Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Robert C. Lafayette


The purpose of this study was to investigate the effectiveness of subtitled video during transactional task practice on oral communicative performance of fifth semester college students of French learning with multimedia courseware. Task practice was provided by the multimedia package, "Practicing Spoken French" (PSF) specifically designed by the researcher for this study. The package, which integrates HyperCard and videodisc technologies, allows for individually customized practice of two video-based oral tasks, description and narration. Drawing on Salomon's theories about the effects of media coding elements and perception of task on cognition and learning, the study used a 2 x 2 factorial design with two levels of subtitling (subtitles, no-subtitles), and two levels of oral transactional tasks (higher, lower). Forty-four fifth semester college students of French were randomly assigned to one of four experimental treatments--subtitles/lower-level task, no-subtitles/lower-level task, subtitles/higher-level task, no-subtitles/higher-level task. Students under each treatment were required to participate in two experimental sessions to complete the four stages of the two practice tasks of their choice--watching a video segment, answering video-related questions, drafting a description or a narration following the video information, and recording an up to 3-minute oral sample, based on what they had drafted. The dependent variable, oral communicative performance, was assessed by applying four six-point rating scales (Effectiveness, Accuracy, Organization, and Fluency) to the subjects' oral samples. Subjects in the subtitles treatments scored significantly higher on the oral performance measure and had significantly better attitudes than subjects in the no-subtitles treatments. The effects of task level and the interaction of this variable with subtitling were insignificant. However, higher-level task subjects outperformed their lower-level task counterparts in the two experimental sessions. Also in both sessions, subtitles/higher-level task subjects obtained the highest scores on the oral performance measure. Insights on CALL issues such as locus of control, time on task, learning environment, word processing, and attitudes were offered, as well as conclusions and implications for speaking assessment, CALL design, and future research.