January 26, 2006

Beyond Strategies: Teacher Beliefs and Writing Instruction in Two Primary Inclusion Classrooms

By Berry, Ruth A Wiebe


Links between teachers' pedagogical beliefs and teaching practices were investigated with respect to process writing instruction. Participants included 5 teachers, 44 general education students, and 23 special education students in 2 elementary multi- age inclusion classrooms. Findings suggested that, although the teachers shared similar views on inclusion and were convinced of the uniqueness of their respective instructional approaches, they nuanced their writing instruction to conform to their implicit theories about teaching, learning, and disability. One set of teachers believed that the writing "breakdowns" of students with disabilities required a structural approach-sequenced, individualized, phonics-based instruction targeting individual performance levels. Another set of teachers advocated a relational approach, wherein students with disabilities are "protected" and "empowered" in learning communities characterized by shared activities, student choice, and interpersonal communication.

Teacher beliefs influence teaching practice and have an impact on students' educational experiences (Beach, 1994; Brantlinger, 1996; Brophy & Good, 1974; Gutirrez, 1994; Stanovich & Jordan, 1998; Vaughn & Schumm, 1996). Nevertheless, the potential links between teachers' theoretical orientations and their instructional practices- which are themselves laden with epistemological associations-are generally left unexplored in studies of classroom instruction. Do teachers align their beliefs with the often implicit epistemological principles underlying the instructional interventions they implement in their classrooms, or is it possible that they alter the interventions to conform to their own underlying beliefs, and does it matter?

Sociocultural perspectives on teaching and learning view knowledge as socially constructed. Corollary to that stance is the hypothesis that particular educational contexts lead to the construction of particular types of knowledge and to unique ways of being a student or teacher (Dixon & de la Cruz, 1992; Tuyay, Jennings, & Dixon, 1995). Thus, learning is located not only in the heads of individual students, but also in the various conversations and activities of which they are a part (McDermott, 1993; Slee, 1997). It has been shown, for example, that differences in context may account for differences in language acquisition and development (see, e.g., Heath, 1983). By revealing the beliefs and orientations underlying teachers' practices, information may be gained about how and why various contexts differentially mediate students' learning.

To examine the influence of teacher beliefs on teaching practices as it might apply to inclusion classrooms, two cases of writing instruction were investigated. Writing instruction was selected for these reasons: (a) Language difficulties, including writing, underlie the academic problems of many students with learning disabilities; (b) because they are perhaps less accountable to district or state curricular mandates than in areas such as reading or mathematics, teachers may have greater latitude in planning writing instruction; and thus (c) teachers may reveal their beliefs about teaching, learning, and literacy in fairly transparent ways that generally characterize the teaching and learning in their classrooms.

Process writing instruction was particularly highlighted in this study because it (a) enjoys widespread use as a component of writing curricula and (b) represents a well-defined approach to writing, in that most educators would recognize process writing as being based on principles of student choice and an emphasis on content over form (Bos, 1988). As a result, it might be assumed that all students who receive writing instruction based on the process approach have essentially similar experiences with teachers who share basic beliefs about writing and instruction. This study suggests that such an assumption might be false and that differences might not fall out along a special education-general education divide. Teachers' different approaches to implementing process writing instruction provided the impetus for this investigation. It was hypothesized that these differences might reflect basic beliefs held about teaching and learning in general.

Review of the Literature

Process writing, recommended as an instructional intervention in the findings of many studies of writing instruction, represents a shift in emphasis from product-oriented instruction (the mechanics of the finished text) to problem solving (ways in which text can be developed; e.g., planning, drafting, revising; Gersten & Baker, 2001; Stringer, Morton, & Bonikowski, 1999). Students with learning disabilities (LD) typically struggle with writing skills such as mechanics (e.g., organization, spelling, capitalization, punctuation), vocabulary, sense of audience, and text structures (Gregg & Mather, 2002). Process writing instruction can provide opportunities for these students to demonstrate writing strengths that are often obscured by difficulties with legibility and mechanics as they learn to recognize and improve dimensions such as content, voice, and creative ideas (Gersten & Baker, 2001; Sexton, Harris, & Graham, 1998; Stringer et al., 1999). Even educators who warn against overreliance on process writing as the primary approach to teaching writing to students who struggle with learning literacy skills have recommended its inclusion in a balanced approach that incorporates teaching mechanics as well as content and creativity (Graham & Harris, 1997).

Intervention Studies and Studies of Effective Teachers

Much research has been undertaken to find the most effective ways of teaching struggling writers. Following Gersten and Baker's (2001) meta-analysis of writing intervention research, Baker, Gersten, and Graham (2003) recommended interventions that have been shown to assist struggling writers. These included teaching the steps of the writing process (planning, drafting, revising); using feedback and interactive dialogue as teaching strategies; explicit teaching of genre conventions; providing frequent, extended opportunities for students to practice writing skills; and incorporating handwriting and spelling instruction into writing programs.

Two studies investigated the writing instruction practices of effective elementary special education teachers of literacy (Rankin- Erickson & Pressley, 2000) and primary grade teachers with high personal efficacy-that is, confidence in their ability to be effective in their teaching (Graham, Harris, Fink, & MacArthur, 2001). Both of these groups of teachers provided their students with frequent writing practice, taught the writing process as well as grammar and usage skills, and expressed a preference for methods that viewed the teacher as coach and placed importance on writing for real audiences. Furthermore, Rankin-Erickson and Pressley found that teachers in their study provided environmental supports such as classroom libraries, chart stories, poems, word lists posted in the classroom, learning centers, and big books. To begin instruction, these effective teachers started with the student's demonstrated skill level rather than where teachers thought the student should be. They encouraged risk taking and viewed writing as enabling purposeful communication for life. They also provided opportunities for using a variety of tasks and genres (e.g., stories, journals, complete books, shared writing) and embedded basic skills instruction within authentic writing experiences. Moreover, they tended to use themes across subject areas.

Teacher Orientations to Writing Instruction

Researchers who have categorized theoretical approaches to teaching writing (Graham, Harris, MacArthur, & Fink, 2002; Hillocks, 1984; Lipson, Mosenthal, Daniels, & Woodside-Jiron, 2000) have described similar categories. One approach, typical of much writing instruction, is skills dominated. Teachers using this approach tend to control topic selection, limit writing time, provide isolated skill instruction, focus on mechanics, and emphasize learning the stages of the writing process. A second, more student-centered approach is characterized by student choice, large amounts of time for exploratory writing, rules arising from students' writing experiences, regular peer and teacher conferencing, and writing for real audiences.

Hillocks (1984) described a third mode that combined elements of both approaches. The environmental mode is characterized by clear objectives, materials designed to engage students with the objectives and with each other in small-group problem-solving tasks, and a progression from teacher-led activities to small-group activities to independent task completion. Hillocks found this orientation to have the greatest mean effect size (.44) of the four orientations identified in his study. Graham et al. (2002) found that many teachers who identified themselves with either the explicit instruction or the natural learning orientations combined aspects from both orientations in actual practice, resulting in instruction resembling Hillocks's environmental mode.

Teacher Beliefs About Disability and Instruction

For most teac\hers, beliefs are formed early, remain highly durable, and acquire emotional dimensions (Pajares, 1992). Jordan and Stanovich (2003) identified two underlying categories of beliefs held by general education inclusion teachers about students with LD and linked these to characteristic patterns of classroom instruction. Pathognomonic teachers believed the disability to be a stable characteristic of the child, who needed to be removed from the general education classroom to receive special instruction by specially trained teachers. Pathognomonic teachers tended to interact less with students with LD and to use more directive, managerial talk with all students in their classrooms. Interventionist teachers believed the disability to be a barrier to learning but amenable to instruction in the general education context. These teachers used more academic than managerial talk with all students and assumed a coaching role to probe and prompt student understanding.

This study posits that teachers' deep beliefs inform their orientations to writing instruction. The fit of two theoretical models, both basically dichotomous-orientations to writing (explicit instruction vs. natural learning) and orientations to disability (pathognomonic vs. interventionist)-was explored relative to the teachers in the study to draw tentative conclusions about possible constellations of beliefs and practices. The study was guided by three research questions:

1. What was the nature of writing instruction in the two classrooms, particularly process writing instruction?

2. What did the teachers say they believed about teaching, learning, and inclusion, especially with regard to writing instruction?

3. Can similarities or differences regarding writing instruction in the two classrooms be linked to the teachers' beliefs?


Conceptual Framework

Clark and Peterson (1986) contended that "the purpose of research on teachers' implicit theories is to make explicit and visible the frames of reference through which individual teachers perceive and process information" (p. 287). Such frames of reference would be important to define if the underlying assumptions of the study were aligned with constructivist views of knowledge as "culturally and historically grounded, as laden with moral and political values, and as serving certain interests and purposes" (Howe, 2001, p. 202). The case to be made here, however, is not that differences in instruction existed, but rather to uncover how and why they existed despite similar teaching contexts and curricula. Clark and Peterson (1986) recognized the possibility that "even within what appear to be relatively homogeneous groups of teachers . . . there is wide variation in the content and orientation of teachers' implicit theories" (p. 291). The hypothesis underlying this study was that teachers must have constructed for themselves different conceptions of teaching and learning that influence their planning and instruction, regardless of the specific content.

Qualitative data are suitable for exploring questions in areas without strongly established categories or theories, which then are expected to emerge from the process of investigation and analysis (Creswell, 1994). The advantage of using qualitative data lies in their close relationship with natural contexts, ability to allow complexity, representation of more than brief snapshots, and capacity to explain the meanings that people ascribe to their lived experience (Miles & Huberman, 1994). Case studies allow the researcher to probe deeply into the phenomenon under investigation and, particularly, to connect hypotheses and understandings to real practice. This study comprises two contrasting cases to provide breadth as well as depth, as understanding the two cases together may provide both a more complete and more general view of the territory.

Setting and Participants

The settings were purposefully selected as reputational cases (Miles & Huberman, 1994) to examine contrasting belief systems at work in inclusion classrooms. Other researchers with access to information regarding the nature of instruction in the classrooms recommended the two classrooms to the author as examples of learning contexts that were similar in type (primary level inclusion), instructional quality (highly effective), and use of the process writing approach, but divergent in epistemological perspective. The goal of the study was to examine relationships between the process writing approach and epistemological perspective.

The research sites were two inclusion classrooms located in different schools in a midsize urban district reflecting a student body composed of approximately 46% European American, 33% African American, 12% Hispanic, 5% Asian American, and 1% Native American students, more than one half of whom received free or reduced-price lunches. The participating primary-level classrooms served a total of 44 general education and 23 special education students, first through fourth grade, marking them as multiage, multi-ability, multigrade configurations accommodating a widely diverse range of student learning characteristics (see Table 1).

The wide diversity of students in these classrooms was seen by the teachers as (a) a natural consequence of combining existing caseloads and (b) providing a desirable heterogeneity with attendant benefits-for example, the opportunity for students of varying levels of ability to help each other and to share different gifts and strengths. The teachers did not view this heterogeneity as detrimental to their teaching goals.

Differences in the makeup of the two classrooms included the following (Ellison statistics first, Watkins second): total number of students (38, 29), mean age in months (88, 103), grade level (87% Grades 1 and 2; 79% Grades 3 and 4), percentage of special education students (29%, 41%), racial/ethnic minorities (84%, 52%), and low socioeconomic status (SES; 100%, 66%; all special education students in both classrooms except one fell into this category).

The teacher participants included two teams of collaborating teachers, each team consisting of one or two full-time general education teachers and one full-time special education teacher. Both teaching teams strongly affirmed two commitments: full inclusion for students with high-incidence disabilities, and co-teaching. For both classrooms, co-teaching meant that (a) all team members jointly planned the overall curriculum, although individual teachers might plan individual lessons or teaching units within the overall plan; (b) many but not all lessons were jointly taught, with one teacher leading and one assisting; either teacher might lead; (c) all teachers taught all students, both general and special education; and (d) all teachers shared classroom management and grading responsibilities.

Case 1 Teachers: Darlene, Monica, and Brace (Ellison School). Darlene (general education) and Monica (special education; see Note 1) had cotaught in this inclusion classroom, called the team room, for 5 years. Both had 8 years of teaching experience. A third general education teacher, Bruce, had, along with his first graders, joined the team in the year the study was conducted. Bruce had 3 years of teaching experience. The classroom occupied a large space with ample accommodation for multiple activities. During the study year, the substitute principal provided a significant source of stress for these teachers due to her stated goal of eliminating the team room. The persistence of the teachers in the face of difficulty indicated a high level of commitment to their instructional program.

Case 2 Teachers: Rhonda and Angie (Watkins School). Rhonda (general education) had 15 years of teaching experience in Title I and general education contexts. Angie (special education) had 26 years of teaching experience in both general and special education contexts. Rhonda and Angie had established their inclusion classroom 2 years prior to the study. During 2 of their 3 years of co- teaching, their assigned classrooms were separated by a long walk down two hallways, which constrained their use of available space. Like Darlene and Monica, they persisted with their collaborative, inclusive arrangement despite marginal institutional support.


Classroom Demographics in the Research Classrooms

In both classrooms, students received all of their academic instruction in the inclusion classroom. Students were pulled out only for ancillary services such as speech, occupational therapy, or social work consultation. Configurations for academic instruction included whole-group, small-group, and individual instruction. Some small groups were heterogeneous and temporary; others tended to be longer lasting and based on performance level.

Data Sources

Teacher Interviews. The main sources of data included audiotaped teacher interviews, videotaped classroom lessons, and field notes. Teacher interviews were sources of belief statements and descriptions of practice. Two semistructured group interviews lasting approximately 90 min were conducted with each team: one prior to data collection and one debriefing interview at the end of the study. The first interview focused on drawing out teacher beliefs and teacher descriptions of writing instruction; therefore, this interview was analyzed for this report. Group interviews were appropriate because the teachers on each team maintained that they spoke with one voice, and indeed, they frequently finished one another's sentences. The interview consisted of 21 questions focusing on five main topics: (a) personal histories regarding collaboration, teaching students with disabilities, and their writing program; (b) opinions and experiences regarding inclusion; (c) goals and expectations with respect to writing instruction; (d) methods for teaching struggling writers; and (e) descriptions of writing instruction, including process writing.

Observatio\ns and Field Notes. A total of 55 classroom observations were conducted by the primary investigator. For this report, classroom observations were used to (a) corroborate and clarify what the teachers described in their interviews, (b) clarify beliefs and practices, and (c) provide specific illustrations of teacher practice. All classroom observations were videotaped. While the camera was running, the primary investigator recorded as much of the dialogue and classroom activity as possible on a laptop computer. Postobservation summaries were written and added. These field notes and summaries provided a description of each videotaped lesson and a basis for later selection of lessons to be fully transcribed.

Data Analysis Procedures

Teacher interviews. Several analytical procedures were applied to the transcripts. First, the transcripts were read and reread to acquire a sense of teacher perspective. Key phrases were marked. Second, a word-count analysis was conducted to identify frequently used terms (Ryan & Bernard, 2000). These were then located in context to help clarify underlying meanings. For example, both interviews contained the word chance. When located in context, it was found that Darlene and Monica used the word to explain that all students, with or without disabilities, deserved a "chance to have their individual . . . needs met," whereas Rhonda and Angie noted that all students should have the "same chance to . . . become a citizen of a community."

Teachers' explicit statements of the form "We believe such and such" were taken as belief statements. Moreover, implicit and unarticulated beliefs were sought by extracting from the interviews both metaphorical and non-metaphorical statements that might convey beliefs. Metaphors can be useful as tools in educational research because they are capable of providing insight into "how we look at the world, or how we construct reality" (Munby, 1987, p. 379), a central aim of this study. Statements were selected to meet the definition of one or more of the categories drawn by Nespor (1987) from Abelson's (1979) theory regarding the makeup of belief systems. According to Nespor, the following four features, taken together, define a belief:

1. existential presumptions: conditions, situations, or understandings that teachers seem to take for granted;

2. alternativity: idealized forms of practice that teachers seek to emulate;

3. affect: topics or ideas that elicit emotional responses; and

4. episodic storage: stories that teachers tell about themselves or their classrooms.

All extracted statements in each of the four categories were then sorted into themes. Some themes crossed more than one category; for example, independence was a theme that appeared in existential presumptions, alternativity, and affect. Further reducing the data, the themes in each category were clustered into metathemes. Memos (Miles & Huberman, 1994) were written to summarize the data for each belief feature.

Using Observations and Field Notes to Refine Interview Analyses. Initial coding of field notes categorized episodes of writing instruction by genre (narrative, expository) and type of participant structure (whole-group lessons, small-group activities, or individual work). An example of the use of classroom observation to clarify teacher explanations is the mention by both teacher teams of their use of "centers." Darlene and Monica planned center tasks, typically worksheet-like, for individual students who worked within study carrels, the barriers intended to help reduce distractions. Students were expected to complete the task exactly. Rhonda and Angie expected all their students to do as much as possible of the same center tasks, which typically involved writing extended text (e.g., a book review). The teachers accepted the students' approximations of task completion that were appropriate for their individual performance level as success, and they encouraged students to consult with each other as they worked (see Note 2).

Validity. Several techniques were employed to establish confidence in the credibility of this study (Lincoln & Cuba, 1985). These included (a) prolonged engagement with participants and setting (10 months); (b) multiple sources (teachers, researcher) and methods (interview, observation); and (c) member checks (the teachers were asked to review a draft of their case results).


The cases are presented individually in this section. Each case will be represented by (a) descriptions of the classroom environment and writing instruction, (b) a brief description of an observed whole-group process writing lesson, and (c) explication of teachers' beliefs based on their interview responses. The following criteria guided the selection of the example lessons: (a) instruction in a genre other than narrative; (b) a process writing lesson, as understood by the teachers; (c) a whole-group structure illustrating teacher-student dialogue; (d) instruction led by the general education member of the partnership; (e) considered by the researcher to be representative of the observed lessons; and (f) considered by the researcher to be a particularly good (i.e., effective) lesson in the specific context.

Please note the following points for clarity. All text marked with quotation marks is taken directly from the transcripts. Because this study deals with teachers' beliefs, it is important to read their exact words to be able to understand their perspectives and the conclusions and implications drawn from and about them. Also, hereafter the term Ellison teachers refers only to Darlene and Monica, not to other teachers at Ellison School, and likewise Watkins teachers refers only to Rhonda and Angle and not to their colleagues, unless specifically referenced. Similarly, the terms Ellison classroom and Watkins classroom refer only to the two participant classrooms, not to other classrooms at these schools.

Case 1: Darlene and Monica

Darlene and Monica described their writing instruction as "very divided up." By this, they meant that it was entirely task- analyzed, detailed, and sequenced. In fact, they suggested to the researcher that "maybe it would be best if you'd just copy this [notebook], because this is very in depth." Students progressed through the curriculum on two tracks simultaneously: the phonics track and the process writing track.

The phonics track entailed a flexible but highly individualized schedule of instruction dependent on students' skill levels. Individualization, the backbone of inclusion for Darlene and Monica, operationalized the egalitarian principles motivating their beliefs and practices, in that all students, with and without disabilities, deserved to be dealt with according to need. Individualization meant each student learning at his or her own pace and place along the same continuum of instruction. Darlene and Monica refused to predict how much progress a particular student might make in the long term, because they wanted to avoid putting "boundaries" on students.

I've got an extremely dyslexic second grade boy. I will be happy if he is able to write first and last sounds, you know, to write a sentence by the end of the year, . . . where we have other second graders who are going to be editing their own papers by the end of the year. . . . It's very, very individual. . . . We just keep going, wherever they can go with it.

Structured writing was a daily activity foundational for process writing, in that it provided a site for direct instruction in phonic, grammatical (grammar, sentence structure), comprehension (vocabulary), and conventional (spelling, punctuation) written language skills. Daily assignments for six to eight ability-based groups were written on chalkboards located around the room. After the teachers explained and modeled each task, students in each group individually completed the assignment. These daily tasks involved working within a text pattern or structure. For example, a lower performing group might deal with number and color word recognition by illustrating written text (e.g., three red balloons, four green fish), whereas a higher performing group might practice phonic skills by completing sentences and responding to a writing prompt.

The process writing track required everyone to learn the processesbrainstorming, drafting, editing, rewriting, and publishing- together. For Darlene and Monica, the term process applied not only to the major phases of a writing project but also to the multiple subphases that eventually culminated in "the whole process of the process writing." Darlene explained it this way:

[We teach] how we learn to stay on a topic; how we divide the big ideas into sentences. . . . Then we go into the story structure. [For example,] . . . we have a process to go through how to write something giving directions. We [have another] process which we go through in terms of story comparison using Venn diagrams. . . . This is all the way down the line.

Process writing, then, was the site for synthesizing all writing knowledge. Writing skills addressed within the framework of process writing focused on meaning-making and included sequencing, transitions, composing interesting sentences, and organization. Perhaps because a chief goal of writing instruction was to be able to get something down on paper, students lacking basic writing skills (e.g., encoding, matching sounds and symbols, spelling a number of sight words) were considered unready for process writing; however, in the interim, they could dictate their text to a teacher. The octopus report illustrates a typical lesson.

The Otopus Report. Typically, process writing assignments were introduced to the whole class, after which students were expected to complete the assignment individually. During the first phase of this wholeclass lesson, Darlene asked students to recall prior knowledge. As they did so, she endorsed and elaborated on their r\esponses. Monica concurrently recorded students' statements on a chalkboard for later referral. This phase was followed by organizing scientific concepts and, finally, detailing technical aspects of the report structure. In the following transcript segment, students recalled and Darlene commented. Marlee was a student with emotional impairment.

Darlene: Who else knows something about an octopus? Marlee?

Marlee: Some octopuses change their color when they're around, their color, they go, they change, they, their color.

Darlene: That's right, when they're around a certain color they change to that color. That's right.

Marlee: They blend so you can't see.

Darlene: They blend into their surroundings; what's that called?

Marlee: Camouflage.

Darlene: Right. Good. Who else knows something about an octopus?

Darlene's strategy was to recognize, evaluate, and accept students' ideas and to add her own elaborations and extensions. In this way, she drew out students' prior knowledge to establish a sort of information bank that would serve as a resource for all. It is likely that without this public knowledge-gathering activity, a number of students, including students with LD, would not have known what they knew. In other words, understanding the extent of one's own knowledge and being able to formally organize it is a complex task. By making available multiple resources for learning information, and then assembling and visually representing a corpus of knowledge about octopuses, the teachers supported students' access to the subject matter knowledge necessary for constructing the report.

At the end of the first phase of the lesson, a list of facts about octopuses (e.g., can change color/camouflage; have a beak inside mouth; can live in boats, caves, tin cans) had been recorded on the board. At this point, Darlene moved into the second phase of the lesson, the organization phase. She asked students to think of "big ideas" to organize the known information just recalled.

Darlene: We have a lot of ideas up here, but we can put them together in groups.

Student: That's what we did yesterday.

Darlene: That's what we did with the jellyfish a couple of days ago. Now, we're going to do this again. I need to have someone to tell me what one of these big ideas is. How can we put some of this stuff together in one big group? ... Alan, can you tell me what one of these big groups is?

Alan: Where they live.

Next, Darlene asked students to list places where octopuses live. As students responded, Monica again recorded their ideas. Later, each student would note the big ideas generated by the group, as well as the supporting details, on a brainstorming worksheet. The brainstorming worksheet served two functions: (a) to make visible the structure required for scientific text construction, involving superordinate concepts and subordinate details, and (b) to demonstrate how the resources of the chalkboard list could be used in the students' reports. The chalkboard list was intended as a resource only, not definitive of the only correct responses.

The third phase of the lesson, the technical support phase, involved a reminder of the students' task, which was to transform details from the class list into sentences for use in the reports. Darlene prompted a model of a complete sentence.

Marlee: You write a sentence.

Darlene: Well, how could I start? What could my first sentence be?

Student: Octopuses live in rocks.

Darlene: Octopus live in rocks. . . . To write your report it has to be in sentences. This is not a list. This is not a group of phrases. These have to be sentences.

Darlene's discourse simplified attentional requirements; students need attend only to her, not to their peers. She filtered student contributions through her talk. This benefited students who might have experienced difficulty attending to constantly shifting conversation. Her talk made knowledge "public, witnessable, and observable" (Macbeth, 2003, p. 258), an important function for the successful outcome of this lesson.

Metaphorical Traces of a Structural Belief System

Beliefs may have stable "core applications" (Nespor, 1987, p. 321); in other words, a perspective may consistently apply to particular situations (e.g., process writing). However, beliefs also possess the attribute of "unboundedness" (Nespor, p. 321)-that is, extensions to less predictable areas (e.g., all teaching behavior). It may be argued that although these teachers revealed specific orientations to their writing instruction, these orientations also provided evidence of systemic frames of reference that underlay all their teaching. Such systemic frames have three functions: (a) delimiting what teachers attend to, perhaps subconsciously; (b) selecting resources to address a situation; and (c) selecting strategies to coordinate resources (Nespor, 1987). Metaphors and direct statements used by the teachers during their interviews provided evidence of existential presumption, alternativity, affect, and episodic storage (Nespor, 1987) that defined beliefs informing instructional practices.

Darlene and Monica believed that structural approaches to curriculum and instruction mediated their students' academic success. Locational and structural metaphors dominated their language (e.g., "next level,""no boundaries,""cover a lot of territory,""systematic approach,""structural analysis") and defined their task-to locate students in the appropriate learning space: "We know what the next step is. We understand where we are in this continuum and where we want to go."

Darlene and Monica felt that they had developed a unique approach so effective it would establish a new "paradigm" for inclusive classrooms whereby students typically considered at risk for academic failure would advance far beyond customary expectations. They alluded to high expectations for student achievement by characterizing their instruction as "openended," saying they "don't have an end goal. We have no clue. . . . We just keep going, wherever they can go with it." To have an "end goal" would be to place inhibiting "boundaries" on the "territory" that students might cover.

Darlene and Monica generated more figurative language about their instructional strategies than about any other topic. The repeated tropes were transparent in meaning: "We've really broken it down,""whatever the structure is we're working with,""We got a process,""That's the start of what we were doing,""It keeps going as their skills develop,""at this point,""all the way down the line,""It's all planned." The implication of structure and sequence implied a scientific and orderly way of proceeding with a moral undertone of "good" (Abelson, 1979, p. 358).

Darlene and Monica had painstakingly constructed a well-defined curriculum to support students' acquisition of writing skills. What underlying assumptions supported this enormous task? Darlene and Monica held that skills deficits were due to "breakdowns" that needed to be "analyzed" and "pinpointed.""Working very specifically," they were always "pushing the next level," so that students moved along on the "continuum of progress." The assumed pedagogical theory of Darlene and Monica seemed to be as follows: Students who lacked writing skills were seen as out of order or broken; therefore, instruction involved directing the process of repairing the malfunctioning part or parts. The role of the student in learning was to follow directions to proceed step by step toward being fixed.

Their inclusive curriculum conformed to an egalitarian perspective of equality of value and need. Monica explained that all students "deserved that fair shot" for "someone to be invested in them" to give them "the equal chance.""Invested" was a metaphor for caring teachers who would "figure out where they're at and what they need." For them, "labeling" was a pejorative term, implying that segregation of children based on disability status was artificial, unnecessary, and improper, with strong moral implications-in Abelson's (1979) word, "bad" (p. 358). Their emotionally charged language resisted the idea that expectations might be based on group characteristics (e.g., disability): "[That idea] is horrible . . . no, no, no, no, no, not at all." For Darlene, at least, these notions may have been rooted in the teaching strategies she had once used with gifted students: "I also discovered . . . that working with gifted kids and working with special edfucation] kids is the same thing. You identify the entry point, and you go from there. . . . There's no difference."

Case 2: Angie and Rhonda

Asked to describe their writing instruction, Rhonda and Angie listed the writing opportunities they considered integral: response journals, mailboxes, Morning Message, Writing Workshop, Book Club, traditional centers, Reading Response, literature response, math journals, and "theme" (integrated science and social studies). Several of these activities underscored their view of writing as a tool for effective communication rather than as a skill in itself, with the potential to enhance both group cohesion in the classroom and personal growth toward independence and self-determination in the world beyond the classroom. For example, response journals both motivated and scaffolded less skilled writers as they shared personal stories and opinions with a teacher. Mailboxes encouraged students to write to each other. The value of mailboxes was underscored for the teachers when they "had this influx of lots of new students with this core group of old kids, and they immediately were talking back and forth [via the mailboxes]."

Relationships were the backbone of inclusion for Rhonda and Angie. They felt strongly that students with LD were more apt to thrive personally and academically as members of a protected community where the norm was helping one another and where teachers strove to empower students as self-directed individ\uals. When all students participated in the social interactions of the classroom and regarded such participation as expected, inclusion was successful.

Supports regarded as crucial for students with LD included teachers' attitudes (establishment of a climate of trust, acceptance of genuine effort at any skill level), the communicational aspect of writing (students wanting to tell teachers about themselves via their daily journals; students wanting to receive letters in their classroom mailboxes), social interaction (students asking one another for spelling help; low-achieving readers asking more skilled readers to read the teacher's response in their journals for them), and words placed around the classroom (word walls, theme walls, poetry on charts). Barriers to learning, in their view, included failing to set the "right tone," resulting in students' being too embarrassed or humiliated to "take a risk and begin trying things," or teachers who expected perfection in mechanical skills (e.g., spelling, handwriting).

Skills Development: Morning Message. A whole-group writing activity, Morning Message (cf. Manage, 2001), was key to the development of writing dispositions, skills, and conventions. Skills first encountered during Morning Message (e.g., capitalizing days of the week) were appropriated during other writing activities such as Writing Workshop, somewhat similar to the extension of board work skills to process writing tasks in Darlene and Monica's classroom.

In Morning Message, a student assumed the role of "author," who dictated a personal experience while a teacher scribed the oral text verbatim (i.e., replete with grammatical errors), omitting the conventions that transform oral text into written text (e.g., capitalization, punctuation). The role of "audience" (other class members) entailed making content suggestions and monitoring meaning and mechanics. The teacher's role, in addition to recording the emerging narrative on large chart paper, involved considering suggested revisions, gaining authorial consent for making changes, and prompting both audience and author when needed. As author and audience composed and edited the narrative, the teacher exploited opportunities for introducing new skills (e.g., capitalization, punctuation, grammar, editing), always maintaining a careful balance between discovery and direct instruction and sensitively scaffolding the participation of children who might have difficulty with the social and academic requirements of the activity (Cooper & Valli, 1996).

Process Writing Instruction. Rhonda and Angie called their process writing activity, which was scheduled for about an hour three or four times a week, Writing Workshop. There were two types of Writing Workshop tasks: assigned Writing Workshop projects, and more open-ended writing opportunities. Students might have several ongoing writing projects in their folders at any time. During an open-ended Writing Workshop lesson, students first reread their latest written story or composition and decided whether to continue with that story or begin a new one. Students who had finished a first draft signed up for a peer conference, followed by revision and a teacher conference. Mini-lessons were taught as needed to introduce or highlight essential skills. Final publication was accomplished by typing the story on a computer.

As students busied themselves writing, either individually or with a partner, the teachers either held writing conferences or moved about the room, checking on students' individual progress and making suggestions: "Are you working on any skill today? ... Good printing? Capitals and periods?" or "How did you know that? Where did you get this information? ... Make sure people know where you got the data." Similar to other teachers who struggle with how to adequately address individual students' needs in contexts that allow a high degree of student autonomy (Lipson et al., 2000), Angie and Rhonda found that knowing at any moment the exact point of development for each student, deciding where to push students' further development, and constantly holding the scope and sequence of the writing curriculum in mind to make on-the-spot decisions was an enormous task.

Lessons concerning new knowledge, such as an unfamiliar genre structure, were presented to the whole group and contained a large amount of teacher input. Other Writing Workshop lessons provided time for students to continue work on previous projects or to begin new ones; these were introduced with a general procedural reminder. Where Darlene and Monica cited "the book" as the backbone of their curriculum, Rhonda and Angie relied on a three-phase activity structure very like that found in Hillocks' (1984) description of the environmental mode of instruction. The first phase involved whole-group instruction with a large amount of teacher guidance in the form of modeling, think-alouds, and interactive negotiation. The second phase entailed smallgroup collaborative tasks, before progressing to the third phase, that of independent work. This sequence suggested a purposefulness about supporting and compelling the evolving independence of all students, but particularly students with LD, by providing opportunities for more knowledgeable others (Vygotsky, 1978) to serve as resources for learning.

The Persuasive Paragraph. In this writing lesson, one of a series on this genre, Rhonda and the students collaboratively constructed a paragraph explaining the group's opposition to wearing school uniforms. After constructing a topic sentence, students contributed ideas for supporting sentences. When appropriate, Rhonda accepted students' ideas with little or no revision.

Dustin, a student with LD, had difficulty with making a coherent contribution; nevertheless, Rhonda's expert scaffolding of his participation produced a usable sentence:

Rhonda: At lunchtime, food might

Dustin: Drop

Rhonda: Drop on your

Dustin: Uniform.

Rhonda: Uniform. And then you would have to what?

Dustin: Clean it.

This sentence could have appropriately completed the paragraph; however, in this classroom, such determination was not entirely the prerogative of the teacher. Typical of the nature of teacher- student interaction in this classroom, Rhonda and Angie encouraged student participation in many lesson aspects. For example, students helped to form the rules that guided their writing by brainstorming with the teachers about "what would make good quality text, what should we do when we are editing.... A lot of the stuff we do, we try and make sure the class is making this decision, that this would make quality writing." This type of student input was possible because a number of students had been members of this writing community for several years and knew how things were done.

Paul suggested a revision of Dustin's sentence to make it more succinct. Rhonda placed Paul's suggestion "on the table," as it were, for class consideration:

Paul: I have an idea for a sentence. When you get dirty you have to wash it, which would be a lot shorter....

Rhonda: Okay, they would, you get dirty and you have to wash it. Do you want to change it, class? Thumbs up if you want to change the dirty one. Remember how we talked about making it shorter, making our writing shorter, using less words but still having a paragraph? If you're not sure, fine. Thumbs up for 'Yeah, let's change it.' Thumbs down for 'No.'

Failing to sense a consensus, Rhonda opened the floor to Ned and Trey, who proposed alternative solutions:

Ned: Um, you could put you would have to keep buying new uniforms and they would get dirty so you would have to wash them over and over again....

Trey: Um, pants would cost $20 or so.

Rhonda: Well, yeah, we don't want to get too specific on how much, because we're not sure. "We would have to keep buying the same clothes and?"

Trey: "It would be way too much money."

Rhonda: "And it would be way too much money." We could add that one right there.

Rhonda acted as a billboard, publicizing students' ideas for consideration and positioning students as policymakers and problem solvers (O'Connor & Michaels, 1996). At this point, Rhonda noted a sense-making problem with Dustin's sentence and challenged him to clarify his idea.

Rhonda: Dustin, you might spill food on that shirt today. Wouldn't you wash it? Well, what's the difference between that and a uniform? ...

Dustin: The buttons might fall off when you're washing your clothes.

Rhonda: ... Oh, and if things start falling off and if they wear out? There's something we didn't think about....

Again, Rhonda molded Dustin's response to fit the paragraph. Throughout the lesson, she seamlessly scaffolded the conversation to meet a range of students' needs. Not all student utterances were filtered through her discourse, and, at times, students' utterances followed each other. This pattern of interaction requires participants' close attention to make a contribution. Sequence and timing are integral. Rhonda's discourse drew students, without risk, into participation in the substantive engagement (Nystrand & Gamoran, 1991) required for meaning exploration and problem solving.

Metaphorical Traces of a Relational Belief System

Rhonda and Angie believed that relational processes provided the means by which students became accepted and valued members of a learning community. They described their classroom as a place where "definitely, everything we do is building community." In their view, communication and community were two sides of the same coin, and the ability to write implied the ability to communicate.

Like Darlene and Monica, Rhonda and Angie expressed strong negative opinions about using disability to segregate students. They likened the idea to drawing a "line" between students in general and special education, a line that demarked a "massive chasm" suggestive of differences so deep that only specialized treatments could ad\dress the needs of students with LD. Rhonda was poignant: "I don't see any lines.... Truly, that question kind of, it's not really upset[ting], it's the word before you get to upset." Describing her imagined response if ever her own child became eligible for special education services, Angie was passionate: "My child, never, I don't care how severe, never will be labeled. Never. Unless they could guarantee me in blood that he'd always be in a full inclusion program. Never. Never. Ever, ever, ever." The "full inclusion" to which she referred must bear an exact likeness, of course, to their own program, which they felt was truly unique in its approach and degree of success with all class members.

Nonetheless, care was required, in their view, to ensure that special populations were truly included. These concerns were rooted, Rhonda suggested, in their own personal experiences. Rhonda recalled her college days: "It's kind of like when I went to State. I never stayed in the dorms, so I never felt a part of that community.... I would really feel left out." Angie recounted similar feelings during her early teaching career when, as a special education teacher, she felt "like an outcast, because in this building ... there were all these cliques of people, and I didn't belong in any one." Rhonda speculated that both being "scarred in our own way" led them to "feel that we have to do this"-that is, to create a sense of community in their classroom.

Rhonda and Angie described their classroom as "heaven for disadvantaged students" where they were "not destroyed" and in fact "protected so they can grow. That's what you do with roses, isn't it? Protect them through the storms?" This atmosphere was the "tone you set and keep." Thus, in their view, students with special needs were vulnerable, requiring a protected context in which they would find trust and safety. Such a context "empowered" students to "take risks and try things." The curriculum, then, entailed the development of tools (e.g., communication, writing skills, social relationships) that enabled students to be self-directed in the world.


In situations that are fraught with ambiguity, as teaching is, teachers draw on their beliefs to guide the multitude of on-the- spot decisions they make every teaching day (Kagan, 1992). This study sought to uncover the implicit teacher beliefs that might explain differences in instruction in what seemed to be, on the surface, similar contexts-that is, process writing in inclusive classrooms. This work extends understanding of teacher beliefs by (a) exploring a method of identifying beliefs by their defining characteristics (i.e., existential presumption, alternativity, affect, and episodic storage); (b) providing a strong example of how teachers' beliefs may form coherent packages or constellations that lead to instructional choices (e.g., how to teach process writing); and (c) for the specific case of teaching writing, uncovering how teachers nuance a specific instructional strategy-process writing- to conform to their underlying beliefs.

Both teaching teams espoused beliefs characterizing them as interventionist rather than pathognomonic teachers (Jordan & Stanovich, 2003). They were adamant that students with LD belong in inclusive contexts. Although disabilities were viewed as barriers to learning, the teachers contended that students without disabilities may also have "issues" that present significant barriers, so that disability was never the salient factor in planning or instruction.

Both teaching teams incorporated instructional practices evocative of Hillocks' (1984) environmental mode-that is, neither totally skills dominated nor totally natural learning. Their lessons centered on specific objectives and problems and used materials designed for both illustrating the principles to be learned and engaging the students in applying them. Nevertheless, within these commonalities, the two teams nuanced their instruction toward structural or relational bases. What resulted was "similar processes operating on disparate contents" (Abelson, 1979, p. 361)-in other words, writing instruction based on different principles of practice.

Thus, designations limited to interventionist and environmental fail to fully explain the nature of instruction in these classrooms. The nuanced display of shared beliefs (disability as barrier), practices (environmental approach), and self-efficacy (uniqueness of approach) are summarized as constellations of beliefs and practices in Figure 1. The four dotted ovals represent shared beliefs and practices. The rounded rectangles indicate nuances regarding student role, teacher role, and curriculum and instruction. The remaining, solid ovals provide details and connections.

Both teams viewed disabilities as barriers for students, striving to benefit from instruction targeted toward either meeting individual needs (Case 1) or building community (Case 2). The nature of the barrier was viewed as either a breakdown needing repair (Case 1) or a vulnerability requiring protection (Case 2).

Similar to curricularist teachers (Lipson et al., 2000), Darlene and Monica preferred structure and order and tended to move students in unison through the stages of the writing process. They believed strongly that a sequential curriculum that built skills consecutively and logically with strong oversight from teachers met their students' needs. The power of this approach for children who may not learn well through informal teaching methods (e.g., whole language) is in its provision for explicit and direct instruction in writing skills (Graham & Harris, 1997).

Similar to inquiry teachers (Lipson et al., 2000), Rhonda and Angie tended to focus on social interactions and activities, regarding the goals of writing to be self-expression, writing for authentic purposes, and communicating with others. The advantage of this approach is in its repudiation of deficit as basis for instruction, emphasis on collaboration, and the development of student agency in exercising choice (Graham & Harris, 1997).

Both teams of teachers saw themselves as highly committed, competent educators who pursued unique and successful approaches to inclusion. In the Ellison classroom, the idea was that successful students made progress to achieve high levels of accomplishment. In the Watkins classroom, students progressed to become competent and self-directed communicators prepared for life beyond the classroom.


Although the aim of this study was to contrast two cases, the researcher sought to avoid dichotomizing instructional patterns. Nevertheless, the use of only two cases might inadvertently predispose the researcher or the reader to reach such conclusions. Further investigation of a larger number of cases would help to avoid this tendency.

This study did not address the developmental needs of the students in the two classrooms. Because the classes were not entirely matched, small but potentially significant differences in age, achievement, and background (ethnicity and SES) might have influenced the teachers' views about teaching and learning. Also, differences in text genre (expository vs. persuasive) during the observed lessons might have encouraged teachers' preference of particular instructional strategies over others. Continued, fine- grained examination of context and teaching variables and their relationship to specific aspects of teacher beliefs would lead to better understanding of how best to instruct all students.

This study also did not examine student writing outcomes with a view to identifying the relationships between teacher beliefs and student achievement. At best, the influence of beliefs on achievement would be indirect (Muijs & Reynolds, 2002). However, were such a connection to be found, our understanding of what works for students with LD would be enhanced.

Directions for Future Research

Although this study did not seek to establish the advantage of one approach over another, it is premature to assume that these distinct instructional approaches had no differential effects on students' learning. Future research should continue to build on prior studies that have examined the implications of instructional decisions to identify their specific benefits and constraints (see, e.g., Englert, Berry, & Dunsmore, 2001; Gutirrez & Stone, 1997; Mariage, 1995; Morocco & Hindin, 2002).

FIGURE 1. Overlapping constellations of beliefs and practices of the Ellison team (top) and the Watkins team (bottom). Dotted ovals at center represent shared beliefs and practices.

Benefits and constraints might also be defined as opportunity-to- learn issues. Special educators would benefit from the development of a line of research that examines learning opportunities afforded by opportunity-to-learn standards implicated in accountability schemes (Banicky 2000; Boundy, 1999; Herman & Klein, 1997; Schwartz, 1995; Ysseldyke, Thurlow, & Shin, 1995). Understanding how teachers' beliefs influence the educational experiences of students with LD may help to redefine opportunity to learn in a more flexible and qualitative way, replacing "absolute measures of opportunity to learn" with "measures that are weighted by the need the student exhibits" (Ysseldyke et al., 1995) for students with disabilities.

Implications for Practice

Kagan (1992) commented that "the more one reads studies of teacher belief, the more strongly one suspects that this piebald form of personal knowledge lies at the very heart of teaching" (p. 85). The findings of this study have implications for teachers and educational researchers who work in or are involved with issues of inclusion.

Teacher educators need to address both personally held beliefs and field-based theories about learning and instruction in both preservice teacher preparation and inservice teacher education. When teachers embrace different epistemologies than those on wh\ich particular instructional interventions are based, well-researched instructional interventions may lose power and fail to achieve their intended outcomes.

At a minimum, teachers must be aware of the effects of their personally held beliefs on their instructional decisions and on their students' school experiences. Those who teach in Ellison-type classrooms should be challenged to consider how to facilitate learning through interactive problem solving and, given what we know about communities of learners and the social acquisition of knowledge, to support learners to use each other as resources. Teachers who teach in Watkins-type classrooms need to ensure that individual learning needs are appropriately recognized and met and to support the active engagement of students with LD in complex activities.

The outcomes of this study suggest that well-defined instructional strategies and models (e.g., process writing) may be variably interpreted, implemented, and perhaps even impaired by teachers' philosophical predilections. However, the alternate argument may also be made, particularly in this case-that teachers' stylistic preferences can be preserved when using recommended and effective instructional strategies. In other words, the process writing model may possess strength and flexibility sufficient to survive the use of variable instructional strategies.


1. Pseudonyms have been provided for all participants and schools.

2. Bruce's contributions to the group interview and his participation in the classroom activities observed were minimally influential and, thus, will be omitted from this analysis. As prime movers in their classroom, Darlene and Monica defined the terms of instruction and carried them out, and Bruce did not seem to challenge their hegemony. In fact, Bruce left the collaboration the following year due to issues of incompatibility.


Abelson, R. P. (1979). Differences between belief and knowledge systems. Cognitive Science, 3, 355-366.

Baker, S., Gersten, R., & Graham, S. (2003). Teaching expressive writing to students with learning disabilities: Research-based applications and examples. Journal of Learning Disabilities, 36, 109- 123.

Banicky, L. A. (2000). Opportunity to learn (Policy Brief Vol. 7). Retrieved February 16, 2003, from http://www.udel.edu/ chep / edbriefs / opplearn.pdf

Beach, S. A. (1994). Teacher's theories and classroom practice: Beliefs, knowledge, or context? Reading Psychology: An International Quarterly, 15, 189-196.

Bos, C. S. (1988). Process-oriented writing: Instructional implications for mildly handicapped students. Exceptional Children, 54, 521-527.

Boundy, K. B. (1999). Opportunity to learn and education reform: Ensuring access to effective education for all students. Retrieved September 24, 2003, from http://specialed .principals.org/discidea/ topdocs/peer/ opportunityib.html

Brantlinger, E. (1996). Influence of preservice teachers' beliefs about pupil achievement on attitudes toward inclusion. Teacher Education and Special Education, 19, 17-33.

Brophy, J. E., & Good, T. L. (1974). Teachers' communication of differential expectations for children's classroom performance: Some behavioral data. Journal of Educational Psychology, 61, 365-374.

Clark, C. M., & Peterson, P. L. (1986). Teachers' thought processes. In M. C. Wittrock (Ed.), Handbook of Research on Teaching (pp. 255-296). New York: Macmillan.

Cooper, D. H., & Valli, L. R. (1996). Designing classrooms for inclusion: Beyond management. In D. L. Speece & B. K. Keogh (Eds.), Research on classroom ecologies: Implications for inclusion of children with learning disabilities (pp. 143-162). Mahwah, NJ: Erlbaum.

Cresswell, J. W. (1994). Research design: Qualitative and quantitative approaches. Thousand Oaks, CA: Sage.

Dixon, C., & de la Cruz, E. (1992, December). Making visible the invisible: Exploring the construction of communities of writers in classrooms. Paper presented at the annual meeting of the National Reading Conference, San Antonio, TX.

Englert, C. S., Berry, R., & Dunsmore, K. (2001). A case study of the apprenticeship process: Another perspective on the apprentice and the scaffolding metaphor. Journal of Learning Disabilities, 34, 152-171.

Gersten, R., & Baker, S. (2001). Teaching expressive writing to students with learning disabilities: A meta-analysis [electronic version]. Elementary School Journal, 101, 251-272.

Graham, S., & Harris, K. R. (1997). Whole language and process writing: Does one approach fit all? In J. W. Lloyd, E. J. Kame'enui, & D. Chard (Eds.), Issues in educating students with disabilities (pp. 239-258). Mahwah, NJ: Erlbaum.

Graham, S., Harris, K. R., MacArthur, C., & Fink, B. (2002). Primary grade teachers' theoretical orientations concerning writing instruction: Construct validation and a nationwide survey. Contemporary Educational Psychology, 27, 147-166.

Graham, S., Harris, K. R., Fink, B., & MacArthur, C. A. (2001). Teacher efficacy in writing: A construct validation with primary grade teachers. Scientific Studies of Reading, 5, 177-202.

Gregg, N., & Mather, N. (2002). School is fun at recess: Informal analyses of written language for students with learning disabilities. Journal of Learning Disabilities, 35, 7-22.

Gutirrez, K. (1994). How talk, context, and script shape contexts for learning: A cross-case comparison of journal sharing. Linguistics and Education, 5, 335-365.

Gutirrez, K. D., & Stone, L. D. (1997). A cultural-historical view of learning and learning disabilities: Participating in a community of learners. Learning Disabilities Research & Practice, 12, 123-131.

Heath, S. B. (1983) Ways with words. Cambridge, UK: Cambridge University Press.

Herman, J. L., & Klein, D. C. D. (December). Assessing opportunity to learn: A California example (