by her teacher preparation program, her own teachers as a child, and the culture of the school). In providing story writing instruction, the teacher used the same general routinized approach that she had applied when teaching personal narratives and other types of writing. Students also continued to generate papers using the same general script they had been using since the start of the school year: selecting a topic, briefly planning what to say, making a draft, sharing it with a peer, revising and editing it, and sharing part or all of it with the class and at home.
While this script for writing was followed by most students, some of them modified it by eliminating a step (e.g., planning) or adding ones (e.g., sharing plans with a peer). The last of these modifications had a ripple effect in the classroom, as almost all of the students started sharing their plans with a peer. To provide students with concrete examples of stories, the teacher read traditional stories to the children (stories taken from her own dominant culture). However, one child brought to class a book of stories from Africa. He asked the teacher if they could try writing stories like those in his book. This request changed the focus of story writing in the class, as the teacher encouraged students to write stories from cultures other than their own, and several of the students' parents were asked to share their favorite stories from their culture with the class.
Not surprisingly, these two basic approaches to conceptualizing writing have led to different views of writing development. For example, Graham (2006) argued that four catalysts spur writing development. These involve changes in writer's strategic or self-regulatory behaviors (e.g., becoming more sophisticated in planning), motivation (e.g., heightened sense of efficacy about one's writing capabilities), knowledge (e.g., increased knowledge about the attributes and structures of different types of writing), and skills (e.g., automatization of handwriting and spelling and proficiency in sentence construction). These catalysts all reside within the individual, and this approach to development is consistent with cognitive/ motivational theories of writing.
In contrast, Schultz and Fecho (2000) offer a different view of writing development—one that is consistent with sociocultural theories of writing. They argue that writing development reflects and contributes to the social, historical, political, and institutional contexts in which it occurs; varies across the school, home, and work contexts in which it is situated; is shaped by the curriculum and pedagogical decisions made by teachers and schools; tied to the social identity of the writer(s), and is greatly influenced by the social interactions surrounding writing.
These two approaches (and the theories underlying them) clearly privilege different aspects of writing and writing development. However, neither is complete, as cognitive/motivational views pay relatively little attention to context, and sociocultural views do not adequately address how individual factors shape writing development.
The primary learning goals in schools during the primary grades are for students to master basic writing skills (such as handwriting, spelling, grammar, and sentence construction); begin to develop the strategic process needed to write effectively (e.g., planning, gathering and organizing information, monitoring, evaluating, revising, and so forth), acquire fundamental knowledge about writing (e.g., knowledge about the characteristics of good writing, needs of audiences, and so forth), learn to use electronic tools for composing (e.g., word processing and publishing tools), start to develop a life-long love for writing, and use writing for various purposes (e.g., communicate, inform, entertain, persuade, reflect, and so forth).
As students move into middle and high school, these same goals remain in play (although it is typically assumed that students have mastered some basic skills such as handwriting and spelling), with an emphasis on increasing students' competence as writers. Although students continue to write for a variety of purposes, using writing as a method for displaying subject-matter knowledge and as a tool for learning about such content becomes more prevalent. In some content areas (e.g., history), discourse genres typically used by that academic profession may be emphasized. At the college level, instructors may continue to emphasize more general writing development (especially for weaker writers), but as students enter classes in their majors, it is expected that they will learn the discourse styles and genres of that academic domain.
While states, school districts, and most schools have relatively clear goals for what students are to learn, there is no consensus on how to teach writing. The most prominent approach to writing instruction in the United States is the process approach (this is the only research-grounded general approach to writing instruction that has been extensively studied). This method is based on both cognitive/motivational and sociocultural views of writing. It involves extended opportunities for writing; writing for real audiences; engaging in cycles of planning, translating, and reviewing; personal responsibility and ownership of writing projects; high levels of student interactions, creation of a supportive writing environment; self-reflection and evaluation; and personalized individual assistance and instruction. This approach may only be effective, however, with teachers who are committed to its use and are trained in how to implement it. In a review of 21 studies, Graham and Perin (2007) found that this approach had little to no impact when these conditions were not met.
One concern with the process approach is that there may not be enough emphasis on explicitly teaching skills and strategic processes. The available empirical evidence indicates that directly and systematically teaching sentence construction skills and strategies for planning, revising, and editing as well as summarizing information can have a positive impact on improving the quality of students' writing (Graham & Perin, 2007). There is also some evidence that embedding more systematic instruction within the context of the process approach has a value-added impact on students' writing.
Students' writing skills can also be enhanced by providing them with assistance that helps them carry out one or more writing processes. Effective forms of support include clear and reachable writing assignment goals; help from peer(s) to carry out some aspect of the writing process; activities that help students generate, organize, and evaluate possible ideas for writing; examples of good writing that serve as a model for students; and technological supports such as word processing (Graham & Perin, 2007).
At present, it is not clear how these forms of effective writing practice should be combined or what amount of each should be provided. It is important to note that the identification of evidence-based practices in writing is far from complete. There are many practices in which there is at least some empirical evidence to support their effectiveness (this includes practices such as teaching vocabulary as a way to improve writing and having students assess their writing using a rubric). It is also certain that there are many scientifically untested practices used by teachers that will prove to be effective when evaluated by researchers. The study of exceptional teachers of writing provides one possible approach for identifying practices that merit such assessment.
There is one research-grounded treatment that has a strong impact on improving how well students write, especially struggling writers. This involves explicitly teaching students how to plan, revise, and edit their papers (teachers model how to use strategies and provide students with guided practice aimed at promoting effective and independent use). In 20 experimental studies reviewed by Graham and Perin (2007), such instruction improved the quality of students' writing in every single investigation. Particularly effective was a specific model for teaching writing strategies. The Self-Regulated Strategy Development model (see Harris, Graham, & Mason, 2006) not only involves modeling and guided practice, students are also taught the knowledge and skills needed to apply the strategies as well as procedures for regulating their use. Furthermore, this instruction is criterion-instead of time-based. This means that instruction continues for youngsters until they reach mastery, instead of providing a set number of instructional sessions.
Many procedures used to assess and evaluate writing quality are drawn from cognitive theories of writing. A cognitive approach to writing supposes that text production is a finalized and complete activity, in which several complex cognitive processes are activated. Therefore, assessment and evaluation of writing quality relies on the final product. Past research has focused on two measures of writing quality: subjective and objective. Subjective (or qualitative) measures use raters to evaluate writing quality based on one or more scales, whereas objective (or quantitative) measures consist of countable indices of writing sub-components.
Holistic, analytic, and primary-trait scoring are the three main subjective methods of evaluating writing quality. Frequently used, holistic scoring reflects a rater's overall impression of the writing, compared to other writing samples in the group. Holistic scoring is norm-referenced; that is, it provides a single score that ranks students within a particular group. Some holistic scoring methods also are criterion-referenced, using pre-determined characteristics of writing quality in the scoring process. Holistic scoring is the most economical method of scoring direct writing assignments (Scherer, 1985). One drawback to holistic scoring is that it does not provide instructional guidance for areas of concern within a writing sample.
Analytic scales and primary-trait scales address the need for assessment to inform instruction. Analytic scales are criterion-referenced and provide separate scores in predetermined areas of good writing, such as organization or development of the composition. They are the most reliable of all direct scoring assessment procedures (Scherer, 1985), although they take longer to score. One specific analytic scale, 6 Traits (and its later version: 6 + 1 Traits) has been used extensively in elementary and secondary schools. The 6 + 1 traits involve ideation, organization, voice, vocabulary sentence structure, conventions, and publication. Although developed for assessment, 6 Traits has been used as a method for writing instruction (see Graham & Perin, 2007, for a meta-analysis on the use of rubrics as an instructional method) and curriculum alignment, despite a lack of empirical research supporting its use for these situations.
Similar to analytic scales, primary trait scales also are criterion referenced. However, they differ from analytic scales in that the scoring guide is developed based on the specific purpose of each writing assignment. They can be used to assess the primary goal of the writing assignment (e.g., coherence of an argument) or to reflect genre-specific requirements (e.g., plot development).
Although in widespread use in both teaching and research, subjective measures of writing quality require scorer judgment and may be unreliable if extensive training and scoring directions are not provided. Other factors increase the variability of human scorers, such as fatigue, mood, and motivation (Freedman & Calfee, 1983). Computer-based scoring can parallel human graders; however, some scholars object to the validity of automated essay scoring (e.g., Ericsson & Haswell, 2006).
Reflecting a more sociocultural theory of writing, portfolio assessment collects several writing samples across different writing purposes and genres to gain a broader view of a student's writing ability. A good model of portfolio assessment includes the final product along with the drafts and revisions, allowing teachers to evaluate progress from start to finish. Some researchers advocate that portfolio assessment encourages students to develop writing over multiple occasions, rather than just a single sitting. Others believe that portfolios encourage more collaboration with peers rather than being teacher-centered, and that they are more consistent with teaching pedagogy (Murphy, 1994). Some districts and states collect student writing portfolios and then assess the written compositions using holistic, analytic, or primary-trait scoring methods. One concern with portfolio assessment is the time and cost to gather and evaluate each student's writing, whether in the classroom or in large-scale testing.
Objective measures of writing include the number of words written, percentage of correctly spelled words, or percentage of correct word sequences, among others. Objective measures are moderately correlated with subjective measures of writing quality (Espin, Shin, Deno, Skare, Robinson, & Benner, 2000); however, they often focus on one aspect of writing and serve as indirect measures of writing quality. Generally, these measures have been used in formative evaluation, allowing teachers to make changes to a student's instructional program based on frequent assessment.
Concerns over how to adequately assess and evaluate the quality of student writing have plagued the field for decades. The multiple purposes of writing, a lack of agreement on how to teach writing, and the increase of large-scale writing assessment have contributed to the debate. Any writing assessment method must meet standards of reliability and validity, while addressing legitimate needs for efficiency and a linkage between assessment and instruction.
Espin, C. A., Shin, J., Deno, S. L., Skare, S., Robinson, S., & Benner, B. (2000). Identifying indicators of written expression proficiency for middle school students. The Journal of Special Education, 34(3), 140–153.
Freedman, S. W., & Calfee, R. C. (1983). Holistic assessment of writing: experimental design and cognitive theory. In P. Mosenthal, L. Tamor, & S. A. Walmsley (Eds.), Research on writing: Principles and methods (pp. 75–98). New York: Longman.
Graham, S. (2006). Writing. In P. Alexander & P. Winne (Eds.), Handbook of educational psychology (pp. 457–478). Mahwah, NJ: Erlbaum.
Graham, S., & Perrin, D. (2007). A meta-analysis of writing instruction for adolescent students. Journal of Educational Psychology, 99, 445–476.
Harris, K. R., Graham, S., & Mason, L. (2006). Improving the writing, knowledge, and motivation of struggling young writers: Effects of Self-Regulated Strategy development with and without peer support. American Educational Research Journal, 43, 295–340.
Hayes, J. (1996). A new framework for understanding cognition and affect in writing. In M. Levy & S. Ransdell (Eds.), The science of writing: Theories, methods, individual differences, and applications (pp. 1–27). Mahwah, NJ: Erlbaum.
Murphy, S. (1994). Writing portfolios in K-12 schools: Implications for linguistically diverse students. In L. Black, D. A. Daiker, & G. Stygall (Eds.), New directions in portfolio assessment: Reflective practice, critical theory, and large-scale scoring (pp. 140–156). Portsmouth, NH: Heinemann.
Nystrand, M. (2006). The social and historical context for writing research. In C. MacArthur, S. Graham, & J. Fitzgerald (Eds.), Handbook of writing research (pp. 11–27). New York: Guilford.
Russell, D. (1997). Rethinking genre in school and society: An activity theory analysis. Written Communication, 14, 504–554.
Scherer, D. L. (1985). Measuring the measurements: A study of evaluation of writing. An annotated bibliography. (ERIC Document Reproduction Service No. 260455)
Schultz, K., & Fecho, B. (2000). Society's child: Social context and writing development. Educational Psychology, 35, 51–62.
Shermis, M. D., & Burstein, J. (2003). Automated essay scoring: A cross-disciplinary perspective. Hillsdale, NJ: Erlbaum.
White, E. M. (1985). Teaching and assessing writing. San Francisco: Jossey-Bass.
Zimmerman, B., & Reisemberg, R. (1997). Becoming a self-regulated writer: A social cognitive perspective. Contemporary Educational Psychology, 22, 73–101.