"Reducing Achievement Gaps in Academic Writing for Latinos and English Learners in Grades 7–12"

AERA 2018 Annual Meeting: “The Dreams, Possibilities, and Necessity of Public Education”
April 13-17, 2018
New York
 
Title: "Reducing Achievement Gaps in Academic Writing for Latinos and English Learners in Grades 7–12"
Authors: Carol B. Olson, Huy Quoc Chung, Rachel Stumpf
 
Abstract

This study reports on findings from a randomized controlled trial of a professional development program that takes a cognitive strategies approach to teaching text-based, analytical writing to students, specifically Latinos and mainstreamed English learners, in a large, urban, low SES district where 77% of the students are designated as English learners.

This study is informed by cognitive (Flower & Hayes, 1981), socio-cognitive (Langer, 1991), and sociocultural theory (Englert, Marige, & Dunsmore, 2006). It specifically focuses on making visible to teachers and their students the thinking tools, or cognitive strategies, research indicates experienced readers (Tierney & Pearson, 1983) and writers (Graham & Perin, 2007) use to construct meaning from texts.

This study utilized a multi-site, clustered, randomized control trial where secondary schools were the sites, teachers were randomly assigned to treatment or control in the intervention, and one of each teacher’s classes was randomly selected to participate in the study. Approximately 3500 students in grades 7-12 were randomly assigned to the classes of 95 teachers (49 treatment and 46 control).

Data sources included existing school district records of student demographic information (gender, race/ethnicity, free and reduced-price lunch, and English learner status), and outcome measures including pre- and post-test on-demand academic writing assessments (AWA) administered by the teacher participants and passing rates from the California High School Exit Exam (CAHSEE). A gain-score regression analysis was applied to the AWA to determine growth over time and logistic regression was used to calculate the odds of passing the exam.

Year 1 treatment students gained 0.99 points more than control students (p<.001). For the first implementation year the effect is 0.96 on the AWA scale, which equals an effect size of .48. Across race/ethnic groups the largest positive effects are for Hispanics and Blacks (1.16 and 1.19, respectively). Program effects are positive and significant for all the language groups, with the very largest occurring for English Learners. For the second year, the effect is 1.30 on the AWA scale, which equals an effect size of .60. The largest gains are for Hispanics, and at 1.47 they are larger than the first year gains for this group. Significant positive effects are found for all four language groups with the largest occurring for RFEPS.
California High School Exit Exam (CAHSEE). In the first program year, 75.2% of controls and 83.7% of treatment students passed the test (∆=8.5%, p<.01). In Year two, 87.7% of treatment 10th graders passed the exam, compared to only 69.3% of the controls (∆=18.4%, p<.001). This difference is more than twice that achieved in Year 1.

Throughout the nation, literacy practitioners and researchers are looking for effective practices to help Latinos and ELs in secondary school improve their academic writing. The robust findings from this study yield promising results that appear to close the achievement gap in writing for Latinos and help ELs to gain ground by narrowing the achievement gap between ELs and their native English speaking peers.

​