Haku

Comparing L2 Speech Fluency in Monologue and Dialogue: A Quantitative Study of Finnish Upper Secondary School Learners of English

QR-koodi

Comparing L2 Speech Fluency in Monologue and Dialogue: A Quantitative Study of Finnish Upper Secondary School Learners of English

This thesis examines second language (L2) speech fluency in monologic and dialogic task mode. The aim of the thesis is to investigate how the mode of performance affects language learners’ speech fluency with a quantitative analysis of fluency and disfluency features in L2 English monologues and dialogues produced by 22 Finnish upper secondary school students. The learners’ speech fluency is analyzed using a set of different fluency measures covering the three aspects of speed (speech tempo), breakdown (pausing) and repair fluency (corrections and repetitions). These measures are then analyzed statistically to examine the differences and correlations between monologue and dialogue fluency measures. The results show that the learners’ performance was statistically significantly more fluent in dialogue in terms of most speed and breakdown fluency measures, namely speech rate, the number of mid-clause and end-clause silent pauses, and the mean length of end-clause, mid-clause and overall silent pauses. However, the mean length of run and the number of false starts indicated a higher level of fluency in monologue, possibly partly due to dialogue-specific interactive features such as turn pauses and interruptions. The results also show that task mode affected the duration but not the frequency of filled pauses, which seems to support the hypothesis that filled pauses should be regarded as means for maintaining fluency rather than signs of disfluency, especially regarding the dialogic mode. The study also employed a set of novel measures of repair extent, and since two of these new measures demonstrated statistically significant differences between the two task modes, it is proposed that both repair frequency and extent should be taken into consideration in future fluency studies. The study has several important methodological implications for the adaptation of monologue fluency measures to dialogue speech and the operationalization of repair extent measures. Overall, the findings demonstrate how studying L2 fluency in interaction can help to provide a more comprehensive picture of L2 speech fluency and highlight its importance for L2 fluency research, assessment and teaching.

Tallennettuna: