BUFing for PAIRED LITERACY in DUAL LANGUAGE CLASSROOMS
In addition to the best practice of teaching Language and Content together Link, newer research has been emerging on paired literacy.
Paired Literacy can be defined by link
A strategic approach to biliteracy instruction where by students are learning to read, write, speak, and listen in two languages, without duplication of instruction, to promote reception and production of 2 or more languages. Paired literacy lessons include intentional design that promotes bilingualism and biliteracy. This approach honors the fact that students acquire literacy skills across languages and transfer skills and content learned in one language to the other Paired literacy, through its structure, supports students and teachers in meeting the shifts of the Common Core.
In the School District of Waukesha, one way we achieve paired literacy is through strategic and structured K-1 English Language Development and 2-8 BUFS (Biliteracy Unit Frameworks). K-8 Paired Literacy is required and is essential to the literacy and language experience of our emergent bilingual students in Dual Language. So strong must our conviction and commitment to paired literacy, that we must as a system protect the time scheduled for ELD and BUF as an absolute right for our students.
Paired Literacy has at its heart the vehicle that content offers to teach Literacy (Reading, Writing, Speaking and Listening). For this reason we must regard our ELD and BUF units as paired units of literacy rather than as "Science" or "Social Studies" units. A few years back Jamie Carmichael and I used to describe our BUFs as both a Blue and a Red car at a stop light. When Green, the Blue car of content and language speeds ahead during concept attainment, slowing over time so the Red car of literacy and language can take over. This visual metaphor should provide us a better understanding of the interplay and the purpose in our teaching.
This year, ELD and BUF units are realigning to reading and writing standards. Next summer we will have clear standards for literacy in our BUFS, as well as common assessment, and common rubrics. We have started with looking at equity across our units of CLM and BUFs and have attended to accessibility of our students to grade level standards. Based upon the work of Cheryl Urow and Karen Beeman, we have begun to align our writing standards first as the productive part of language is the assessable part, and will add the reading standards that support the productive writing. Teachers can creating targeted I CAN statements, common assessment products and common criteria for measuring student learning.
These teams are doing incredible work for our faculty, work which will be shared in 2019.
If you would like more access to research on PAIRED literacy...keep reading!
Effects of a Paired Literacy Program on Emerging Bilingual Children’s Biliteracy Outcomes in Third Grade
For the past four decades, there has been a contentious debate about how to best teach emerging bilingual children.1 For the most part, the discussion has focused on the language of instruction, and more specifically on whether children’s home language should be used for instruction (Francis, Lesaux, & August, 2006). In the United States, the most common non-English language used for instructional purposes is Spanish (Center for Applied Linguistics, 2008). Several reviews of literature and empirical studies have compared the performance of emerging bilingual students in bilingual education and English-only programs and concluded that home language literacy instruction promoted higher reading outcomes in English than all English instruction (e.g., August & Shanahan, 2006; Genesee, Lindholm-Leary, Saunders, & Christian, 2006; Greene, 1997; Rolstad, Mahoney, & Glass, 2005; Slavin & Cheung, 2005). Even when outcomes in English are similar for comparable students in transitional bilingual education and English-only programs (e.g., Slavin, Madden, Calderón, Chamberlain, & Hennessy, 2011), bilingual education is still advantageous because it helps students become bilingual and biliterate (Goldenberg, 2008, 2013). This outcome must be acknowledged, as bilingualism and biliteracy are superior academic, social, cultural, and cognitive outcomes than monoliteracy.
Despite the possible advantages of bilingual education, most emerging bilingual students in our nation’s schools are in English-medium programs (Goldenberg, 2008). Yet, their literacy achievement in English has tended to be consistently low on large-scale assessments such as the National Assessment of Educational Progress (NAEP). The underachievement of emerging bilingual children can be seen as an indication that existing literacy approaches are not adequately meeting these students’ language and literacy learning needs. As previously stated, language of instruction has been at the core of the bilingual education debate; however, the examination of the quality of instruction has been practically left out of the conversation. The present study aims to contribute to this conversation by examining the type of literacy instructional approach within the realm of bilingual education that might be most effective over time. More specifically, it examines the potential of “paired literacy,” an instructional approach that provides literacy instruction in the students’ home language and English, at different times during the day, from the beginning of formal schooling. Recent studies have investigated the impact of this simultaneous biliteracy approach, though more are needed (e.g., Baker et al., 2012; Berens, Kovelman, & Petitto, 2013). Such research is important to elucidate aspects or qualities of instruction to best meet the needs of the rapidly increasing number of emerging bilingual children who generally attend underresourced schools in working class neighborhoods (Goldenberg, 2013).
The purpose of this study is to examine whether the earlier and coordinated implementation of Spanish and English literacy instruction provides an academic advantage to students over a sequential literacy model. We hypothesize that paired literacy instruction has the potential to improve students’ literacy achievement in both languages and that these benefits are cumulative. This study included a cohort of Spanish-English emerging bilingual students who participated in a paired literacy model in Grades K-3 and students from the same schools who received sequential literacy instruction in Grades K-2 and were just beginning to participate in the paired literacy model in third grade.
The following research questions guided the study:
- Research Question 1: How do the biliteracy outcomes of students in a paired literacy model compare to those of students in a sequential literacy model in Grade 3 as measured by informal writing and reading assessments and what effect does it have?
- Research Question 2: How do the reading outcomes of students in a paired literacy model compare to those of students in a sequential literacy model in Grade 3 as measured by a state standardized assessment?
This section briefly describes the two types of literacy instruction for emerging bilingual children that are compared in this study: paired and sequential literacy.
The paired literacy model examined in this study consisted of literacy instruction provided in both Spanish and English from the beginning of schooling. Within this approach, English literacy instruction is not delayed while children are learning to write and read in the home language. In contrast to models that construct artificial barriers to separate Spanish and English literacy instruction, this paired literacy approach stresses that literacy in the two languages can “codevelop” (Barrera, 1983, p. 171). The goal of this paired literacy model is never to transition students to all English instruction. Rather, the goal is to promote bilingual and biliteracy development. In their review of research, Francis and colleagues (2006) found that paired literacy was the least commonly implemented biliteracy approach. It is also important to note that some paired literacy models documented in the literature start with literacy instruction in two languages, but they discontinue instruction in the non-English language within a few years as children acquire the skills and language proficiency to be successful in English-medium instruction (Francis et al., 2006; Slavin & Cheung, 2005).
The sequential literacy model that students in this study received consisted of initial literacy instruction in Spanish with the primary goal of transitioning students to English-only literacy instruction. As such, literacy instruction in Spanish and English was offered sequentially (Spanish literacy instruction ceased by first or second grade as students transitioned into English literacy instruction). Sequential literacy instruction is often part of transitional bilingual education models such as early-exit programs and some types of dual-language programs that provide initial literacy instruction in the students’ dominant language (e.g., 90/10 models). In sequential literacy programs, literacy instruction in the second language is introduced after the child acquires a certain level of proficiency in that language. Dual-language programs that use a sequential literacy model do not cease home language literacy instruction when literacy instruction in English is introduced.
The paired literacy approach that was the focus of this study is an essential instructional component of a biliteracy model called Literacy Squared. This model is designed to provide biliteracy instruction in Spanish and English beginning in kindergarten and continuing through fifth grade. It was developed to address the needs of Spanish-English simultaneous bilingual children. Simultaneous bilinguals are those children who are acquiring two languages from an early age (Baker, 2006). They represent the “new normal” (Escamilla et al., 2014), as they continue to be the largest language minority group in many areas of the country (Capp et al., 2005).
Table 1 presents a summary of the distinctive characteristics of the two types of literacy instruction that were compared in this study: paired and sequential literacy instruction. A more detailed description of these biliteracy models is provided in the “Method” section.
|
Table 1. Features of Paired Literacy (Literacy Squared) in the Treatment Group Versus Sequential Literacy in the Comparison Group.
Recent reviews of research have concluded that bilingual education programs can facilitate reading achievement in English and maintain bilingualism (e.g., Goldenberg, 2013; Lindholm-Leary & Genesee, 2010). These results are consistent with those reported in previous research reviews (e.g., Francis et al., 2006; Greene, 1997; Rolstad et al., 2005; Slavin & Cheung, 2005). For example, Slavin and Cheung (2005)conducted a meta-analysis of 17 studies that met their inclusion criteria. These 17 studies investigated program effectiveness (bilingual vs. English-only programs), 13 of which were conducted in the elementary grades. Ten studies of bilingual education programs that used paired literacy models in the elementary grades were identified for this meta-analysis (i.e., Alvarez, 1975; Campeau, Roberts, Oscar, Bowers, Austin, & Roberts, 1975 [four studies]; Cohen, 1975; Huzar, 1973; Maldonado, 1977; Plante, 1976; Ramirez, Pasta, Yuen, Billings, & Ramey, 1991, as cited in Slavin & Cheung, 2005). Slavin and Cheung (2005) defined paired literacy models as those in which “children were taught reading in both English and their native language at different times each day from the beginning of their schooling” (p. 272). The results of their review of research indicated that nine of 13 studies that focused on the elementary grades showed positive effects of bilingual programs on students’ English reading outcomes and the other four studies found no differences. The median effect size for these 13 studies was + 0.45 and the mean effect size was + 0.33.
Francis and colleagues (2006) evaluated the impact of bilingual education programs as compared with English-only programs using meta-analysis and estimated effect sizes for 20 studies that met their inclusion criteria. Of the 20 studies they reviewed, 14 studies were conducted in the elementary grades. Their review included nine studies of paired literacy models in which literacy instruction was provided in Spanish and English at different times of the day. These are the same paired literacy studies reviewed by Slavin and Cheung (2005), with the exception of the study by Ramirez and colleagues (1991), which Francis et al. (2006) categorized as a study comparing Transitional Bilingual Education programs (early-exit and late-exit) to English immersion programs.2 The meta-analytic results indicate a positive effect for bilingual education on English reading outcomes that is “moderate in size” (Francis et al., 2006, p. 397).
We now shift from meta-analyses of early research to discuss individual studies on the effectiveness of paired literacy instruction. Of interest to our investigation is that paired literacy was the instructional approach used in the bilingual education programs examined in these individual studies. Baker et al. (2012)conducted a longitudinal quasi-experimental study comparing the impact of a Spanish-English paired bilingual model with an English-only program on the English reading outcomes for first- to third-grade students attending high-poverty schools. Paired literacy instruction within the bilingual education program consisted of 90 min of daily Spanish reading instruction using the Houghton Mifflin Lectura program in Grades 1 to 2, and 60 min in Grade 3. Reading instruction in English (Houghton Mifflin curriculum) was provided daily for 30 min in first and second grades, and at least 60 min in third grade. Results of piecewise growth modeling analysis indicated that students receiving paired literacy instruction consistently obtained greater gains in English oral reading fluency across grade levels than students who received English-only literacy instruction. The estimated effect sizes in Grades 1 to 3 indicate a small to moderate effect (+ 0.33 to + 0.53) for the paired literacy model on oral reading fluency in English, which is consistent with the results of the meta-analyses mentioned above. Regression analysis showed that the two groups performed comparably on standardized measures of reading comprehension in first and third grade; in second grade, the paired literacy group performed better than students in the English-only group, and outcomes were statistically significant. They concluded that paired literacy instruction is feasible, and Spanish reading instruction does not delay the acquisition of reading in English.
In another quasi-experimental study, Berens et al. (2013) examined the effect of a 50:50 dual-language program that used paired literacy instruction and a 90:10 dual-language program that used a sequential literacy model in Grades 2 to 3 on Spanish and English reading and language proficiency outcomes. The performance of students who were “bilinguals from home” (home language was Spanish or English and Spanish) as well as that of “bilinguals from school” (home language was English-only; pp. 42-43) on standardized reading and language assessment tasks in both languages were compared between the two dual-language programs and to those of socio-economically matched monolingual English speaking children in English-only schools. All participating schools reported using a “whole-word approach” to teach reading in English rather than a phonics-based approach (p. 41). In the 50:50 model, children received daily and equal amounts of academic instructional time in both languages. Although the onset of formal reading instruction differed by home language, children were exposed to print in both languages beginning in kindergarten. In the 90:10 model, 90% of instructional time was in Spanish and 10% in English in kindergarten and first grade. As students progressed across grade levels, the amount of Spanish instruction decreased while instructional time in English increased by 10% every year, reaching a balance of 50% of instructional time in both languages by fifth grade.
The impact of the two types of literacy instruction (paired and sequential literacy) on reading and language learning was reported for each language. For the English tasks, “bilinguals from school” in both the paired literacy (50:50) and the sequential literacy (90:10) approaches outperformed “bilinguals from home” in all the tasks. This group of English-only home students from both literacy models also outperformed monolinguals in English-only schools on most of the tasks in English. For the Spanish tasks, “bilinguals from home” in the paired literacy model outperformed “bilinguals from school” in all but one task, and in the sequential literacy model, they outperformed “bilinguals from school” in all tasks. In addition, the combined group of bilingual students (bilinguals from home and bilinguals from school) in the paired literacy group performed better than bilinguals in the sequential literacy group on English Irregular Words and Passage Comprehension tasks. In contrast, bilinguals in the sequential literacy model performed better than bilinguals in the paired literacy approach on English Phonological Awareness and Reading Decoding tasks. Berens et al. (2013) concluded that paired literacy instruction seems to provide bilingual reading advantages on some reading outcomes (i.e., English Irregular Words and Passage Comprehension tasks) but not on others (English Phonological Awareness and Reading Decoding tasks).
As seen in the reviews of research and recent individual studies discussed above, in no case were English-only literacy approaches found more effective than paired literacy approaches. However, currently paired literacy programs are implemented least frequently; thus, more research-based evidence on the effects of this type of biliteracy instruction is needed. Additionally, there is little research comparing paired literacy to sequential literacy instruction, the most common approach to literacy instruction within bilingual education programs in the country.
This longitudinal study employed a quasi-experimental design to determine whether paired literacy instruction provides an academic advantage to emerging bilingual students over sequential literacy instruction. The sustained effects of Spanish and English writing and reading outcomes of students in the treatment group (paired literacy instruction in Grades K-3) were compared with biliteracy outcomes for students in the comparison group (sequential literacy in Grades K-2 and beginning participation in the paired literacy model in Grade 3).
Context
The 13 participating schools in the treatment group were implementing Literacy Squared, a comprehensive biliteracy framework made up of four main components: paired literacy instruction, professional development, assessment of biliteracy skills, and research. This study focuses on the paired literacy instruction component. It utilizes data collected from participants in the Salem-Keizer School District. In Salem-Keizer, prior to the implementation of Literacy Squared in 2009, emerging bilingual students identified as English language learners (ELLs) requiring language services received sequential literacy and English as a second or other language (ESOL) instruction.
Paired literacy instruction
In paired literacy instruction, a distinctive feature of Literacy Squared, Spanish literacy and literacy-based English language development (ELD) are purposefully connected to facilitate biliteracy acquisition (see Table 1). Spanish literacy instruction provides the foundation for literacy development. It uses teaching approaches and materials that are authentic to the Spanish language. Literacy-based ELD instruction builds on the literacy foundation developed in Spanish without being duplicative. It focuses on developing the language students need to successfully engage with and produce texts in English and provides opportunities for children to apply and practice literacy skills and knowledge acquired in Spanish to the learning of literacy in English. Language environments are purposefully connected in various ways such as through a shared theme, genre, literacy objectives, or the use of bilingual texts. Another distinctive feature of the Literacy Squared model is that both Spanish literacy and literacy-based ELD instruction are provided daily starting in kindergarten and continuing through fifth grade. Within the daily-allotted Spanish literacy and literacy-based ELD instructional blocks, equal time is spent in reading, writing, oracy, and metalanguage.
Additionally, the Literacy Squared model fosters authentic reading and writing in the two languages through coordinated biliteracy units that emphasize explicit and interactive teaching approaches and use high-quality texts (trade books) from a variety of genres. The literacy instructional approaches include modeled, shared, collaborative, and independent writing and reading; teacher-led small groups; and word work, while also attending to oracy and metalanguage instruction in both language environments. Writing instruction is intricately connected to reading, so as to promote its reciprocal relationship. Oracy instruction involves teaching the language skills necessary for students to achieve the literacy tasks through the explicit teaching of vocabulary, language structures, and opportunities for meaningful dialogue and language use. Direct attention to cross-language connections is provided via metalanguage, where students learn to analyze each language (its sounds, symbols, grammar, vocabulary, structures, and uses), as well as to understand the relationship between languages in the context of authentic writing and reading activities. Metalanguage development is promoted through cognate instruction, the use of bilingual books, the strategic use of language (e.g., preview-review, anchor charts that compare the two languages), and a researcher-developed cross-language strategy that validates translation as a constructive process that involves sophisticated analysis of language and communication across languages and cultures (Escamilla, Geisler, Hopewell, Sparrow, & Butvilofsky, 2009; Escamilla et al., 2014). The majority of time spent within each language environment is whole group, as instruction is planned to help students achieve grade level standards. Collaborative and small group instruction is intended to reinforce skills and strategies taught via whole group instruction.
Each year, teachers implementing Literacy Squared received professional development to support their implementation of the paired literacy model. The first year, professional development was focused on introducing the paired literacy model, as well as the idea of holistic biliteracy and authentic bilingual assessment to help guide instruction, with some basic introductions to Literacy Squared strategies and ideas for lesson planning. Teachers were also supported in creating daily schedules to ensure that they met the daily time allocations (for the daily time allocations of the paired literacy model, see Table 2). Over the 4 years of implementation for this longitudinal study, they also received more in-depth training on Literacy Squared strategies and demonstrations of biliteracy unit plans specifically designed to teach paired literacy to help support the planning of their instruction, and they were also provided with collaboration time to plan with other teachers.
|
Table 2. Daily Language Allocations for the Treatment Group (Paired Literacy Instruction).
Sequential literacy and ESOL instruction
Prior to the implementation of Literacy Squared, students in the comparison group were in sequential literacy programs. Sequential literacy consisted of Spanish instruction in kindergarten through first or second grade, after which literacy instruction was provided only in English, as compared with students in the Literacy Squared model who received instruction in both languages beginning in kindergarten. Students in the comparison group received 30 min of pull-out ESOL instruction via the “Carousel of IDEAS” curriculum (Ballard & Tighe, 2005), and they were grouped by language proficiency. Eight of the 13 schools had programs in which students participated in K-1 Spanish instruction until they were transitioned to all English instruction in second grade. Five of the 13 schools had an extra year of Spanish support, and students received instruction in Spanish until third grade, when they were transitioned to ESOL classrooms. Once students were transitioned to English-only programs, they received all of their instruction in English, and they continued to receive 30 min of pull-out ESL instruction.
In contrast to the Literacy Squared model, instruction in both the Spanish language arts and ESOL block of the sequential literacy model followed different programs, and instruction was not connected (see Table 1). Prior to the implementation of Literacy Squared, the district focused professional development on ELD. Also, the district introduced and provided extensive professional development on the implementation of their Comprehensive K-12 Literacy Model. This district-created literacy model is based on authentic reading and writing approaches similar to some of the teaching approaches emphasized in Literacy Squared (i.e., print-rich environment; modeled, shared, guided, and intentional independent reading and writing; word study; purposeful speaking and listening, and critical thinking).
In third grade, all students in the 13 schools who had previously received sequential literacy instruction began the paired literacy program, though when writing data were collected, they had been participating for fewer than 4 months, and when reading data were collected, they had been participating for about 8 months. Further, it is important to note that the first year of Literacy Squared implementation in these schools did not include high levels fidelity of implementation (FOI), as teachers were learning about its components and the framework supporting it. The professional development in the first year was primarily focused on orientation to the Literacy Squared framework and setting up teachers’ schedules so that they met the minimum time allocations for literacy instruction in both languages. They did receive training on a specific strategy (theDictado), as well as how to view their students’ language and literacy abilities holistically, and they saw some lesson demonstrations. However, based on coaches’ reports and teacher observations that we conducted in February of the first year, implementation was not in place at medium to high levels until the following year. Recognizing that the time allocations for sequential literacy instruction varied by school, Table 3 illustrates what these language time allocations looked like for students who had Spanish literacy instruction through first or second grade.
|
Table 3. Daily Language Allocations for Comparison Group (Sequential Literacy Instruction).
Setting and Participants
Participants in this study included 358 Spanish-English speaking third-grade students from 13 elementary schools within the Salem-Keizer school district in Salem, Oregon. All 13 schools participated as Literacy Squared research schools beginning in 2009-2010. School demographics are represented in Table 4. Although the schools have varying percentages of emerging bilingual learners, all students included in this study were identified as ELLs. Furthermore, with the exception of two schools, at least 71% of students receive free or reduced price lunch (FRL).
|
This study included 167 students in the treatment group who participated in paired literacy starting in kindergarten in 2009-2010 through third grade in 2012-2013. Students were included in this group if they had complete Spanish and English data sets for language proficiency, writing, and reading for four consecutive years (2010-2013) and were not on individualized educational plans (IEPs). Students on IEPs (n= 15) were omitted from the analysis, as we have no way of knowing the reasons for their IEP or the severity of their learning disability, and the majority of them were outliers in the data set. The comparison group (sequential literacy) consisted of 191 students from the same schools as the treatment group. Students in the comparison group were just beginning participation in the Literacy Squared program in 2009-2010, but had received sequential literacy from kindergarten through second grade. Students were included in the comparison group if they had complete Spanish and English data sets for language proficiency, writing, and reading in third grade (2009-2010) and were not on IEPs. Like the treatment group, students on IEPs (n = 19) were omitted from the analysis, as the reasons for their IEP or the severity of their learning disability were unknown to the research team, and the majority of them were outliers in the data set.
As illustrated in Table 5, students in both the treatment and the comparison group were similar with regard to gender and ethnicity. Additionally, students in both groups were comparable in their English language proficiency designation as measured by Oregon’s English Language Proficiency Assessment (ELPA; see Table 5).
|
It is important to note that students in both the treatment and control groups attended the same schools, and neither the neighborhoods nor the demographics changed throughout the duration of this study. While ideally we would have administered a pre-test to determine potential differences between the two groups, our relationship with this district did not begin until 2009 when the children in the comparison group were already in third grade. Therefore, pretesting to determine potential between group differences was not possible. However, no major differences were apparent between the treatment and control groups when we began the study, as they were comparable demographically and in language proficiency (see Table 5).
Although some teachers changed grade levels over those years, teacher turnover was low, and other than the fact that new teachers were hired to fill newly created positions due to program growth, teachers and teacher demographics were fairly consistent over the years. Of the 11 third-grade teachers in 2009-2010 in the comparison schools, nine of them (82%) were still in Literacy Squared schools in 2012-2013. Of those nine, one third were still teaching third grade, one third were Literacy Squared teachers at other grade levels, and one third had become English Language Acquisition (ELA) specialists, or Literacy Squared coaches, illustrating that teachers in both groups were fairly similar. In addition, teachers of both the treatment and comparison groups had similar years of teaching experience and educational levels. As shown in Table 6, about a third of teachers in both groups had been teaching for 1 to 3 years, and just over 50% had 7 or more years of teaching experience. About 80% of teachers in both groups reported that they held a master’s degree or were in the process of obtaining one (see Table 6). Thus, we are confident that differences in instruction were likely due to implementation of the treatment (Literacy Squared) and the training and coaching that accompanied it.
|
Data Sources
Writing
Spanish and English writing samples were collected annually from all students during the winter. Children were given 30 min to respond to a prompt provided by Literacy Squared. Writing sample data were first collected in Spanish, and 2 weeks later they were collected in English. Spanish and English prompts were similar to facilitate comparisons of language development, but different to avoid translation. The third-grade writing prompt in Spanish was Escribe lo mejor que te ha pasado en la escuela. Y ¿por qué piensas que fue lo mejor? (Write about the best thing that has happened to you in school. Why do you think it was the best?), and the English writing prompt was “Write about the best thing that has ever happened to you. Why was it the best thing?” The Literacy Squared Writing Rubric was used to score all samples (see Appendix). This rubric measures emerging bilingual students’ Spanish and English writing skills on a single rubric. It was designed using a holistic theory of bilingualism to honor the notion that emerging bilingual students’ biliterate writing skills and knowledge are distributed across languages, thus abilities are assessed in both languages. The rubric measures three writing constructs: content (0-10 points), structural elements (0-5 points), and spelling (0-6 points) with a total of 21 points. The rubric was purposefully designed to distribute weight differently to each construct. The ability to communicate a message, as measured in the content construct, carries more weight than spelling individual words correctly.
Each spring, participating teachers are trained on the Literacy Squared Rubric and they score a random selection of K-5 students’ Spanish and English samples. Inter-rater reliability is established during the training (for more information on the training and inter-rater reliability see Butvilofsky & Sparrow, 2012). Construct and content validity were also established on this rubric. To establish construct validity, the Literacy Squared Writing Rubric was compared to similar writing rubrics created for children learning to write in Spanish and English in U.S. contexts (e.g., 6+1 Trait Writing® rubric in English and Spanish, Culham, 2003; Authentic Literacy Assessment System [ALAS] in English and Spanish, García, 2005). Furthermore, the rubric was revised twice to establish construct validity. The Literacy Squared research team established content validity through expert review, continuously looking at and scoring writing samples, and revising the rubric accordingly.
Reading
Students’ biliterate reading skills and abilities were assessed annually in May using informal parallel measures: for Spanish, the Evaluación del desarrollo de lecto-escritura–segunda edición (EDL2; Celebration Press, 2007b), and for English, the Developmental Reading Assessment–Second Edition (DRA2; Celebration Press, 2007a). Third-grade reading was further examined using outcomes from the high-stakes state assessment, the Oregon Assessment of Knowledge and Skills (OAKS).
EDL2 and DRA2
The EDL2 and DRA2 are criterion-referenced assessments available in both Spanish and English, and they assist in documenting emerging bilingual students’ biliterate reading achievement over time. Their purpose is to identify students’ reading levels using a benchmark text in which reading accuracy, fluency, and comprehension are assessed. These tools are also informative in helping teachers design and deliver instruction for children. Scores on the EDL2 range from Level A (0) through Level 60, and scores on the DRA2 range from Level A (0) through Level 80, though increments between scores are uneven. The EDL2 and DRA2 have been studied and determined to be valid and reliable measures of reading in Spanish and English (Pearson Education, 2009).
OAKS
Data from the OAKS reading exam were collected as a formal reading measure to examine students’ third-grade reading outcomes. The OAKS is a criterion-referenced assessment based on the Oregon Content Standards, and students in Grades 3 to 5 take the reading exam. Scores are based on an achievement scale, and scale scores are converted to the following categories: Exceeds, Meets, Nearly Meets, Low, and Very Low. Third-grade students can take the reading OAKS in Spanish or English, as decided by their teacher or principal, though it is important to note that neither the state nor the district disaggregates the data by language.
FOI
Even though we have no data on FOI of the sequential literacy model, Literacy Squared was invited into the district because district administrators felt that their sequential model was not effective, and each school was implementing it in a different way. Once we began implementing Literacy Squared, we documented FOI via observation checklists, the collection of lesson plans, and student artifacts.
Data Analysis
Independent-samples t tests were conducted to compare means on the four measures (Spanish and English writing and reading) obtained by students in the treatment group with students in the comparison group. The independent variable for this study was third-grade students’ participation in paired literacy via Literacy Squared. Dependent variables were Spanish and English reading scores as measured by the EDL2 and DRA2 and Spanish and English writing scores as measured by the Literacy Squared Writing Rubric. In order to understand the practical group means differences, Cohen’s d was calculated to generate effect sizes for each assessment in each language.
In addition, frequencies were run to determine the percentage of students who met or exceeded the OAKS performance standards. It is important to note that effective beginning in the 2011-2012 school year, the state changed the achievement standards (cut scores) as a way to prepare for higher standards with the new Common Core State Standards (CCSS). Because each group had different cut scores, in order to make a comparison possible, we used students’ scale scores from their respective years, but used the 2009-2010 cut scores to determine which students met and exceeded the performance standards so that we could compare the outcomes of the treatment group with those of the comparison group.
We examined the difference in the biliteracy outcomes of third-grade students participating in paired literacy via Literacy Squared and those of students who were in a sequential literacy model prior to their participation in Literacy Squared, which did not begin until third grade. Table 7 presents mean writing and reading scores in both languages for both groups. Students who participated in the paired literacy model from kindergarten through third grade had higher mean scores than students who were in the sequential literacy model in all areas (Spanish and English writing and Spanish and English reading), as illustrated in Table 7.
|
Table 7. Mean Scores and Standard Deviations for Spanish and English Writing and Reading.
The significance of the differences between groups as well as the effect of the paired literacy model on students’ outcomes were examined by conducting independent-samples t tests and calculating Cohen’s dfor outcomes in Spanish and English writing and Spanish and English reading (Table 8). Differences between groups were statistically significant for each outcome measure in each language with moderate to large effect sizes (0.42 to 0.90). In Spanish writing, students participating in paired literacy had significantly higher scores (M = 12.92, SD = 2.6) than those in sequential literacy programs (M = 11.44, SD = 2.1); t(356) = 5.97, p = .000. Similarly, students in paired literacy had higher English writing scores (M = 12.58, SD = 2.3) than those in sequential literacy (M = 11.44, SD = 2.1); t(356) = 5.97, p = .000. When examining the results of reading, similar outcomes emerge, as in Spanish reading, the paired literacy group performed significantly higher (M = 33.87, SD = 6.6) than the sequential literacy group (M = 30.40, SD = 9.6); t(339) = 4.03, p = .000, and in English reading, the paired literacy group also outperformed (M = 29.56, SD = 8.3) the sequential literacy group (M = 24.43, SD = 10.1); t(355) = 5.27, p = .000. Furthermore, Spanish and English writing had effect sizes of 0.63 and 0.90, respectively, and Spanish and English reading had effect sizes of 0.42 and 0.56, respectively. These results suggest that paired literacy instruction does have an effect on student biliteracy outcomes and that when compared with students in sequential literacy models, those receiving paired literacy have significantly stronger biliteracy outcomes in writing and reading.
|
Table 8. Independent t-Test Results for Spanish and English Writing and Reading.
In examining OAKS reading outcomes, we calculated the percentage of students in each group who met or exceeded the performance standards (Table 9). A larger percentage of students in the paired literacy group (69%) met or exceeded the performance standards on the OAKS than in the sequential literacy group (49%). This difference is even greater when comparing the two groups’ outcomes in English only. Before Literacy Squared began in the district, almost all students took the assessment in English. Thus, the sequential literacy group outcomes represent all English scores. However, about 57% of students in the paired literacy group took the assessment in Spanish. While the district and the state do not disaggregate student OAKS data by language, when doing so for the paired literacy group, 79% of students who took the test in English met or exceeded the performance standard, which is in stark contrast to the 49% of students who did so in the sequential literacy group.
|
Table 9. Percent of Third-Grade Students Meeting or Exceeding OAKS Reading Performance Standards.
Overall, findings indicate that students who were in the paired literacy treatment group scored considerably higher in Spanish and English writing and reading than students who were in the sequential literacy comparison group on all measures. Furthermore, we found that there were significant differences between groups in Spanish writing, English writing, Spanish reading, and English reading, and a larger percentage of students in the paired literacy group met or exceeded the OAKS performance standards. These findings suggest that early and sustained paired literacy instruction leads to stronger literacy outcomes in both languages than sequential literacy instruction. Our findings also show that the simultaneous teaching of literacy in Spanish and English does not hinder the acquisition of English literacy, nor does it impede the development of Spanish literacy. Rather, it strengthens English literacy learning while fostering biliteracy.
A possible explanation to account for the stronger biliteracy outcomes in the paired literacy cohort is the connection of language environments where literacy instruction in two languages is purposefully coordinated. Students in the paired literacy model not only participated in paired literacy instruction from the beginning of kindergarten, but also literacy-based ELD was designed to build on what they were learning in Spanish literacy. As a result, it is likely that students were able to apply new literacy skills and knowledge from one language to the other (e.g., narrative writing skills, reading fluency, comprehension), which strengthened their biliteracy development. Additionally, the effect sizes observed in the paired literacy group are considerably larger than those of previous studies of paired literacy models (e.g., Baker et al., 2012; Francis et al., 2006; Slavin & Cheung, 2005). Again, a plausible explanation may be the emphasis of the Literacy Squared model on the early and sustained teaching of literacy in both languages and professional development focused on paired literacy instruction and biliteracy development. We concur with Francis and colleagues’ (2006) suggestion that emerging bilingual children “may learn to read best if taught in both their native language and in English from early in the process of formal schooling” (p. 397).
We also found that the effect of the paired literacy model was stronger in English than in Spanish, indicating that students can experience accelerated gains in English literacy while continuing to develop Spanish literacy. This is an important difference between paired literacy and sequential literacy models, which discontinue home language instruction once students have reached certain levels of literacy achievement.
There was a difference in the total time spent on daily literacy instruction across the two groups, with the paired literacy group spending 1 hr more on literacy instruction in first grade and 30 min more in second grade. While we recognize that this extra time on literacy instruction might have contributed to the outcome differences across the two groups, this finding reaffirms the importance of allocating sufficient time for literacy instruction in the two languages from the beginning of schooling. Our findings also suggest that even though the paired literacy cohort received less literacy instruction in English than in Spanish during the first 3 years of school (see Table 2), in third grade the effects of this model were stronger in English, particularly in English writing. This finding challenges the “time on task” argument (i.e., more instructional time in English will produce better English outcomes) that is often used in conversations about the most effective programs for emerging bilingual children. An explanation for this stronger effect on English writing may be that at the heart of Literacy Squared model is an enhanced attention to writing. In previous studies (Butvilofsky, 2010; Butvilofsky & Escamilla, 2012, 2013; Butvilofsky & Sparrow, 2012; Escamilla & Hopewell, 2010; Sparrow, Butvilofsky, & Escamilla, 2012; Sparrow, Butvilofsky, Escamilla, Hopewell, & Tolento, 2014), we found a strong correlation between writing and reading outcomes in both languages. This finding is supported by the notion that writing and reading are not only interconnected, but also that writing leads to reading development (Ferreiro, 2002). We view writing as an essential aspect of paired literacy instruction that is particularly beneficial for emerging bilingual students.
FOI data show that while each school had low, medium, and high implementers, all schools had more medium and high implementers than low implementers. This finding makes us confident that we have reasonable levels of FOI, and that paired literacy via Literacy Squared is being implemented in the classrooms. (For more on FOI in Literacy Squared, see Sparrow, 2010; Sparrow, Butvilofsky, Wiley, & Escamilla, 2012).
Because we did not begin our relationship with the Salem-Keizer school district until 2009, we do not have pre-test data on the comparison group participating in this study. However, as previously mentioned, because students attended the same schools, school demographics did not change, and 82% of the third-grade teachers who were implementing sequential literacy were still involved in the implementation of Literacy Squared in the 2012-2013 school year, we have no reason to believe that groups differed significantly before the change in programming.
The students in the comparison group started participating in Literacy Squared in the third grade. Even though at the point of data collection they had been participating in Literacy Squared for less than 1 year, the small amount of paired literacy instruction they received might have benefited their literacy development. We see this as a limitation because we cannot know with any certainty how the results would have differed had these students continued with sequential literacy instruction or been transitioned to all English instruction in third grade. However, if anything, we believe differences between groups would have been even larger had students not begun receiving paired literacy via Literacy Squared. Furthermore, with the exception of introducing the teachers to one strategy (theDictado), the first year of Literacy Squared implementation focused on introducing teachers to the Literacy Squared framework and beginning to shift their paradigms to a holistic bilingual viewpoint of bilingualism. As teachers were beginning to grasp the idea of paired literacy, there was a small shift in their instructional practice, but they were not yet implementing the program with fidelity.
Results from this study reaffirm previous evidence that teaching emerging bilingual students to read and write in their home language and in English promotes reading achievement in English, as well as biliteracy development. The study enhances the focus on language of instruction, and at the same time contributes to the conversation about qualities of instruction by comparing the impact that two fundamentally distinct approaches to literacy instruction—paired literacy and sequential literacy—had on students’ writing and reading outcomes over time. Essential features of the paired literacy approach that were examined in this study include (a) a holistic biliteracy framework that promotes equal attention to oracy, metalinguistic skills, writing, and reading in both languages for the development of biliteracy; (b) purposeful connection between Spanish and English literacy environments without duplicative instruction; (c) the use of teaching methods authentic to the Spanish language; and (d) text-based literacy instruction in English to promote English language and literacy development.
In addition, this paired literacy approach proposes a framework for the development of biliteracy units. Key elements of this framework for the development of biliteracy units include standards-conscious planning of coordinated literacy instruction in the two literacy environments, explicit instruction on cross-language connections, and the incorporation of explicit teaching and collaborative approaches while attending to a gradual release of responsibility model. The implementation of this holistic biliteracy framework alongside the intensive professional development and teacher support provided by the Literacy Squared team in collaboration with the district ELA directors and coordinators might have influenced the biliteracy gains observed in the paired literacy group. We are hopeful that additional research on long-term paired literacy approaches, combined with more opportunities for professional development, will result in improved and sustained children’s biliteracy outcomes.
Given the superior outcomes of paired literacy approaches noted in existing reviews of research and individual empirical studies, including the present one, we contend that schools that already provide sequential literacy instruction have the human and financial resources to implement paired literacy and should do so. The schools included in this study already had the bilingual teachers needed to implement a paired literacy model. All that was necessary to change their programs from a sequential approach to a paired literacy approach was leadership from their principals to change the scheduling, training, and coaching to focus on the new holistic biliteracy paradigm. Rather than confusing children, teaching them to read and write in two languages simultaneously from the start strengthens literacy learning in English while developing biliteracy. Building capacity in both languages provides an academic advantage, as shown by the results of this study. Thus, there is no need to delay the teaching of literacy in English until a strong foundation in Spanish literacy has been developed.
Lastly, we would like to suggest that it is not just the pairing of two languages for literacy instruction that contributed to the results of this study, but the intentionality of the pairing. That is, how the Spanish and English literacy environments are intentionally connected to maximize cross-language learning, biliteracy development, and the use of teaching methods authentic to each literacy environment. This is the innovation and what we see in this study as the potential of paired literacy instruction.
Existing evidence about the effectiveness of paired literacy instruction is by no means conclusive. Although research has emphasized the potential of this type of literacy instruction for emerging bilingual children, more research is needed to establish a more robust body of knowledge on this topic. Longitudinal studies of paired literacy are particularly needed. Despite empirical evidence showing the cumulative effects of bilingual instruction provided over long periods of time, a number of studies have stopped the examination of bilingual instructional approaches at Grade 3. Research on biliteracy that ends at Grade 3 underestimates the potential of bilingual education. With that in mind, we will continue the present study and follow the same students through fourth and fifth grade. Also, studies that utilize both quantitative and qualitative research methods are needed to determine not only the effects of paired literacy on students’ biliteracy outcomes, but also the role of various instructional aspects on those outcomes.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
The author(s) received no financial support for the research, authorship, and/or publication of this article.
| August, D., Shanahan, T. (2006). Developing literacy in second-language learners: Report of the national literacy panel on language-minority children and youth. Mahwah, NJ: Lawrence Erlbaum. Google Scholar | |
| Baker, C. (2006). Foundations of bilingual education and bilingualism. Clevedon, UK: Multilingual Matters. Google Scholar | |
| Baker, D. L., Park, Y., Baker, S. K., Basaraba, D. L., Kame’enui, E. J., Beck, C. T. (2012). Effects of a paired bilingual reading program and an English-only program on the reading performance of English learners in Grades 1-3. Journal of School Psychology, 50, 737-758. Google Scholar, Crossref, Medline, ISI | |
| Ballard, W., Tighe, P. (2005). Carousel of ideas: English language development program. Orange County, CA: Ballard & Tighe. Google Scholar | |
| Barrera, R. (1983). Bilingual reading in the primary grades: Some questions about questionable views and practices. In Herrera Escobedo, T. (Ed.), Early childhood bilingual education: A Hispanic perspective (pp. 164-184). New York, NY: Teachers College Press. Google Scholar | |
| Berens, M. S., Kovelman, I., Petitto, L. A. (2013). Should bilingual children learn reading in two languages at the same time or in sequence? Bilingual Research Journal, 36, 35-60. Google Scholar, Crossref, Medline | |
| Butvilofsky, S. A. (2010). Towards the development of a biliterate pedagogy: A case study of emerging biliterate writing (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (UMI No. 3419442) Google Scholar | |
| Butvilofsky, S. A., Escamilla, K. (2012). Literacy Squared® Phase II: Colorado Case Study Technical Report Year Two, 2010-2011. Retrieved from https://scholar.google.com/scholar?hl=en&q=Literacy+Squared+Phase+II%3A+Colorado+Case+Study+Technical+Report+Year+Two%2C+2010-2011&btnG=&as_sdt=1%2C6&as_sdtp= Google Scholar | |
| Butvilofsky, S. A., Escamilla, K. (2013). Literacy Squared® Phase II: Colorado Case Study Technical Report Year Three, 2011-2012. Retrieved from http://literacysquared.org/ABOUT%20Page/CO%20Case%20Study_Technical%20ReportLiteracy%20Squared%20Phase%20II%20Year%203_Final.pdf Google Scholar | |
| Butvilofsky, S. A., Sparrow, W. (2012). Training teachers to evaluate emerging bilingual students’ biliterate writing. Language and Education, 1, 1-21. Google Scholar | |
| Capp, R., Fix, M., Murray, J., Ost, J., Passel, J. S., Herwantoro, S. (2005). The new demography of America’s schools: Immigration and the No Child Left Behind Act. Washington, DC: Urban Institute. Retrieved from http://files.eric.ed.gov/fulltext/ED490924.pdf Google Scholar | |
| Celebration Press . (2007a). Developmental reading assessment. Parsippany, NJ: Author. Google Scholar | |
| Celebration Press . (2007b). Evaluación del desarrollo de la lectura [developmental reading assessmen]. Parsippany, NJ: Author. Google Scholar | |
| Center for Applied Linguistics . (2008). Languages of instruction in TWI programs. Retrieved from http://www.cal.org/twi/directory/language.htm Google Scholar | |
| Culham, R. (2003). 6+1 traits of writing: The complete guide, grades 3 and up. New York, NY: Scholastic. Google Scholar | |
| Escamilla, K., Geisler, D., Hopewell, S., Sparrow, W., Butvilofsky, S. (2009). Using writing to make cross-language connections from Spanish to English. In Rodriguez, C. (Ed.), Achieving literacy success with English language learners: Insights, assessment, and instruction (pp. 141-156). Worthington, OH: Reading Recovery Council of North America. Google Scholar | |
| Escamilla, K., Hopewell, S. (2010). Transitions to biliteracy: Creating positive academic trajectories for emerging bilinguals in the United States. In Petrovic, J. (Ed.), International perspectives on bilingual education: Policy, practice, and controversy (pp. 69-93). Charlotte, NC: Information Age. Google Scholar | |
| Escamilla, K., Hopewell, S., Butvilofsky, S., Sparrow, W., Soltero-González, L., Ruiz-Figueroa, O., Escamilla, M. (2014). Biliteracy from the start: Literacy squared in action. Philadelphia, PA: Caslon. Google Scholar | |
| Ferreiro, E. (2002). Relaciones de (in)dependencia entre oralidad y escritura (Vol. 21) [Relationships of (in)dependence between talk and writing]. Barcelona, Spain: Gedisa Editorial. Google Scholar | |
| Francis, D., Lesaux, N., August, D. (2006). Language of instruction. In August, D., Shanahan, T. (Eds.), Developing literacy in second language learners: Report of the National Literacy Panel on Language Minority Children and Youth (pp. 365-414). Mahwah, NJ: Lawrence Erlbaum. Google Scholar | |
| García, E. E. (2005). Teaching and learning in two languages. New York, NY: Teachers College Press. Google Scholar | |
| Genesee, F., Lindholm-Leary, K., Saunders, W. M., Christian, D. (2006). Educating English language learners. New York, NY: Cambridge University Press. Google Scholar, Crossref | |
| Goldenberg, C. (2008). Teaching English language learners: What the research does—and does not—say. American Educator, 32(2),8-44. Google Scholar | |
| Goldenberg, C. (2013). Unlocking the research on English learners: What we know—and don’t yet know—about effective instruction. American Educator, 37, 4-11. Google Scholar | |
| Greene, J. P. (1997). A meta-analysis of the Rossell and Baker review of bilingual education research. Bilingual Research Journal, 21(2/3), 103-122. Google Scholar, Crossref | |
| Lindholm-Leary, K., Genesee, F. (2010). Alternative educational programs for English learners. In Improving education for English learners: Research-based approaches (pp. 323-382). Sacramento: California Department of Education. Google Scholar | |
| Pearson Education . (2009). DRA2 K-8 Technical manual: Developmental Reading Assessment. Upper Saddle, NJ: Pearson. Retrieved from http://s7ondemand7.scene7.com/s7ondemand/brochure/flash_brochure.jsp?company=PearsonEducation&;sku=DRA2_TechMan&vc=instanceName=Pearson&config=DRA2_TechMan&zoomwidth=975&zoomheight=750Google Scholar | |
| Ramirez, J., Pasta, D. J., Yuen, S., Billings, D. K., Ramey, D. R. (1991, February). Final report: Longitudinal study of structural immersion strategy, early-exit, and late-exit transitional bilingual education programs for language-minority children (Report to the U.S. Department of Education). San Mateo, CA: Aguirre International. Google Scholar | |
| Rolstad, K., Mahoney, K., Glass, G. (2005). The big picture: A meta-analysis of program effectiveness research on English language learners. Educational Policy, 19, 572-594. Google Scholar, SAGE Journals, ISI | |
| Slavin, R. E., Cheung, A. (2005). A synthesis of research on language of reading instruction for English language learners. Review of Educational Research, 75, 247-284. Google Scholar, SAGE Journals, ISI | |
| Slavin, R. E., Madden, N., Calderón, M., Chamberlain, A., Hennessy, M. (2011). Reading and language outcomes of a multiyear randomized evaluation of transitional bilingual education. Educational Evaluation and Policy Analysis, 33, 47-58. Google Scholar, SAGE Journals, ISI | |
| Sparrow, W. (2010). An examination of the challenges associated with fidelity of implementation in an innovative biliteracy program (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (UMI No. 3408973) Google Scholar | |
| Sparrow, W., Butvilofsky, S., Escamilla, K. (2012). The evolution of biliterate writing development through simultaneous bilingual literacy instruction. In Bauer, E. B., Gort, M. (Eds.), Early biliteracy development: Exploring young learners’ use of their linguistic resources (pp. 157-181). New York, NY: Routledge. Google Scholar | |
| Sparrow, W., Butvilofsky, S., Escamilla, K., Hopewell, S., Tolento, T. (2014). Examining the longitudinal biliterate trajectory of emerging bilingual learners in a paired literacy instructional model. Bilingual Research Journal, 37, 24-42. Google Scholar, Crossref | |
| Sparrow, W., Butvilofsky, S., Wiley, E., Escamilla, K. (2012, April). Assessing fidelity of implementation of an instructional model: Targeting student trajectories toward biliteracy. Paper presented at the Annual Meeting of the American Educational Research Association, Vancouver, British Columbia, Canada. Google Scholar |
Author Biographies
Lucinda Soltero-González is an assistant professor of education in the division of Educational Equity and Cultural Diversity at the University of Colorado, Boulder. Her work has appeared in a variety of journals including the Bilingual Research Journal, International Journal of Bilingual Education and Bilingualism, Journal of Early Childhood Literacy, and Theory into Practice. Her research interests include the development of bilingualism and early biliteracy in young Spanish speaking children and biliteracy practices in U.S. schools.
Wendy Sparrow is a research associate at the University of Colorado, Boulder and a director of the Literacy Squared Project. Her research interests include simultaneous bilingual children’s biliteracy development and fidelity of implementation in bilingual/dual language programs and interventions. Her work has been published in a variety of journals, and she coauthored Biliteracy from the Start: Literacy Squared in Action(Caslon, 2014).
Sandra Butvilofsky is a research associate at the University of Colorado, Boulder. Her research interests include biliterate writing and reading development of simultaneous bilingual children in the elementary grades as well as classroom-based research. Sandra has over 18 years of experience in bilingual education, and has been involved in Literacy Squared since 2006, assisting in developing the program, conducting research, and providing professional development.
Kathy Escamilla is a professor in the Division of Educational Equity and Cultural Diversity at the University of Colorado, Boulder. Her research interests include looking at educational opportunities particularly in the development of biliteracy in Spanish speaking children in U.S. schools. Kathy is the lead author of Biliteracy from the Start: Literacy Squared in Action (Caslon, 2014).
Susan Hopewell is an assistant professor of education in the division of Educational Equity and Cultural Diversity at the University of Colorado, Boulder. Susan co-authored Biliteracy from the Start: Literacy Squared in Action (Caslon, 2014). She is interested in issues of language, culture, equity, and identity, especially as they impact—or are affected by—literacy practices.
No comments:
Post a Comment