Linguistics Colloquium Series: A Talk by Dr. Ellen J. Serafini
Analyzing use of second language (L2) knowledge over time by adult learners of Spanish at three levels of linguistic experience
Past Event
Thursday, March 27, 2014 7:30 PM to 9:00 PM EDT
#3300
Second language (L2) development in adults who were exposed to the L2 after puberty is highly variable over time both within the same learner and between different learners (R. Ellis, 2008). This stands in stark contrast to the uniform native or native-like outcomes observed in children acquiring their first, or additional languages (e.g., Birdsong, 2005; DeKeyser, 2000; Johnson & Newport, 1989; Long, 2007). In order to adequately study the gradual, variable, and dynamic nature of gaining proficiency in another language as an adult (Selinker, 1972, 1992), longitudinal designs and robust statistical methods are needed to capture multiple levels of change over time. However, this is a current methodological shortcoming as the majority of research in second language acquisition (SLA) continues to be cross-sectional in nature and employ traditional inferential statistical analyses to analyze development (Ortega & Iberri-Shea, 2005). In light of these gaps, the current study aims to provide both a cross-sectional and longitudinal analysis of intraindividual and interindividual variation in L2 learners' use of Spanish grammar at three points during and after a learning period. Eighty-seven adult learners in their second, fourth, and sixth semesters of university Spanish instruction completed three versions of an elicited oral imitation task assessing knowledge of ten linguistic structures in Spanish, adapted from Bowles (2011). Tests were distributed using a counter-balanced, split-block design at the beginning and end of a semester of instruction and one month later (3.5 months total). Both traditional inferential statistical analyses and innovative hierarchical linear modeling techniques revealed significant improvement and retention for all proficiency groups over time on both measures as well as significant performance differences between groups at each point in time but no significant differences were found in mean rate of growth over time. Nonetheless, inspection of individual growth trajectories showed highly variable development for lower proficiency learners in comparison to less variable developmental trajectories for learners with the most linguistic experience. Results contribute to building a comprehensive view of change in interlanguage systems over time in learners with increasing levels of experience.
Hosted by GMU Linguistics Club.
Sponsored by Linguistics Program, Department of English.