Academic journal article Foreign Language Annals

Does Measuring L2 Utterance Fluency Equal Measuring Overall L2 Proficiency? Evidence from Five Languages

Academic journal article Foreign Language Annals

Does Measuring L2 Utterance Fluency Equal Measuring Overall L2 Proficiency? Evidence from Five Languages

Article excerpt

Introduction

Understanding the relationship between a second language (L2) learner's oral fluency and oral proficiency has important implications for L2 acquisition theories, pedagogical practices, high-stakes testing, and so on (de Jong, Groenhout, Schoonen, & Hulstijn, 2013; Ginther, Dimova, & Yang, 2010; Segalowitz, 2010; Skehan, 2009). For example, determining the relationship between these two may provide a better understanding of how the L2 system develops (Segalowitz, 2010; Skehan, 2009), which in turn may generate better teaching methods for improving oral L2 fluency and, as a result, help learners develop greater L2 proficiency (de Jong & Perfetti, 2011). Moreover, knowing how fluency development may vary according to proficiency from one L2 to another can allow educators to know what to expect and how to appropriately weigh various aspects of fluency as they promote the proficiency development of their learners. Finally, those interested in high-stakes testing might examine whether automatically measuring L2 fluency could replace more laborious and expensive manual rating systems for estimating L2 proficiency (Ginther et ah, 2010; Ushigusa, 2009). The purpose of the current study is to better understand the relationship between oral fluency and L2 proficiency, especially as it relates to different L1-L2 combinations and at different ACTFL proficiency levels and sublevels.

Literature Review

The term fluency, especially in an L2 context, has several definitions, most of which refer to fluidity or ease of speech (Guillot, 1999; Kormos, 2006; Riggenbach, 2000; Schmidt, 1992; Segalowitz, 2010). In its broad sense, as defined by Lennon (1990), L2 fluency is often synonymous with overall L2 proficiency. In this sense, when laypeople use the word fluency, they often mean proficiency (Guillot, 1999). More narrowly, fluency can refer to language that is produced fluidly and smoothly as one combines words and sentences in speech. In this sense, the term is used in most L2 research to complement accuracy (i.e., error-free language) and complexity (i.e., language spoken with situation-appropriate forms) as one of the three main aspects of proficiency (Towell, 2012). However, these three factors may influence each other differently depending on the speaker's L2 proficiency (Skehan, 2001, 2009; Taguchi, 2008). Segalowitz (2010) further divided this narrow sense of fluency into three subcategories: cognitive fluency, a speaker's ability to plan and execute L2 speech smoothly and easily; perceived fluency, a native speaker's subjective impression of a nonnative speaker's ease of producing speech; and utterance fluency, often called temporal fluency, the measurable features for the ease and smoothness of L2 speech, such as the speech rate, number of hesitations, and number and length of pauses. This study focused on utterance fluency and sought to understand its relationship with overall L2 proficiency.

Tavakoli and Skehan (2005) have suggested that utterance fluency has three aspects: speed fluency, the rate at which speech is delivered; breakdown fluency, disruptions in the ongoing flow of speech; and repair fluency, how often speakers make repairs, corrections, or false starts. Each of these aspects of L2 utterance fluency is determined by measuring specific acoustic features of speech (Derwing, Munro, Thomson, & Rossiter, 2009; Fillmore, 1979; Goldman- Eisler, 1961; Lennon, 2000). While many measures of L2 utterance fluency exist, some of the most typical are as follows: for speed fluency, syllables per second (Hilton, 2009), pruned syllables per second (i.e., the number of syllables minus hesitations or other disfluencies) (Rossiter, 2009), the number of runs or turns (Ginther et ah, 2010), and mean length of run in syllables (Ginther et ah, 2010) are most frequently considered. For breakdown fluency, most researchers examine the number of pauses (Trenchs-Parera, 2009) and length of pauses (Rossiter, 2009). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.