1
Bringing Together Reading and Writing to Enhance Reading Comprehension
Bringing Together Reading and Writing
for the Purpose of Enhancing Reading Comprehension
James L. Collins and Jaekyung Lee
State University of New York at Buffalo
Paper presented at the annual meeting of the National Reading Conference, Miami FL
December 2, 2005
Introduction to the NRC Symposium
This symposium summarizes the progress and findings of the first year of the Writing Intensive Reading Comprehension (WIRC) study funded by the Institute of Education Sciences, U.S. Department of Education[1]. It reports on progress made during Year I toward achieving the study’s two major goals: (1) A new integrated reading-writing curriculum entitled Writing Intensive Reading Comprehension for enhancing reading comprehension and writing performance by teaching reading and writing simultaneously and interactively, and (2) A set of empirical data which speaks to the effectiveness of WIRC compared to the teaching of reading comprehension and writing separately. In the first paper I will discuss the theoretical underpinnings for integrating reading and writing in the WIRC study. In the second paper Tim Madigan will describe thinksheets, the curriculum component at the heart of the WIRC intervention, and he will focus on the development and evaluation of thinksheets during Year I of the study. In the third paper Jeff Fox will describe the design and evaluation of measures and methods for assessing the effectiveness of the WIRC intervention. Taken together, the three reports address first-year development and assessment issues in preparation for the Year II randomized controlled experiment which we are currently conducting.
Bringing Together Reading and Writing
for the Purpose of Enhancing Reading Comprehension
The Writing Intensive Reading Comprehension (WIRC) study[2] is investigating the hypothesis that bringing together reading and writing will improve reading comprehension and writing performance in low-performing urban public elementary schools. At the heart of the intervention being developed and tested is a curriculum innovation employing thinksheets step-by-step guides which students use to write about reading they are doing and a set of instructional methods for teachers to use the thinksheets interactively with students. Such thinksheets have been shown in previous research to be effective tools for guiding the writing processes of struggling students (Tierney & Readence, 2000; Collins, 1998; Collins & Collins, 1996; Collins & Godinho, 1996; Raphael, Kirschner, & Englert, 1988; 1989; Englert, 1995; Englert & Mariage, 1990). By developing interactive thinksheets and testing their effectiveness through experimental research, the WIRC study challenges the conventional assumption that low-achieving students cannot use writing to make sense of their reading.
Year I efforts toward creating a new integrated reading-writing curriculum focused on the development of the thinksheets and instructional methods comprising the WIRC curriculum. Both the thinksheets and instructional methods were under constant development, observation, critique, and revision. These efforts were guided by several types of inquiry: an ongoing review of research on reading-writing relations, the teaching of reading comprehension, and the teaching of writing; process studies conducted to inform the development of the thinksheets and instructional methods; pilot testing of thinksheets with teachers and students in fourth and fifth grade classrooms; critical evaluation and revision of completed thinksheets; interviews with teachers; and a year-long study which analyzed the WIRC thinksheets by examining how they were used by a group of eight low-performing fourth graders in one of the project classrooms. Data in the year-long study consisted of writing tests the fourth graders took early and late in the year and all their intervening thinksheets, classroom observations, and teacher interviews between the two tests. In this paper I’ll present highlights of Year I in four categories: building a conceptual base for the study; developing , piloting, and revising the thinksheets and the instructional methods for using them; recruiting schools and training teachers for Year II; and developing process and outcome measures.
Building a Conceptual Base for the Study
An ongoing review of research was crucial to the WIRC Year I development process because it clarified what it means to seriously consider teaching reading and writing simultaneously to students who struggle with both. Historically, reading and writing have been separated in American education (Fitzgerald & Shanahan, 2000), and this separation has several dimensions: social values, as when reading is valued more than writing (Kaestle, 1985); political divisions, as in the case of reading and writing having different professional organizations (Clifford, 1989); and pedagogical and developmental perspectives, as in theories portraying reading and writing as mutually irrelevant (Shanahan, 1988). At the same time, we have a rich tradition of empirical research pertaining to connections between reading and writing (Shanahan, in press). Current educational practices across the curriculum and across the grades separate reading and writing, especially for low-performing students, even though research over several decades indicates reading and writing are connected developmentally and should be connected instructionally as well. Shanahan (in press) recognizes the complexity of combining reading and writing instructionally and argues for the need for design experiments that show how to do that most productively. We are, of course, in complete agreement, since we are currently conducting such an experiment in Year II of the WIRC study.
Based on our ongoing review of the literature and our Year I experience, we agree with Shanahan (in press) on the need for research in two areas, connecting reading and writing instructionally, and studying how classroom discourse can support reading-writing connections:
1.“Reading and writing instruction can be usefully combined, but instruction in one or the other is unlikely to be an adequate replacement for the other if the goal is to develop students who can read and write well.” P. 16, emphasis in original
2.“There continues to be a paucity of research into the relations between oral language development and literacy, particularly with regard to studies that show how to use oral language towards better reading and writing skills or how to teach or support language development in ways that positively impact literacy.” P. 21
As I will show in this section, developing and testing a specific instructional reading-writing connection and embedding it in relevant useful forms of classroom discourse was a major focus of our Year I efforts.
Foremost among our theoretical and conceptual underpinnings for bringing together reading and writing and designing the WIRC curriculum was Bereiter and Scardamalia’s (1987) distinction between knowledge telling and knowledge transforming. Bereiter and Scardamalia use this distinction to describe contrasting writing processes children exhibit when asked to compose aloud. Knowledge telling is using witing in a straightforward manner to unreflectively tell what one knows, and knowledge transforming is using writing to reflect on, restructure, and thus enhance what one knows. Their analysis concludes with the argument that knowledge telling is most likely an instruction-induced writing process, the result of the academically ubiquitous tendency to use of writing as a means of testing students on recently taught course content. Our findings during the grant application process and early in Year I are consistent with Bereiter and Scardamalia (1987) in that we observed that teachers are seemingly universally accustomed to using writing primarily after reading to have students tell what they know about the reading. This conclusion is based on data from our observations during a pilot study we completed as part of the grant application which observed eight lessons, two in each of four English language arts classrooms in four different schools in the target district, and on initial observations in four classrooms in the two low-performing urban elementary schools we recruited to work with us on developing the WIRC intervention during Year I. The instrument we used to collect these observations is in Appendix A. The default setting for writing about reading in our target district (and Bereiter & Scardamalia, 1987, would argue throughout the profession) is that writing is a means of testing students by having them write to tell what they know about a selection they have read. This model of writing about reading is shown in Figure 1, an adaptation of Bereiter & Scardamalia’s (1987) knowledge telling model of the writing process.
Figure 1. Knowledge Telling model of the writing about reading process.
The large arrow superimposed on the model in Figure 1 represents our observation that knowledge telling is a rather straightforward process of telling what one knows, or remembers, about recently completed reading. We observed that students in our project classrooms before we introduced thinksheets generally either began writing soon after receiving a task or they did not write much at all; both of these modes of response may be dependent on memory of what was read, suggesting that knowledge telling is a useful metaphor for describing how students respond to tasks where writing is a form of testing what students recall from their reading. Without exception, students in our pre-thinksheet classroom observations responded by writing quickly or not writing much, without consulting the textbooks containing the literary selections they were writing about, even though their teachers had told them they could use their books.
The popularity of the knowledge-telling model may be reflective of the education profession’s apparent love-hate relationship with the idea of integrating reading and writing for low-performing students. Scholars are generally in favor of recognizing connections between reading and writing, but in fact no one has brought them together in meaningful ways for low-performing students; instead, for such students the dominant pedagogical assumption is still “Read first, Write later,” as in our universally observed condition of read a selection, close the book, and then write about the selection.
From these observations, we decided that the problem of bringing together reading and writing has both cognitive and social practice components (Purcell-Gates, Jacobson, & Degener, 2004; Hull & Schultz, 2002). For example, struggling readers and writers probably do not write much beyond the requirements of classroom practices which favor writing as a form of testing. Other instructional practices may be reductive and limiting, as in the case of the five-paragraph theme which many have argued is formulaic and constraining (Johnson, Smagorinsky, Thompson, and Fry,2003).
Consistent with cognitive/social practice theory, the WIRC study assumes that individuals appropriate the cognitive tools and signs that are available to them in their cultural contexts (Vygotsky, 1934/1986, 1978; Bakhtin, 1981; Wertsch, 1991; Collins, 2003). In this view, literacy is achieved by using the tools of literate interaction appropriate within cultural settings, including homes, neighborhoods, and schools. This view is supported by constructivist perspectives on language and thought (Vygotsky, 1934/1986; 1978), studies of reading and writing relations (Tierney & Pearson, 1983; 1994a; 1994b), reading (Morrow & Gambrel, 2000), writing (Collins, 1998), and standards and assessment (Valencia & Wixson, 2000). From the premise that literacy is achieved by using the tools of literate interaction appropriate within cultural settings the WIRC team concluded that reading and writing should be brought together in school in ways which reflect how they come together in students’ lives outside of school. This proposition challenges two mainstream assumptions: that comprehension is best developed by reading whole texts from beginning to end, without writing being called into the service of comprehending portions of text; and that writing about reading is primarily a test-like process of writing down what you remember about a recently read passage.
In accordance with our ongoing review of review of relevant research and our classroom trials of thinksheets, we made two crucial decisions early in Year I. The first was to distinguish between “Cover-to-Cover” reading and “Targeted Reading,” and the second was to include both but to favor targeted reading in developing thinksheets for the WIRC intervention. Cover-to-cover reading refers to reading a selection from beginning to end, usually for pleasure rather than information; it is the style of reading Rosenblatt (XXXX; XXXX) calls aestheticbecause it is the style of reading children and adults often enjoy doing for its aesthetic rewards think of a winter evening, a fireplace, an easy chair, a piece of homemade fudge, and a good book. Targeted reading refers to directed reading aimed at achieving a focused purpose, such as getting specific bits of information from a specific text; it is the style of reading Rosenblatt (XXXX; XXXX) calls efferent, the style adults do when reading while at work (for example, on reports such as this) and it usually involves reading to get information and doing something as a result fixing a flat tire, playing a video game, printing directions to a new destination. Interestingly, targeted reading shows a good match with Bereiter & Scardamalia’s Knowledge Transforming model of the writing process. By beginning with a search for relevant bits of information in the text they are reading, and then organizing these related but discrete bits into a coherent whole, students are encouraged to create and structure meaning, and thus construct and transform their knowledge of their reading. Figure 2 shows the knowledge-telling process in the dual contexts of academic and social literacies at the heart of the current WIRC conceptual model for bringing together reading and writing.
Figure 2. Knowledge Transforming Model of the Reading-Writing Process
The WIRC knowledge transforming model, again an adaptation of Bereiter & Scardamalia’s model (1987), represents our observation that knowledge transforming is not a straightforward process of telling what one knows, or what one remembers, about recently completed reading. Students still tell what they know in the final step of the model, but prior to that they do a considerable amount of constructing their knowledge using the thinksheet we have provided and teacher and/or peer assistance. We observed repeatedly in Year I that students using the knowledge transforming approach and one of our thinksheets begin writing only after re-reading portions of the selection they are writing about, rather than right after receiving the task as in the pre-thinksheet observations described above. Whereas students using the knowledge telling model write with their books closed, students using our thinksheets and the knowledge transforming approach do what we have come to call “two-handed reading.” They write with one hand on the book they’re writing about and one hand on the thinksheet they are using.
Following Purcell-Gates (Purcell-Gates, Jacobson, & Degener, 2004) who developed the cognitive/social practice theory of print literacy development referred to above, and in company with Gee (2003) who points out that kids only read the manual after they get stuck playing a video game, the WIRC intervention made full use of targeted reading in designing thinksheets. Our observations lead us to claim that the thinksheets foster purposeful, goal-directed reading by targeting certain information in texts being written about. Simply stated, the thinksheets insure there will be no more writing with the textbook closed.
Two key principles to guide thinksheet design evolved from our emphasis on targeted reading:
- Targeted reading looks for information in a purposeful fashion. Thinksheets should lead students to identify and record information that is necessary to complete a specified task.
- Targeted reading is organized, not random. Thinksheets should help students in a focused, deliberate manner to find and arrange information to solve problems involved in comprehending and writing about their reading.
The next section of this report will describe how we used these principles to develop the WIRC thinksheets during Year I.
Developing, Piloting, and Revising Thinksheets and Instructional Methods
The WIRC intervention is designed to test a solution to the problem of low performance on reading comprehension and related writing tasks. As we said earlier, the heart of the intervention is a curriculum innovation which uses thinksheets and a set of instructional methods for teachers to use the thinksheets interactively with students. Thinksheets are step-by-step guides to writing about reading, but they are not meant to be stand-alone worksheets. Throughout our work in developing the WIRC intervention, we have been careful to not allow educators to view the use of thinksheets as a solution in itself, without devoting instructional resources needed to use the thinksheets wisely. We do not want teachers to hand thinksheets to students and then walk away. From the beginning of our work, we envisioned teachers using thinksheets interactively and discursively with students in focused reading-writing workshops where discussions and individual conferences and teacher modeling guide the use of thinksheets for students, including students who struggle with literacy.
The logic entailed in thinksheet design is this. Writing is an instrument for the social construction of knowledge. With writing we take inchoate ideas and develop them by interacting with the ideas of others, especially in the case of the WIRC research, the ideas of others gained through reading, discussing, and conferencing. When writers cannot transform inchoate knowledge by spelling it out in detail, they have a problem (or usually several problems). The purpose of thinksheets is to help solve these problems. A thinksheet is not a worksheet, to be filled in independent of the selection under study or in the absence of interaction through discussion and conferencing. And it is not a self-teaching device, to be completed independent of the teacher. A thinksheet records the result of reading, writing and conversing on paper it presents anticipated problems in writing about reading at the same time it presents solutions. Having students write their way through reading comprehension problems may be better than only talking them through the same problems. In the teacher interviews in Year I of the WIRC project, teachers often told us that their students can do the work when the teachers talk them through it, but the teachers added that talking students through their difficulty doesn’t enable the children to do the work by themselves. Kids also have to solve the problem themselves, even as we lead them to a solution.