English-Related Out-of-Class Time Use by Japanese University Students

Abstract

This project explored aspects of English-related out-of-class time use by Japanese university students. The aim was to identify the salient temporal and motivational features of the episodes. The study had three parts, two longitudinal components and one cross-sectional component. Data from these three components consisted of out-of-class English time use information collected through a time diary (8,838 episodes). The episode data was collected from longitudinal study participants (n = 125) and cross-sectional study participants (n = 642) who maintained the time diary for at least one week. Data also include interviews with longitudinal participants (n = 40) about their time use and motivation and motivational information collected through an L2 motivational self system survey administered to cross-sectional study participants (n = 1,399) that was modeled after the survey developed by Taguchi, Magid, and Papi (2009). Two sets of time use and interview data were collected longitudinally (one semester per dataset) through Longitudinal Study 1 (ninitial = 66, nfinal = 15), which included participants from three universities in Western Japan, followed by Longitudinal Study 2 (ninitial = 59, nfinal = 25), which included participants from two universities in Western Japan. Interviews were with participants from these studies. Participants maintained a record of their out-of-class English-related time use during the semester. These data provide an overview of the out-of-class time use of Japanese university students during a full Japanese academic year. Longitudinal Study 1 data were collected during the fall semester, the second term at Japanese universities. Longitudinal Study 2 data were collected during the spring term, the initial term. Longitudinal Study 1 participants reported 2,529 episodes and Longitudinal Study 2 participants reported 3,322 episodes of out-of-class English access during the study period. One interview was held with the Longitudinal Study 1 participants (n = 15), at the end of fall semester. Two interviews were held with Longitudinal Study 2 participants (n = 25), one during the term and one following summer holiday. Data were examined for their temporal patterns and the contextual and affective features of the time use episodes. The cross-sectional component collected data from participants (n = 1,399) at 11 universities in Western Japan. These participants provided data for the motivational survey (n = 1,399) and at least one week of out-of-class English time use (n = 642) data. The participants in the cross-sectional study reported 2,987 out-of-class English access episodes. Episode data for all three components (K = 8,838) and the motivational survey data (n = 1,399) were analyzed at the person, group, and amalgamated episode levels for the patterns of participants' time use using ANOVA and nonparametric procedures. The data were also examined using nonparametric procedures (Kruskal-Wallis) to exam the affective variables (anxiety, enjoyment) by the contextual variables (purpose, location, persons present) of that time use. The episode data regarding participants' (n = 1,399) time use and motivational survey data were analyzed using ANOVA, factor, Rasch, multiple regression analyses, and structural equation modeling. The analyses of the time use data considered the temporal features of the episodes (time of day, day of week, hebdomadal pattern), the contextual features of the episodes (purpose, location, persons with), and the affective features of the time use (enjoyment, anxiety). The analyses of the motivational data considered two models of the L2 motivational self system, an intention to learn model (ILM) and a time use model (TUM). The ILM replicates the model found in the final solution proposed by Taguchi et al. (2009) for their model of the L2 motivational self system. In this model, the outcome factor is Intention to Learn, a factor labeled Criterion Measures by Taguchi et al. In the TUM, the outcome factor of Intention to Learn is replaced by actual time use on out-of-class English access. Time use results from all three components of this study indicated most out-of-class episodes occur when the participant is alone at home either studying or listening to English music. The most typical episode was listening to music, either alone at home or while commuting. A similar pattern of out-of-class English access was found for participants in all three components of this study. Study-related episodes were not considered enjoyable but also were not seen as causing anxiety. The amount of out-of-class time varied widely between participants, with one longitudinal study participant devoting 40 hours per week to English outside of class. For longitudinal study participants, the time use episode data, along with interviews, indicated that habit was a primary driver of out-of-class English access, with participants showing stable patterns of time use, whether for enjoyment or study, during the term. For the most part, once participants in the longitudinal components for this project had established a routine it remained fairly consistent during the term. The interviews clarified the initial motivators and drivers of the participants' English-related out-of-class time use. Interviews also confirmed the importance of habit in out-of-class time use patterns. Moreover, the interviews also indicated that the participants' L2 motivation was set in junior high school, though initial interest may have begun much earlier. Moreover, results from the three components showed that none of the participants met the time requirements of the Japanese Ministry of Education, Sports, Technology and Culture (MEXT, 2002, 2009b) for out-of-class time allocated to study for their English courses, a 2:1 ration, for every week that they participated in the study. Only a few of the participants met this requirement for out-of-class English access during any week of the study and only if all purposes, including enjoyment, were considered. This study also addresses the call that Dörnyei (2000) made for research examining the links between motivation and behavior in L2 learning. One unique aspect of this study is the use of a behavioral variable, Time Use, in addition to the survey-assessed latent trait, Intention to Learn, to explore the links between motivational profile and actual behavior. Motivational results show limited support for Taguchi et al.'s (2009) structural model following the same paths that they used in their model of the L2 motivational self system. More informative is the difference in the loading of the motivational profile factors on the outcome variable, Intention to Learn, which was the Criterion variable in Taguchi et al.'s final model, and the loading of the motivational profile factors on the measure of actual out-of-class time accessing English. Results show that Ideal L2 Self loaded strongly (.94) on the criterion, Intention to Learn, but much less strongly (.35) on actual out-of-class time use. However, the confirmatory factor analysis also indicated only a marginal fit to the model. Of importance, however, is that participants who rated highly on their Ideal L2 Self rated highly on the Intention to Learn, but not nearly as highly on actual time use on English. This suggests that responses on surveys of motivation to learn a second language are measures of intention and should not be used to predict or explain actual language learning behaviors. Like the results found by M. P. Eccles et al. (2012), who looked at intention and behavior in medical intervention studies and concluded that surveys targeting intention did not indicate actual behavior, language researchers might need to take a more critical approach to any interpretation of survey-based results as explaining actual learner behavior. This study set out to begin the process of understanding language learners' out-of-class time to language learning and exploring the links between out-of-class behavior and the learners' motivations. Through the longitudinal and the cross-sectional components, the study clarified the ways in which language learners use their out-of-class time allocated to second language access, the amount of time that they allocate, and the characteristics of this time use. It examined the connections between motivation and behavior and began the process of linking motivational studies with actual behaviors called for by Dörnyei (2000). If, as this study has suggested, there is little connection between the level of motivation and the type of motivated behavior that is being targeted, then researchers need to reconsider the current construction of such instruments and search for alternative ways to include actual measures of behavior in L2 motivational surveys. Given the large body of research on motivation in language learning, the results of this study indicate that researchers might want to consider ways include a measure of actual behavior or interpret results of surveys more conservatively when making claims of links between motivation and actual intention.