Enhancing the Quality of Biliteracy Instruction and Writing Outcomes in Spanish/English Bilingual Programs

• In Event: Dynamic Bilingual Literacy Instruction as Social Justice Practice: Educational Equity for Emergent Bilinguals

Abstract

I. Objectives
This paper presents results of a quasi-experimental study designed to examine the writing outcomes and biliterate writing strategies of students in two comparable one-way Spanish/English dual language schools in Texas (Table 1). Strategies used in the treatment school were based on principles of dynamic bilingualism and equity-oriented literacy instruction (Table 2). Research questions are:
1. What gains do emerging bilingual (EB) students in study schools make on biliterate writing assessments?
2. How do writing outcomes compare for students in the treatment and control schools?
3. What biliterate writing strategies are used by study students?

II. Perspectives
The biliteracy model in the treatment school is the flexible multiple biliteracy model (García, 2009). García’s model links literacy contexts across languages and makes explicit cross-language connections (Authors, 2014). The holistic bilingualism framework posits that bilinguals’ language systems are cohesive rather than mutually exclusive (Cenoz & Gorter, 2011; Grosjean, 1982).
III. Methods
This study employed a quasi-experimental mixed methods design (Creswell, 2009). Researchers collected writing sample data from two cohorts of students at the treatment (Blandon) and control (Ponder) schools three times over a two-year period (2015-2017) -- a pre-test and one post-test at the end of each academic year. Writing samples were analyzed using a researcher-developed biliterate writing rubric that yielded numeric composite writing scores (Authors, 2014). These data were used to calculate student writing growth. Multiple regression analysis was used to assess the statistical significance of differences in pre-test and post-test scores between schools, holding constant pre-test scores and clustering by standard errors of treatment versus control. Biliterate writing samples were examined qualitatively by analyzing a randomly selected set of 8 high and 8 low scoring samples using the writing rubric (Authors, 2014).

IV. Data sources
Data sources were Spanish and English writing samples scored through a ‘side-by-side’ method using the Biliteracy Writing Rubric to examine student writing holistically and simultaneously (Authors, 2014). The rubric has been determined to have content validity and inter-rater reliability (Butvilofsky & Sparrow, 2012). The qualitative portion of the rubric enabled analysis of students’ dynamic biliterate writing strategies at the discourse, sentence, word, and phonics level (Authors, 2014).

V. Results
There was no statistically significant difference in pre-test composite writing scores between schools (Table 3). Students in both schools made statistically significant growth in both languages (Table 4). Multiple regression analysis of post-test scores indicates statistically significant differences of 1-1.5 composite score points between schools for both cohorts in the Spanish post-test and for Cohort III in the English post-test (Table 5) indicating a strong association between treatment strategies and student writing outcomes. Qualitatively, lower performing students produced less text but all students demonstrated emerging biliteracy in writing and the utilization of cross-language strategies (Tables 6-9).

VI. Scholarly significance
Few mixed-methods longitudinal studies have examined growth in student writing in Spanish and English in the U.S. This study presents a novel way to assess and interpret trajectories to biliteracy in writing. Data from the rubrics can be used to inform equitable dynamic bilingual writing instruction.
 

Authors

• Kathy Escamilla, University of Colorado - Boulder

• Caitlin Fine, University of Colorado - Boulder

• Susan Hopewell, University of Colorado - Boulder