The Stretch Model: including L2 student voices

The Stretch Model is a model of first year composition (FYC) that “stretches” the first semester's class over two semesters in order to help writing students who arrive at university with low test scores to succeed in their composition courses. Originally piloted in 1994 at Arizona State University (ASU), the Stretch Model of composition has been found to be effective in terms of retention and persistence of first language (L1) writers (e.g., Glau, 1996; 2007). It has become known at ASU and abroad as the Stretch Program. Since 1997, a separate track of the Stretch Program has been solely for second language (L2) writers, and L2 writing students are now roughly 17% of the program's population. Until fairly recently, there was no attempt to collect L2 data to support the Stretch Program's claims for effectiveness for the L2 population. As many universities across the nation have garnered inspiration for their own programs ("Stretch Award" 2016), and L2 writers have the potential to be in any composition class (Matsuda, Saenkhum, & Accardi, 2013), it is imperative to include the voices of L2 writers in the analysis of the Stretch Program. This study addresses the need for L2 writers' voices to be included in the analysis of the Stretch Program at Arizona State University. From the quantitative analysis of 64,085 students’ institutional data records, and qualitative analysis of 210 student surveys, findings include L2 writers have the highest rates of passing, but the lowest rates of persistence in the three-semester first year composition requirement when compared to Stretch L1 students and the traditional FYC population. Survey data also lends L2 student perceptions to complicate the main features of the Stretch Program including perceived writing improvement, having the same teacher and classmates for two semesters, and having more time to work on their writing. The quantitative findings are consistent with Snyder’s (2017a) analysis of the 2012 fall Stretch Program L1 and L2 cohorts.