The English Literacy Development of Bilingual Students
Within a Transition Whole-Language Curriculum

Stephen B. Kucer
Fordham University at Lincoln Center

Cecilia Silva
Texas Christian University

Abstract

This research project investigated the English literacy development of bilingual (English/Spanish) students beginning their formal transition into English literacy in a whole-language classroom. The curriculum consisted of four components: theme-based literacy activities, teacher reading, free reading, and free writing. Instruction was meaning-focused, and written language conventions were demonstrated through the use of contextualized mediations. Literacy growth was measured through pre/post-reading miscue and retelling analysis, holistic and analytic writing analysis, and spelling assessments. Analysis of the reading miscue data indicated statistically significant improvements in the students' ability to produce more meaningful sentences. Additionally, retelling analysis showed significant gains in the overall number of retelling units and matches produced by the students. Analytic writing analysis showed improvement in capitalization and spelling, as well as the number of words produced in a story. Spelling improvement was also noted in an analysis of words produced in isolation. Holistic analysis of the students' written stories did not indicate significant improvement, nor did the analytic analysis show an increase in the number of sentences or the conventional use of punctuation. Results suggest that literacy development may require differentiated mediation, i.e., some tasks may require more direct mediation than others.

 

Sections of the Article

 

Introduction

During the last decade, a number of educators have begun to question the effectiveness of whole-language instruction for multicultural and multilingual students (Delpit, 1995; Pearson, l989; Perez, l994; Reyes, l992). In many respects, their concerns are not unlike those expressed by educators working with monocultural and monolingual students; the issues of focus (meaning vs. conventions), instruction (direct vs. indirect), and learning (whole to part vs. part to whole) are all present.

Multicultural and multilingual students are frequently—though not always—students of color and from lower socioeconomic homes. As such, they may represent cultural and linguistic communities which traditionally have been assigned low status by the dominant American culture. The schools, as institutions representing mainstream interests and ideologies, oftentimes replicate or reinforce an assignment of low status to minority communities. Therefore, just as there is socioeconomic stratification in the larger society because of varying degrees of access to social and economic privilege, there is stratification of access to knowledge, language, and culture in the classroom setting.

Because of such stratification and because of the gap between the culture and language of the home and school for particular groups, many students of color are especially affected by the instruction in the classroom. Critics of whole-language instruction argue that because students of color from families with low socioeconomic status frequently fail to bring to school those discourse styles and language skills commonly brought by children from families of middle socioeconomic status, it is mandatory that schools explicitly teach these styles and skills. Failure to do so will result, the critics maintain, in students who are unable to join the club—mainstream American society—because they lack the credentials to do so. For middle-class students, family credentials—i.e., money, power, status, education—can more easily make up for any gap left by whole-language instruction.

The research investigating the impact of whole-language instruction on bilingual populations has produced mixed results. Furthermore, there is scant research on the impact of various curricula on students who are in the process of being transitioned from written discourse in one language to another (Gersten, 1996a; Gersten, 1996b; Gersten & Jimenez, 1996). Most existing studies of bilingual learners in whole-language contexts have examined literacy development in the students' home language, typically Spanish, or have examined English reading or writing development after the initial transition from home to school language has already occurred. Additionally, in these studies literacy has been narrowly defined and tended to focus only on reading or writing.

This study builds upon, and extends existing research on biliteracy teaching and learning. Specifically, it investigates the impact of a whole-language literacy curriculum on the initial English reading and writing development of bilingual (Spanish and English) and monoliterate (Spanish) students. In contrast to the subjects of previous studies, these students are just beginning their formal transition into English literacy. And, rather than examining a single process, this study asseses the development of both reading and writing in their various dimensions. The following two questions guided the investigation:

Theoretical and Research Framework

In the last 20 years, the whole-language paradigm for the teaching and learning of literacy has become both more popular and more controversial. Initial reservations were expressed by those within the traditional English literacy community, but significant challenges have also come from researchers and teachers of color.

A growing number of researchers have questioned the effectiveness of whole-language teaching for multicultural and bilingual students, especially given the purported lack of explicit and direct instruction in such classrooms. Delpit (1986, 1988, 1995) has argued that minority children, especially those living in poverty, are in need of explicit instruction in the conventions of standard written discourse. Because of the gap between the discourse of the home and of the school for many of these students, she claims the inductive learning of written language conventions is not possible. Direct access (via direct instruction) to the conventions of the power code of written language would "help children acquire the culture of power because it would give them access to a major medium of power, written language" (Teale, 1991, p. 541). To withhold such instruction, Delpit suggests, is to withhold access to the dominant American culture and all of the rewards this culture provides.

Delpit maintains that although all children must learn the conventions associated with "edited English," many minority children need to have such conventions highlighted in a more direct and explicit manner. Guided-discovery learning techniques and collaborative groups do not afford such students the support necessary for the learning of conventions because, as one student told Delpit, "We (students in the class) can sit around in groups all day talking to each other, and we're never going to learn to write `standard' English because nobody knows it" (1991, p. 543).

According to Delpit, explicit instruction need not be oppressive or result in student rejection of community values and identities if certain conditions are met. First, teachers need to explicitly validate and acknowledge the discourse brought to the school by the students. Because one's language is a critical part of one's identity, to reject the students' language may signify a rejection of the students. The teacher's role, therefore, is to add to the students' linguistic repertoire, not to eliminate or replace one mode of discourse with another. Secondly, teachers need to acknowledge that a conflict exists between home and school discourse. Frequently, students who appear unable to learn, are, in fact, unwilling to learn. Their behavior is a form of resistance to an institution that fails to validate their homes and their community (Ogbu, 1992). Finally, teachers need to openly discuss with students the unfair ranking of non-dominant discourse by society. This discussion needs to validate the linguistic integrity of various discourses and the injustice in the bias against individuals who speak a non-dominant form of English.

Perez (1994), Reyes (1992), and Reyes and Laliberty (1992) have made similar observations about whole-language classrooms for linguistically diverse student populations. In researching such classrooms, they have found teachers who are reluctant to mediate the students' interactions with print and students who fail to learn inductively the conventions of written language through mere exposure. Perez (1994) investigated the Spanish reading and writing development of 20 students in four whole-language classrooms (in kindergarten, first, second, and fourth grades) during the spring semester of the school year. In these classrooms, reading and writing activities focused on story meaning; the teaching and use of graphophonics and conventional spelling were not emphasized.

At the end of the semester, Perez found that, in general, the children had improved in their ability to construct meaning in reading and writing. Most students showed improvement in their production of grammatically and semantically acceptable sentences when reading orally. These students used both graphophonic and contextual cues to generate meaning from written language. Additionally, all children in the study demonstrated improved comprehension as assessed through the use of retellings. In writing, students who were comfortable using invented spelling produced longer and more varied texts. However, those children with the least developed knowledge of letter-sound relationships showed little improvement in their reading and writing. These children tended to have difficulty with meaning construction and to write less. According to Perez, the whole-language classrooms failed to directly and explicitly provide these learners with the linguistic knowledge necessary for successful reading and writing, and the children failed to acquire such knowledge inductively.

In a year-long study, Reyes (1991) investigated the impact of a process approach to literacy instruction using dialogue journals and literature logs in a sixth-grade whole-language classroom. In the dialogue journals, students and their teacher wrote to one another about topics or issues of interest, and students were allowed to write in either Spanish or English. In contrast, the children's literature logs were personal responses to stories that had been read, did not involve a response by the teacher, and were written in English. Similar to Perez, Reyes found that without explicit attention, students did not improve their control over written language conventions in Spanish or English. Rather, they tended to repeat their errors from one dialogue or log entry to the next.

In a follow-up study of fourth graders writing in Spanish, Reyes and Laliberty (1992) investigated the effects of a Spanish language writing program with fourth-grade students. In contrast to Reyes' previous study, in this classroom students were explicitly taught the conventions of written language through the use of mini-lessons as well as revision and editing conferences. Spelling was also taught on a daily basis. Although no pre and post data were formally analyzed, Reyes and Laliberty suggest that the Spanish-speaking children made significant gains in writing. These gains were in the development and organization of ideas as well as in the use of such conventions as punctuation, capitalization, and spelling. In addition, the growth of the Spanish-speaking students was comparable with that of the European-American, English-speaking students in the same classroom.

In a study examining the interaction between Spanish and English orthographies, Fashola, Drum, Mayer, and Kang (1996) compare the English spelling patterns of Spanish-speaking children with those of English-speaking students. The Spanish-speaking students were in the second, third, fifth, and sixth grades, spoke Spanish at home, and were classified by the schools as limited English-proficient. The English-speaking students were native speakers and not proficient in Spanish.

All students were given a list of 40 common English words to spell. In an examination of the spelling errors of Spanish-speaking students, two patterns were discovered. First, students produced misspellings by adjusting their perceptions of English phonology to fit within the Spanish phonological system. Sounds that exist in English but not in Spanish, such as the sound of oo in look and the b in cable, were mapped onto the closest Spanish sounds, such as o and v. Secondly, with sounds that exist in both English and Spanish, students frequently applied Spanish phonological and orthographic rules rather than English rules. For example, the English word hero was spelled jero, since the sound represented by h in English corresponds to the sound that is marked by j in Spanish.

In their discussion of instructional implications, the authors recommend that teachers working with Spanish-speaking students must explicitly point out those phonological or orthographic rules in English that differ from those in Spanish. Students must be given the necessary mediation so that they develop the strategic knowledge of where and when to apply English rules. They further indicate that if transitional spellers are not explicitly taught these rules, there is the danger that students will remain in a transitional stage and their "orthographic pidgin" will become a permanent structure in their English writing.

In summary, these studies sound a note of caution to advocates of whole-language instruction for all students. The research being reported replicates, in part, these existing studies but extends it to transitional students. Little is known about this population, the type of instruction they typically receive, or the impact of instruction on their English literacy development.

 

The Setting

Students

The research was conducted in a third-grade classroom in a large metropolitan area. The 26 students in the class were Mexican-American, bilingual, and from working class homes. Linguistically, most of the children entered kindergarten speaking Spanish predominantly and were in Spanish literacy programs through the second grade. Accompanying the Spanish literacy programs was instruction in oral English. In accordance with the school district's traditional bilingual education model, the purpose of the kindergarten through second-grade language and literacy programs was to provide the children with a foundation in spoken English and written Spanish so as to bridge instruction from the native language to English.

Because of their oral English and Spanish literacy abilities, the students were ready for formal transition into English literacy in third grade. English and Spanish abilities were assessed in several ways and involved several evaluators. The second-grade bilingual teacher identified students for transition based on her working knowledge of the children's academic and conversational English abilities (Cummins, 1981; Cummins & Swain, 1986). The students were also reading at "grade level" in the Spanish reading basal.

In addition to the above determiners of readiness for transition, at the beginning of the third grade, the third-grade teacher administered the Bilingual Syntax Measure II Test (BSM) (Burt, Dulay, & Hernandez-Chavez, 1978). The BSM is a language assessment instrument approved by the state of California to assist districts in identifying and placing second-language learners in appropriate classroom settings. This one-on-one test evaluates the students' use of English syntax and vocabulary, and places students in one of five English language proficiency levels (level one being the least proficient and level five the most proficient). In order to be transitioned, students had to receive a score of three or better.

The third-grade transition program integrated the bilingual students with the monolingual English students in all subjects except language arts. During language arts, all third-grade students in the school were grouped by reading ability. One teacher had the most proficient monolingual English-speaking students and a second teacher had the average monolingual English-speaking students. A third teacher taught the bilingual students who were to be transitioned into English literacy. The few monolingual English students reading one or more levels below grade were taught by the school's reading specialist. The result of this grouping was that almost all third-grade Spanish-speaking students were in the transition classroom.

Informal discussions with the first- and second-grade teachers, formal discussions with the principal and district curriculum consultant, and an examination of curricular materials indicated that the Spanish literacy programs experienced by the bilingual children in the first and second grades were fairly segmented and skills-oriented in nature. In fact, the principal and curriculum consultant noted that the school in general had just begun to move toward the wholistic curriculum suggested in the state's language arts framework (California State Department of Education, 1987). To a large extent, a basal reader, speller, and grammar book "framed" the instruction. Sound-symbol correspondence and vocabulary were explicitly taught in an isolated and sequenced manner, as were spelling words and punctuation, capitalization, and penmanship. Phonics was especially emphasized, given the fairly consistent relationship between letters and sounds in Spanish. Because the bilingual students in the school had historically progressed at a slower pace than the English-monolingual students, the principal supported a shift to a more wholistic form of instruction for the bilingual students with the expectation that such a curriculum would better serve this population.

There was no evidence to suggest that the students had experienced the use of thematic units or had encountered such instructional strategies as reader response groups or the use of contextual clues to understand unknown words before entering third grade. In addition, the students lacked experience with writing as a process (i.e., the use of writing and editing conferences to move a written piece of discourse from an initial draft to final publication). Therefore, in the third grade the students discussed in this study were encountering a curriculum with a different instructional focus, in addition to transitioning to formal literacy instruction in English.

Teacher

The teacher, originally from Colombia, was bilingual and biliterate in Spanish and English, and came from a bicultural, middle-class background. For 11 years she had taught elementary school, primarily in bilingual settings. At the time of the study, she was finishing her Ph.D. in a university language, literacy, and culture program that was whole-language oriented. An active participant in the study, she is the co-author of this article. The teacher believed that by providing literacy instruction in an environment rich in contextual support, she could facilitate learning for her second language students and thought that the use of thematic units provided such a context. She viewed thematic units as providing second language learners with the linguistic and extralinguistic support that would facilitate reading and writing development. She defined herself as a whole-language advocate and, being opposed to the explicit teaching of skills in isolation, felt that reading and writing strategies were best taught within contextualized situations.

She was experienced in working with transitional students and in the assessment of their English literacy behaviors—oral readings and retellings, written stories, spelling samples, and informal observations of the students as they engaged in classroom literacy activities. Based on her literacy assessments of the students and her ongoing interactions with them, the teacher viewed the students in this study as being overly concerned with language conventions and reluctant to take risks when engaged in English reading and writing. Therefore, her primary curricular goal was to develop a literacy program that would encourage students to actively engage and interact with print, to develop a range of strategies for constructing meanings via written discourse, and to develop conceptual and generalizable knowledge through the study of themes.

Researcher

For one academic year, the researcher was a participant and observer in the third-grade transitional whole-language classroom (Kucer, l991, 1992). He was monolingual in English, from an Eastern European-American middle-class background, and had known the teacher for six years. The researcher was in the classroom on Wednesday, Thursday, and Friday mornings during the entire language arts period (8:30-10:45 a.m.). As a participant observer, he watched and talked with the students as they went about their daily literacy activities. At no time during the year, however, did the researcher engage in any direct instruction of the children.

The Transition Literacy Curriculum and Instructional Strategies

The transition whole-language program was intended to provide a supportive instructional environment for the children as they moved into English literacy. The teacher developed a literacy curriculum for the students that consisted of four components: theme-based literacy activities, teacher reading, free reading, and free writing (Kucer, Silva, & Delgado-Larocco, 1995; Silva & Delgado-Lorocco, 1993; Silva & Kucer, 1997). The themes engaged the children in integrated activities related to a particular topic under study. Four thematic topics were experienced by the students during the year: Getting to Know About You, Me, and Others; Getting to Know About Amphibians and Reptiles; Getting to Know About Things That Scare Us; and Getting to Know About Plants and Seeds. The first topic was initiated by the teacher in an attempt to build a community of learners who were respectful and supportive of one another. The other theme topics were selected by the students.

The theme-related activities were designed to help students develop conceptual and generalizable knowledge about the topic and to promote literacy development. Lessons involved art, music, and math, as well as oral and written language. Materials came from the sciences, social sciences, and literature, were available in English and Spanish, and represented a range of discourse types (narrative, expository, poetic, dramatic) and resources (books, magazines, filmstrips, records, movies). Basal readers, spellers, grammar books, and other types of textbook materials and worksheets were absent from the curriculum.

Embedded within the thematic units were a number of learning events that were repeated throughout the year, regardless of the theme under study. On a regular basis, students experienced paired reading, reader response groups, compare/contrast activities, expert groups, learning logs, writing conferences, modified cloze procedures, and strategy wall charts. Although during these times students might express themselves in either English or Spanish, generally English was the language of choice and of instruction.

In terms of emphasis, the strategy wall charts, reading and writing conferences, paired reading, and the modified cloze procedure played particularly significant roles in the curriculum. In fact, they were the primary instructional strategies through which the teacher specifically and explicitly promoted student reading and writing development. As previously noted, because of her assessments and ongoing interactions with the students, the teacher believed that the students needed to develop a wider range of strategies so as to increase their effectiveness as users of written language. In particular, she wanted the students to learn a variety of ways to overcome "blocks to meaning" when they read and wrote (Kucer, l995). Blocks meant those things encountered by the students that halted their ongoing generation of meaning. In this class, blocks typically were encounters with: (a) something not recognized, known, or understood during reading; (b) difficulty "getting into" or engaging deeply with what was being read; (c) not knowing what to write next or how to express an idea within a piece of writing; and (d) difficulty spelling a word during writing.

During the first semester, the teacher collaborated with the students in developing a series of strategy wall charts. Each chart had a heading related to one of the four blocks frequently experienced by the students. The wall charts were introduced to the students over a two-month period, the teacher beginning the introduction of each chart by asking, "What can you do when—?" and recording student responses. For example, students were asked what they could do when they encountered something they did not know or understand during reading. Students brainstormed various strategies and the teacher listed these on the corresponding chart, including, at times, strategies of her own. Throughout the year, these charts were reviewed with the students and new strategies were added. Eventually, the teacher typed the problems and solutions on 8 l/2" by 11" paper and gave a copy to each student for easy reference. The teacher referred to the charts in a variety of instructional settings and students were encouraged to use them when reading and writing within the theme as well as when engaged in free reading and writing. The appendix illustrates how the charts appeared at the end of the academic year.

The use of response groups was a second mediational structure that was intended to help the students develop a wider range of interactions with what they were reading. Within the themes, students were regularly given the opportunity to select a book from a set of texts that focused on a particular thematic issue. After reading the book independently or with a partner, response groups were formed. In response groups, the students chorally read the book with the teacher a second time and then discussed the text from a variety of perspectives, including things they had difficulty understanding. As each difficulty was shared, the teacher and the students in the group discussed and "tried out" various solutions. These solutions were taken from the existing strategy wall charts or new solutions were developed and later added to the charts. When the problem was an unknown word, for instance, the teacher and students might reread the previous paragraph, read the paragraph following the unknown word, and discuss the relative importance of needing to identify the word. Regardless of what the problem happened to be, the teacher would demonstrate and "walk the students through" various solutions. Although "sounding it out" appeared on the strategy wall chart and was frequently used by the students, the children were encouraged to develop the use of contextual clues as well.

The teacher also frequently had the students work in pairs when reading self-selected theme books. In paired reading, the students chorally read from a single copy of the material and used strategies from the wall chart when they encountered difficulty. These dyads were largely formed by the students themselves based on having selected the same book from the text set.

Within every theme, students engaged in a variety of writing activities, for example, the creation of fictional stories about dinosaurs or an investigation of characteristics common to all reptiles. These texts were frequently published and, whether written as part of a thematic unit or during free writing, always involved the children in at least one conference. In small groups, each child read his or her draft and the students and teacher discussed what they liked about the piece and why. Conferences were concluded with suggestions for improving the draft. These suggestions tended to focus on the addition of new information to the text. Both types of evaluative comments were recorded by the teacher on a conference form and given to the author as a guide for making revisions.

On occasion, after revisions had been made following the initial conference, an editing conference would occur. In editing conferences, the students and the teacher revised such surface-level errors as punctuation, capitalization, and spelling. There was, however, no instructional emphasis on the mechanics of writing. Given the reluctance of the students to take risks, the teacher felt that students needed more experience as writers before focusing their attention on mechanics. She viewed a focus on mechanics at this point as counterproductive to the students' active involvement in the writing process. The teacher, therefore, served as the primary agent in the correction of surface-level problems before the texts were published. Following the editing, the teacher would either type the stories or have the students recopy them. The stories were then illustrated, front and back covers made, and bound. After students shared their published texts with the class, the books were placed in a plastic tub and were available during free reading.

The final mediational strategy used by the teacher was a modified cloze procedure. The teacher believed that this activity effectively highlighted cognitive processes that the students needed to learn and utilize during reading, that is, the use of contextual cues. The cloze activity was modified in a number of ways. First, words were deleted at points in the text where the teacher felt there was enough contextual information to support the generation of meaningful predictions on the part of the students. Also, students were not expected to generate the exact word that had been deleted. Rather, through the use of various context clues and strategies—rereading previous portions of text, reading on and returning, etc.—students generated meaningful responses. Finally, in the modified cloze activity, single words as well as groups of words were deleted from the text.

The modified cloze lessons, using material related to the theme under study, were presented to small groups. The typical pattern of instruction was as follows:

As noted earlier, cloze lessons were utilized by the teacher because of her belief that they highlighted cognitive processes needed by the students during reading. In describing the use of this mediational strategy we also note that as the year progressed the use of this strategy was challenged by the students. Interviews with the teacher and students, as reported by Kucer (1991, 1992), revealed a gap between the way the teacher and students understood the intent of this strategy. Although the teacher's intention was to provide students with a cognitively authentic strategy that would support them in using contextual cues, the students did not share this interpretation. The students, instead, discussed the purpose of the task in terms of learning new vocabulary, or believed that the purpose of the activity was solely to learn how to do the task in and of itself, or simply viewed the cloze strategies as preparing them to do similar tasks with more difficult texts in the future.

The second curricular component was teacher reading. During this time, the teacher read aloud short stories, trade books, and articles related to the theme. As she read, the teacher frequently responded to the text and encouraged the children to do so as well. On occasion, the teacher would also share with the children her particular reading behaviors. For instance, if she read a sentence that did not make sense to her, she would reread the sentence and discuss with the children why she had done so. Or, if she changed words in the text as she read, but had maintained the author's meaning, she would highlight this behavior, noting that this is something frequently done by good readers.

Following teacher reading, students engaged in free reading. Throughout the room were plastic tubs of paperback books and magazines on different topics, representing various discourse modes, and written in English and Spanish. The children selected their own reading material and were provided opportunities to share what they were reading. Although the children were never assigned book reports or any such activities to demonstrate that, in fact, particular materials had been read, a daily log was kept in which they recorded what had been read and the number of pages.

Free writing, in contrast to theme writing that focused on the topic under study, required students to select their own topics and to determine which texts to publish. These writings were done in bound notebooks and the teacher intended that the stories written might take several days to complete. In fact, throughout the year students were repeatedly encouraged to continue their writing from the previous day. However, the pattern that evolved was more similar to that of a diary, i.e., each day the students began a new story. Approximately once every six weeks, the students selected a text to publish and engaged in the publishing cycle previously described.

It is important to reiterate that there was no isolated instruction in written language conventions such as phonics, spelling, punctuation, or capitalization. Even within the reader response and writing conferences, these conventions seldom became the focus of instruction. Although some educators supportive of whole-language curricula have advocated the use of mini-conferences that focus on a particular language convention (Cazden, 1992), such instructional interventions were not part of this classroom.

 

Data Collection

Procedures

Throughout the year a variety of data were collected: field notes, curricular and instructional lesson plans, teacher interviews and reflections, oral readings and retellings, story writing, and spelling. The general purpose for the field notes and teacher interviews and reflections was to document the content and structure of the literacy curriculum as it unfolded throughout the year. During each classroom visit, the researcher hand-recorded descriptive field notes (Goetz & LeCompte, 1984; Spradley, 1980). Descriptive field notes were general accounts that described classroom events, including the lessons introduced, teacher directions, questions and responses to and from the students, and the content and focus of the instructional activities. Following the observations, the researcher held daily discussions with the teacher. Discussions focused on the purpose behind the activities observed and future curricular plans. These discussions were recorded as well.

The hand-recorded field notes were subsequently expanded and more fully developed into a narrative on a computer file. Expansions were written on the same day as the classroom visit so that the observations were of recent memory. In these expanded accounts, details and particulars not captured in the condensed notes were added. Analytic and interpretative insights, such as instructional patterns that emerged and the possible reasons for these patterns, were also noted. The previous description of the whole-language curriculum and primary mediational strategies is a product of this analysis of field notes.

In order to assess student literacy growth, all students in the class were asked to (a) orally read and retell a short story, (b) write a story about an exciting experience, and (c) spell 57 words from the third-grade speller. The same readings, writing topic, and spelling words were used in the pre/post assessments and were all in English. There were a number of reasons for using these measurements in a pre/post test format. First, as documented earlier, much of the existing research on bilingual literacy development has used similar or identical measures. Given that this study is conceived as an extension, in part, of existing research, it was desirable to incorporate some of the same measures. Secondly, the school in which the study was conducted was interested in pre/post measures that might be used to document the impact of the curriculum on "at risk" students. In addition, given that spelling was not explicitly taught in the classroom under discussion, there was the desire by school administrators to have specific evidence on spelling growth in the words being explicitly taught to the students (largely monolingual European-Americans) in the other two third-grade classrooms. Finally, reading miscue and retelling analysis and holistic and analytic writing measures are frequently employed to assess student growth in wholistic classrooms.

Oral readings and retellings followed the procedures involved in miscue analysis (Goodman, Watson, & Burke, l987). Each child independently read a story of approximately 500 words that was selected from graded stories in Readings for Taping (Goodman & Burke, l972), and that was slightly above the child's ability level. As noted by Goodman, Watson, and Burke (l987), "the material should be difficult enough to challenge readers' strategies, but not so difficult that they are unable to continue independently" (p. 38). In addition, the reading must generate at least 25 miscues—i.e., deviations from the print, such as omissions, substitutions, and insertions. All oral readings and retellings were audiotaped.

Before reading began, the students were informed that they would probably encounter some "things" not known or recognized and that they were to do the best they could, but to continue reading as no assistance would be provided. Following the reading of the story, the students would be asked to retell everything they could remember without referring to the text. The first story read by the students was at the 1.0 grade level, and each student continued to read successively more difficult texts until a story generated approximately 25 miscues. The story that first produced 25 miscues was used to assess the students' reading behavior. The average grade level of the texts read was 2.5.

Writing and spelling data were collected by asking students to independently "write a story about an exciting thing that has happened to you." Students were given as much writing time as necessary but received no assistance. In addition, a test of 57 words, three randomly selected from each unit in the school's third-grade speller (Cramer, Hammond, Lim, Prejza, & Triplett, 1981), was administered in the fall and spring of the year. Each word was read to the students, used in a sentence, and then read again. After all 57 words had been presented, particular words were repeated as requested by the students.

 

Data Analysis

In order to assess student literacy growth, pre/post readings, retellings, writings, and spelling assessments were compared and contrasted. These data were formally analyzed through use of miscue and retelling analysis, holistic and analytic writing measures, and scoring of words spelled conventionally. Before analyzing the data, all identifying information—student name and date—was removed.

Miscue analysis evaluates the degree to which students utilize the interacting semantic, syntactic, and graphophonic systems of language when reading (Goodman, Watson, & Burke, l987). An evaluator experienced in miscue analysis listened to the taped readings of all students and marked all miscues. Each sentence as finally read was then judged as to whether or not it was syntactically acceptable and made sense within the context of the story (language sense). Following the initial marking and coding of miscues, a second experienced evaluator reviewed the assessments. When discrepancies arose, they were discussed and resolved by the two evaluators. The Friedman Two-Way ANOVA Analysis by Ranks was used to evaluate the significance of change between all pre and post reading miscue scores.

Retellings were analyzed through the inductive development of the retelling taxonomy shown in Table 1 (Kucer & Silva, 1996). All retellings were first transcribed and, along with the texts read, segmented into clauses (a group of words containing a subject and a verb). The retold clauses were then analyzed in terms of their relationship to the text clauses. This correlation of clauses resulted in the taxonomic categories of match, substitution, addition, summary, conflict, and rearrangement. Using the taxonomy, two evaluators independently classified each retold clause as a match, substitution, addition, summary, or conflict. The last category in the taxonomy, rearrangement, examined the degree to which the order of the story retold matched the order of the story read. Following the classification of clauses, the evaluators held a conference on the results and resolved any differences in classification. The Friedman Two-Way ANOVA Analysis by Ranks was used to evaluate the significance of change between retelling pre and post scores.

Table 1
Retelling Taxonomy

Category	Description
Match	The idea expressed in the retelling matches an idea in the text. The surface structure may be different, but the deep structure is the same.
Substitution	The idea expressed in the retelling is a substitution for an idea in the text. A substitution represents a semantically acceptable modification of an idea expressed in the text.
Addition	The idea expressed in the retelling is not found in the text but is semantically acceptable. An addition may represent implicit text meanings or an inference is feasible.
Summary	At least two separate ideas in the text are condensed into one general idea in the retelling.
Conflict	The idea expressed in the retelling contradicts an idea expressed in the text.
Rearrangement	The order of the ideas and their interrelationships expressed in the retelling are at variance with the order of the ideas and their interrelationships expressed in the text.

 

The Friedman Two-Way ANOVA Analysis by Ranks was also used to evaluate the significance of change in both holistic and analytic writing measures. Growth in writing was assessed through the development and use of a 4-point holistic rubric (l low - 4 high) that focused on organization, development, appropriate vocabulary, and well-formed syntactic structures. (Table 2.) Pre and post texts were typed and given to three readers experienced with holistic scoring techniques. Before scoring began, the readers were introduced to the rubric, provided with anchors for each level, and given experience using the rubric on sample texts. After scoring was complete, texts receiving three identical scores or two identical scores and a third score no more than one level different were identified. The majority score was then assigned to these texts (e.g., a text receiving scores of 2, 2, and 3 was scored as a 2). Readers held a conference on those few texts not meeting this criteria and reached agreement as to the final score assigned.

The pre/post stories were analytically evaluated for changes in length (number of words and sentences), spelling, capitalization, and punctuation. In the analysis of story sentence length and appropriate use of capitals and periods, a sentence was defined as a free-standing syntactic structure consisting of no more than two independent clauses and their dependents linked by appropriate connectives.

Table 2

Holistic Writing Rubric

Score	Description
4	A 4 is a superior composition, although it may have flaws. It does all or most of the following: a. Clearly responds to the prompt b. Develops a coherent and well-organized narrative that includes a correct sequence of events c. Uses specific details to develop ideas d. Displays appropriate vocabulary and syntax
3	A 3 is a good composition, but not clearly outstanding. It does all or most of the following: a. Clearly responds to the prompt b. Develops a coherent narrative although it may not be as organized as 4 and may occasionally wander c. Uses specific details, although the narrative may not be as fully developed as 4 d. Uses appropriate vocabulary and syntax but may not be quite as proficient as 4
2	A 2 is an adequate composition, with some or most of the following characteristics: a. Attempts to respond to the prompt but may wander partially or completely off the topic b. Is not developed into a coherent narrative, and may lack organization c. Lacks supporting detail and development of ideas d. Lacks variety in vocabulary and syntax
1	A 1 is an inadequate composition and will compound the problems of 2, with most of the following characteristics: a. May not respond to the prompt at all or may go completely off the topic b. Lacks focus or has no coherent organization c. Has few supporting details or a complete lack of development d. Lacks variety in vocabulary and syntax

 

As well as evaluating spelling growth in the context of a written story, the analysis assessed spelling development through a comparison of words spelled conventionally on the pre/ post spelling test. A paired t-test was used to determine the significance of change in words spelled conventionally in pre/ post tests.

 

Results and Interpretation

Reading and Comprehension Development

Table 3 summarizes the pre/post miscue analysis scores and indicates a significant improvement in language sense. Students produced more sentences that were both syntactically and semantically acceptable within the context of the story in their post readings than in their pre readings. What these data demonstrate is that students improved their ability to coordinate the use of contextual cues and graphophonics to produce meaningful sentences. Given the teacher's focus on the use of context during the reading process, as demonstrated in the reading strategy wall chart, it is not surprising that there was improvement in this area. This finding is similar to that of Perez (1994), who also found that whole-language curricula improved student use of a range of meaning-making strategies. Perez's study, however, focused on Spanish (first language) rather than transitional English (second language) literacy.

Table 3

Means and Friedman Exact Tests for Reading Miscue Analysis

		M		Chi-Square
	Fall		Spring
Language Sense	38.90		52.72	11.0000***
***p < .001

 

The results of the analysis of pre- and post- reading retellings are presented in Table 4. As indicated, the overall number of clauses retold by the students doubled significant gain. Similarly, there was a significant difference in the number of matches produced by the students in the fall (4.67) and those produced in the spring (9.67). Although changes in the other retelling categories were not significant, it is worth noting that there was a trend toward significance.

Table 4
Means and Friedman Exact Tests for Reading Retellings

Retelling Category		M		Chi-Square
	Fall		Spring
Retelling Units	8.67		20.22	5.4444*
Matches	4.67		9.67	4.5000*
Substitutions	0.11		0.33	1.0000
Additions	2.00		4.67	3.5714
Summaries	0.67		1.11	1.0000
Conflicts	1.22		0.78	0.6667
Rearrangements	0.89		0.044	1.0000
*p < .05

 

Once again, the increase in matching clauses is similar to the increase in comprehension found by Perez. However, retellings were not part of the curriculum in this classroom. Nor were students engaged in answering recall questions or encouraged to take a literal stance towards the texts they read. Rather, students had opportunities to participate in response groups and the use of response strategies. These strategies encouraged students in developing a wider range of stances to written discourse—e.g., what did you like, what did you learn, what was confusing—that were not in any way similar to a retelling.

Writing Development

As indicated in Table 5, student writing grew by .45 on a 4-point holistic scale. This growth was not statistically significant, although there was a trend toward significance. The lack of statistical significance is noteworthy, given the frequency with which students drafted, conferred, revised, and published their writings. However, as previously discussed, the focus of the writing conferences was fairly general in nature. Students discussed what was liked and suggested ideas for the improvement of the piece. Suggestions for improvement tended to emphasize the addition of information (idea development).

Table 5
Means and Friedman Exact Tests for Holistic Writing Scores

		M		Chi-Square
	Fall		Spring
Holistic Writing Scores	1.30		1.75	3.6000

 

Given this lack of growth, it appears that the students needed more direct and explicit instruction, perhaps through the use of focused mini-lessons as employed by Reyes and Laliberty (1992). In such lessons, one particular dimension of writing—e.g., idea development, vocabulary, or syntax—is highlighted for those students requiring additional support. In these mini-lessons, not only is the dimension discussed, but students are explicitly shown how to use it as well. Following the lesson, students apply what they have learned to drafts they are revising for publication. One example that points to the need for more focused mini-lessons appears in the researcher's field notes in which he comments on a student who was requested to add more information about an animal that was in his story. The student returned to his seat and stared at his paper, not knowing where to add the new information. After a few minutes, he shrugged his shoulders and simply attached the new idea to the last sentence in the story. However, the placement of the information at this point in the text made little sense.

The pre/post stories were also analytically evaluated for changes in length (number of words and sentences), spelling, capitalization, and punctuation (periods). The increase from fall to spring scores was significant in story word length, spelling, and capitalization (Table 6). The gains found in the use of capitals and conventional spelling are in contrast to the general lack of growth in conventions reported by Perez (1994) and Reyes (1991).

Once again, in light of the dimensions of literacy that were and were not highlighted in this classroom, these findings are rather interesting. Although in writing conferences there was usually the suggestion to add new information, the students' focus was on the addition of new sentences. Expansion of existing sentences through the inclusion of adjectives or adverbs was never discussed. Additionally, following student revision, it was usually the teacher who corrected surface-level errors. She would then either type the texts on the computer or have students recopy them for publication.

Table 6
Means and Friedman Exact Tests for Analytic Writing Scores

Writing		M		Chi-Square
	Fall		Spring
Number of Sentences	3.26		4.16	1.0000
Story Word Length	35.68		57.37	4.2632*
Spelling	31.32		49.79	6.3684*
Capitalization	2.05		2.84	4.0000*
Periods	1.63		2.32	2.5714
*p < .05

 

Finally, growth in spelling was also assessed through the use of pre and post tests. As indicated in Table 7, the increase from fall to spring scores was significant. It should be remembered that the students did not engage in spelling instruction as it was presented in the other two third grades in the school. That is, students did not go through the usual routine of studying twenty or so new words each week, with words introduced on Monday and tested on Friday. However, the students did engage in writing as part of the thematic units, as well as free writing activities, throughout the year and had to deal with spelling in these contexts.

Table 7

Means and Paired t-Tests for Spelling Word List

		M		Chi-Square
	Fall		Spring
Spelling	21.26		32.84	4.43**
**p < .001

 

The increase of words spelled correctly that were taken from the spelling basal is all the more interesting when the increase is compared to the spelling growth in the school's two other ability-based classrooms (above average and average monolingual English readers). The same pre/post test was given to these classroom as well. On average, all classrooms—above, average, and transition—increased the number of words spelled correctly by approximately 12 words. This similarity is striking, given that the students in the other two classrooms received direct instruction in spelling words that were tested through the use of the spelling basal from which the same words were selected.

 

Summary and Conclusions

This research investigated the English literacy development of bilingual (English/Spanish) students beginning their formal transition into English literacy in a whole-language classroom. Framing the investigation was a question regarding the effectiveness of transitional literacy programs in which reading and writing activities are meaning-focused and students are expected to learn written language conventions through social interactions and contextualized mediations.

The English literacy transition students in this study, like the Spanish literacy students in Perez's research (1994), demonstrated improved reading abilities and comprehension as assessed through the use of miscue and retelling analysis. Writing results were mixed: the students improved in story word length, spelling (story and word lists), and capitalization. Analytic writing evaluation, however, did not indicate an improvement in the number of sentences produced in a story or in punctuation (periods), nor did a holistic rubric show an improvement in overall writing abilities. Perez (1994), Reyes (1991), and Reyes and Laliberty (1992) report gains in the use of conventions and/or idea development and organization only with the use of explicit instruction.

The improvement in spelling on both measures (story and word lists), and the fact that students improved on the word lists as much as the monolingual English literacy classrooms in which spelling was explicitly taught, sheds light on the concerns expressed by Fashola, Drum, Mayer, and Kang (1996). They argue that without explicit instruction, transitional students may develop an "orthographic pidgin" that will become a permanent structure in their English writing. Although this study examined group effects within a single academic year, student spelling did not appear to be fossilized. That is, student development of the conventional English spelling system continued to be dynamic and changed throughout the year.

The varied impact of the curriculum on student literacy development, as well as the varied relationship of these findings to those of related studies, highlights the complex nature of learning and instruction for bilingual learners. Clearly, it is overly simplistic to assert that significant growth in certain dimensions of literacy will not occur without direct instruction. Conventional spelling, use of capitals, story word length, and literal comprehension were dimensions not directly targeted by the teacher in this classroom. Nevertheless, growth was significant in these areas. Similarly, it is overly simplistic to assert that students will improve their literacy abilities by being immersed in a garden of print; that is, students will improve in their reading and writing due to the maturation process, regardless of instruction. Students in this study did not demonstrate growth in all areas, e.g., holistic writing, story sentence length, or use of periods.

The students' uneven performance across tasks suggests the need for what we term differentiated mediation and Cazden has called "instructional detours" (1992). In differentiated mediation, students would continue to be engaged in authentic and meaningful literacy activities as found in this transition classroom. However, when it is determined that a child is encountering repeated difficulty with a particular dimension of written language, focused instructional events would be developed that explicitly teach over time the matter in which the child is experiencing difficulty. In these lessons, not only is the child shown what to do, but also how it is accomplished. Such mediation is differentiated because not all children receive the instruction, only those in need. Additionally, the degree of "explicitness" varies depending on the child.

It is beyond the scope of this paper to explore in any depth the concept of differentiated mediation. However, a variety of instructional frameworks have been developed that can provide such mediation: demonstrating, contingency management, feedback, instructing, mini-lessons, questioning, instructional detours and cognitive structuring (Cazden, 1992; Gallimore & Tharp, 1990; Kucer, 2001; Pressley, Rankin, & Yokoi, 1996; Pressley et al., l989; Roller, 1996). Roller, in particular, provides a general outline of how instruction of this type might appear. In her work with struggling readers, she documents how the teachers initially cues the use of a particular reading strategy by telling the student what to do and when to do it. In time, the teacher provides a more general prompt—e.g., Did that make sense?—and allows the child to select the appropriate strategy to apply. Once the student's independent and appropriate use of various strategies has been established, the teacher verbally reinforces their use by commenting on what the student has done. Finally, the teacher supports the development of metacognitive awareness by asking the student to identify and explain the strategies employed.

Further research in the use and impact of differential mediation is particularly needed at this time when more school districts are mandating that teachers move toward a literacy curriculum that focuses on the direct teaching of isolated skills. Rather than assuming that each and every literacy skill must be directly taught to all students, we need to expand our understandings regarding the various types of mediations that may and may not be necessary to support students in general and second language learners in particular.

This study examined literacy growth within one transitional classroom. Although the data generated in this study add a new dimension to our understanding of literacy development for transitional students, comparative data with other types of classrooms are necessary to further explore the different ways in which students interact with literacy curricula.

 

References

Burt, M., Dulay, H., & Hernandez-Chavez, E. (1978). Bilingual syntax measure II. San Francisco, CA: Harcourt Brace Jovanovich.

California State Department of Education. (1987). English-language arts framework. Sacramento, CA: Author.

Cazden, C. (1992). Whole-language plus: Essays on literacy in the United States and New Zealand. New York: Teachers College Press.

Cramer, R., Hammond, W. D., Lim, L., Prejza, J., & Triplett, D. (1981). Spelling: Words and skills. Glenview, IL: Scott Foresman.

Cummins, J. (1981). Four misconceptions about language proficiency in bilingual education. NABE Journal, 3, 31-45.

Cummins, J., & Swain, M. (1986). Bilingualism in education. New York: Longman.

Delpit, L. (1986). Skills and other dilemmas of a progressive Black educator. Harvard Educational Review, 56, 379-385.

Delpit, L. (1988). The silenced dialogue: Power and pedagogy in educating other people's children. Harvard Educational Review, 58, 280-298.

Delpit, L. (1995). Other people's children: Cultural conflict in the classroom. New York: New Press.

Edelsky, C. (1986). Writing in a bilingual program: Habia una vez. Norwood, NJ: Ablex.

Fashola, O., Drum, P., Mayer, R., & Kang, S. (1996). A cognitive theory of orthographic transitioning: Predictable errors in how Spanish-speaking children spell English words. American Educational Research Journal, 33, 825-843.

Gallimore, R. & Tharp, R. (1990). Teaching mind in society: Teaching, schooling, and literate discourse. In L. Moll (Ed.), Vygotsky and education. Cambridge, England: Cambridge Press, 175-205.

Gersten, R., (1996a). The language-minority student in transition: Contemporary instructional research. The Elementary School Journal, 96, 217-220.

Gersten, R. (1996b). Literacy instruction for language-minority students: The transition years. The Elementary School Journal, 96, 227-244.

Gersten, R., & Jimenez, R. (Eds.) (1996). The language minority student in transition. The Elementary School Journal, 96 (3).

Goetz, J. P., & LeCompte, M. D. (1984). Ethnography and qualitative design in educational research. Orlando, FL: Academic Press.

Goodman, Y., & Burke, C. (l972). Readings for taping. New York: Macmillan.

Goodman, Y., Watson, D., & Burke, C. (l987). Reading miscue inventory. New York: Owens.

Kucer, S.B (1991). Authenticity as the basis of instruction. Language Arts, 68, 532-540.

Kucer, S. B. (1992). Six bilingual Mexican-American students' and their teacher's interpretations of cloze literacy lessons. The Elementary School Journal, 92, 555-570.

Kucer, S.B. (1995), Guiding bilingual students "through" the literacy process. Language Arts, 72, 20-29.

Kucer, S. B. (2001). Dimensions of literacy: A conceptual base for teaching reading and writing in school settings. Mahwah, NJ: Erlbaum.

Kucer, S. B., Silva, C., & Delgado-Larocco, E. (1995). Curricular conversations: Themes in multilingual and monolingual classrooms. York, ME: Stenhouse.

Kucer, S. B., & Silva, C. (1996). Reader response strategies in a whole language bilingual classroom and their effects on student comprehension. Paper presented at the annual conference of the National Teachers of English, Chicago, IL.

Ogbu, J. (1992). Understanding cultural diversity and learning. Educational Researcher, 21, 5-14.

Pearson, P. D. (l989). Reading the whole-language movement. The Elementary School Journal, 90, 23l-24l.

Perez, B. (l994). Spanish literacy development: A descriptive study of four bilingual whole-language classrooms. Journal of Reading Behavior, 26, 75-94.

Pressley, M., Goodchild, F., Fleet, J., Zajchowski, R., & Evans, E. (l989). The challenges of classroom strategy instruction. The Elementary School Journal, 89, 30l-342.

Pressley, M., Rankin, J., & Yokoi, L. (1996). A survey of instructional practices of outstanding primary-level literacy teachers. The Elementary School Journal, 96, 363-384.

Reyes, M. de la Luz. (1991). A process approach to literacy using dialogue journals and literature logs with second language learners. Research in the Teaching of English, 25, 219-313.

Reyes, M. de la Luz. (l992). Challenging venerable assumptions: Literacy instruction for linguistically different students. Harvard Educational Review, 62, 427-444.

Reyes, M. de la Luz, & Laliberty, E. (l992). A teacher's "Pied Piper" effect on young authors. Education and Urban Society, 24, 263-278.

Roller, C. (1996). Variability not disability: Struggling readers in a workshop classroom. Newark, DE: IRA.

Silva, C., & Delgado-Larocco, E. (1993). Facilitating learning through interconnections: A concept approach to core literature units. Language Arts, 70, 469-474.

Silva, C., & Kucer, S. B. (1997). Expanding curricular conversations through unification, diversity, and access. Language Arts, 74, 26-32.

Spradley, J. (1980). Participant observation. New York: Holt, Rinehart, & Winston.

Teale, W. (1991). A conversation with Lisa Delpit. Language Arts, 68, 541-547.

 

Appendix
Wall Charts