Getting to Level-Three and Beyond: the Relationship of Program Design, Initial Level, and Time-on-Task to Russian L-2 Outcomes in the Study Abroad Environment

Dan Davidson, Bryn Mawr College

Scholarly and policy-level interest in the conditions affecting second language gain within the study abroad environment continues to attract the attention of researchers (e.g., Freed, SSLA, 6-2004, et al), but relatively little empirical work has been undertaken to date on L-2 gain (esp. among the LCTL’s) at the superior level and above. The present study examines the roles of time-on-task, initial levels of proficiency, and program design on the attainment of superior-level proficiency in Russian during study abroad.

The National Flagship Language Initiative (NFLI) is a U.S. government-sponsored effort to support the study and acquisition of a group of strategically important world languages at the “superior-level” (ILR 3, 3+) and beyond. In 2003, ACTR was invited by NFLI to design and implement an overseas Flagship program in Russian, beginning in the 2004-5 academic year, for a group of no fewer than 8 full-time graduate students, who could demonstrate 2/2+ levels of proficiency in two skills at the time of application to the program. Approximately 50 US students applied for the program, which is based at St. Petersburg University, of whom 9 ultimately were accepted and funded for participation in the program. A similar applicant/participant ratio has obtained for the 2005-2006 Russian Flagship Year.

Participants underwent multiple batteries of proficiency-based testing (ACTFL, ACTFL/DLI, and European Framework/(TORFL) at the beginning, mid-point, and end of the year-long Flagship program. The present study analyzes outcomes measurements from the first year of the Flagship Program and presents comparisons with gains measurements in ACTR’s summer, semester, and academic year (non-Flagship) programs, which, for experimental purposes, may serve as a “control group” for comparing the effects of certain adjustments in program design that are unique to the Flagship Program.

While the overall number of Russian Flagship participants is still small (N=18), the overall patterns of language gain for the Flagship participants coincide with or exceed those predicted statistically for this level by ACTR’s on-going standard analysis of gain (“Predictors,” Brecht, Davidson, Ginsberg, 1995; Davidson 2004). In addition, the parallel use of U.S. and Russian (EU) testing instruments within the Flagship Program permits a degree of comparison of the behavior of these two widely used testing systems for pedagogical and research purposes.