Due to an educational system that tends to prioritize receptive reading and listening skills and a concomitant lack of opportunities to engage in meaningful spoken production in the target language outside of the class, many English as a Foreign Language (EFL) learners in Japan struggle to develop their English speaking abilities even after many years of study. For many of these learners, post-secondary coursework presents a first opportunity to engage in consistent active oral production of the language in interactive communicative situations, but how language abilities develop in this context over time is not well understood. This longitudinal, mixed methods research study explores the development of interactive second language (L2) speaking proficiency of Japanese university students throughout their first year of post-secondary study. The participants were 92 female students from five sections of a required oral communication course of which I was the instructor: 34 from two mid-proficiency sections and 58 from three low-proficiency sections. In addition to the standard curriculum for this course, these participants also completed a series of ten 10-minute discussion tasks in groups of three to four participants on personal information topics of general interest over the course of two 15-week semesters that comprise the academic year. All of the group discussion tasks were video-recorded and carried out under the same procedures: without any pre-task planning time but with a post-task transcription and reflection assignment. The first group discussion task was administered for training and instructional purposes and excluded from the data analyses leaving seven tasks conducted roughly one month apart throughout the academic year as well as one additional repeated task with the same group assignments completed at the end of each semester. In total, nine discussion tasks were carried out for analysis in this study. A number of quantitative measures were collected for this study at various points throughout the academic year. First, all of the group discussion performances were rated using an adapted rating scale consisting of criteria dedicated to complexity, accuracy, and fluency (CAF) as well as interactional engagement and overall communicative effectiveness. Additional measures consisting of participant ratings assigned to topic difficulty and group member familiarity were collected via questionnaire after each discussion task. Furthermore, measures of extraversion and English-speaking self-efficacy were collected along with the three repeated tasks that spanned the entire year. Finally, initial and end-of-year scores on the TOEIC-IP test consisting only of listening and reading sections were used as measures of general English proficiency. To provide a richer understanding of the participants’ perceptions of their development as well the social and individual difference factors considered most directly relevant to the research questions, an embedded qualitative component consisting of two focus groups and one individual interview was also carried out. The results indicated that the participants in this study made small but significant improvements in their mean interactive English speaking abilities across both the three repeated tasks and between the most temporally distant earliest and latest new tasks administered in the first and second terms, accounting for relatively large amounts of the variance in the measures. Furthermore, the social and individual difference factors investigated were found to jointly predict 56% of the variance seen in the interactive English speaking measures for the three repeated tasks. Of these factors, the participants’ initial general proficiency scores, English speaking self-efficacy ratings, and averaged group member initial proficiency scores were found to be much stronger predictors than their topic difficulty, group member familiarity, and extraversion ratings. When the ratings given along the five rating scale dimensions were inspected, strong relationships were revealed among all of them, though particularly among the CAF rating measures. Tracking these relationships over time revealed that they tended to strengthen from one task to the next with the most readily identifiable changes found among the relationships between accuracy, interactional engagement, and overall communicative effectiveness. Finally, the qualitative component of this study revealed that the participants interviewed perceived changes in their conscious attention during the group discussion tasks shifting from accuracy concerns to speaking fluently/effectively, a general lack of concern for and attention to complexity throughout, and improvement in their interactional engagement over the course of the academic year. Furthermore, the interviewees were found to generally ascribe a high degree of influence on their resulting group discussion task speaking performances to both the topic prompts assigned and their group members’ personalities as well as to the complex interplay between group member proficiency and familiarity among other factors. Although a number of limitations should be taken into account, it is believed that the results of this study provide new, important insights into the longitudinal development of interactive English speaking proficiency in an EFL university context where learners have little engagement with target language beyond limited classroom contact hours. / Teaching & Learning