Individual and interactional speech fluency in L2 English from a problem-solving perspective : A mixed-methods approach

Abstract

This study examines second language (L2) speech fluency and resources for maintaining it based on monologue and dialogue data. The study extends previous L2 speech fluency research by making three theoretical–methodological contributions: adopting a mixed-methods approach (including quantitative and qualitative analyses), developing a problem-solving framework for fluency analysis that incorporates fluency resources (stalling mechanisms and communication strategies), and measuring and assessing fluency based on L2 dialogue data. The study is situated within L2 oral proficiency and second language acquisition research and has implications for L2 teaching and assessment. The study is based on four original publications (Articles I–IV). In the articles, Finnish learners’ monologue and dialogue speech fluency is examined with two data sets including L2 English samples from ninth graders, upper secondary school students, and university students of English. L1 Finnish samples are also examined in Article III. Fluency is analyzed using measures related to temporal (individual) fluency, interactional fluency, and problem-solving mechanisms, with an added perceived fluency perspective (raters’ assessments) in Article IV. The results demonstrate that the qualitative analysis complements the quantitative analysis in various ways, notably highlighting the role of stalling mechanisms in compensating for local disfluencies. Based on the analyses, more advanced students also speak more fluently. L2 fluency is also linked to some extent to L1 fluency. In L2 dialogue data, (between-)turn pause length in particular distinguishes between learners at different levels. Based on the study, interactional fluency can be viewed as consisting of two main components: temporal fluency (turn pauses) and cohesive devices (other-repetitions and collaborative completions). The study demonstrates the benefits of triangulating quantitative and qualitative methods in fluency research and applying a problem-solving perspective to fluency analysis. The approach to interactional fluency that is developed in the study can also be used as a basis for further research. Based on the results, the influence of L1 speaking style on L2 fluency and the joint nature of fluency in L2 dialogue contexts should be more widely acknowledged in L2 assessment and teaching.