In lieu of an abstract, here is a brief excerpt of the content:

199 14 ■How Does Foreign Language Proficiency Change over Time? Results of Data Mining Official Test Records Amber bLoomFieLD, STeven roSS, megAn mASTerS, KASSAnDrA gynTHer, AnD STePHen o’ConneLL University of Maryland ■ STUDIES INVESTIGATING CHANGE IN foreign language skills over time have explored a number of factors affecting skill loss, including achieved proficiency in the foreign language , the amount of time since foreign language input markedly decreased (e.g., since the conclusion of formal language training), and use of the foreign language during this period of reduced input. In addition to the general interest in identifying factors that impact the likelihood of losing foreign language proficiency skills over time, there is interest in exploring rate of loss to address language policy questions, such as the frequency with which foreign language skills should be formally tested. This paper discusses the results of preliminary data mining of test records for more than eight hundred foreign language professionals. Longitudinal analyses explored how listening, reading, and speaking proficiency test ratings change over time. In addition, analyses investigated the extent to which lag time between test occasions (i.e., days between test events), participation in formal language courses, and initial test rating impact the rate of loss in test ratings. Overall trends show patterns of improvement in test ratings for reading and listening; no significant pattern of change was found for speaking skills. Latent growth analyses indicated that individuals who participated in foreign language courses had faster rates of improvement for reading and listening. Event history analyses of cases that showed a loss in ratings found that speaking was the most vulnerable skill, with a sizable proportion of individuals showing loss in their test ratings when the time between tests exceeded two years. Reading and listening proficiency ratings were more stable, with a large proportion of individuals maintaining their rating even when the time between tests exceeded three years. Rate of loss did not significantly differ based on initial reading or listening test score or participation in formal language courses. The paper discusses the implications of these findings for policies on foreign language testing frequency and future research on factors that may mitigate attrition; it also explores the 200 A. Bloomfield, S. Ross, M. Masters, K. Gynther, and S. O’Connell applications of the statistical approach used here to other research questions related to language learning. Literature Review The study of how foreign language proficiency changes over time is a relatively new area of research, described as deriving its principal impetus from a conference at the University of Pennsylvania in May 1980 (e.g., Clark and Jorden 1984; De Bot and Weltens 1995; Lambert and Freed 1982; Weltens 1987). Most studies in this area have focused on which aspects of the foreign language are lost, and in what order (e.g., syntax versus lexical knowledge first; Jordens, De Bot, and Trapman 1989). Although this approach is valuable for describing those aspects of language most vulnerable to loss and for comparing the pattern of attrition to the pattern of acquisition, it does not address the question of what factors (e.g., regularly reading newspapers in the L2) affect change in general language abilities, such as reading comprehension over time or how quickly skills begin to show loss. Studies that have investigated factors influencing change in more general foreign language skills have examined duration of the period of reduced input (e.g., time since the end of formal language training), achieved proficiency level prior to this period, amount of target language use during this period, and other factors. However, as many of these studies involve children or examine language skills at only one or two time points, this research is often difficult to generalize to the language skills of adult speakers over an extended period of time. The current study expands on previous research by examining a database of adult foreign language professionals’ proficiency test records, systematically archived over a period of six years, to determine how reading, listening, and speaking skills change over time; the rate at which loss occurs; and how change is affected both by participation in formal language courses and initial proficiency level. Period of Reduced Input and Change in Language Skills Duration of the period of reduced input has been defined as the time since the end of formal training (e.g., Bahrick 1984) or since the end of intensive language exposure , such as an immersion experience (e.g., Snow, Padilla, and Campbell 1988). As its name implies, the period of reduced...

Share