Spaan Fellow Working Papers in Second or Foreign Language Assessment, Volume 3, 2005 1 English Language Institute, University of Michigan Language Learner Strategy Use and English Proficiency on the Michigan English Language Assessment Battery Xiaomei Song Queen’s University Using a 43-item strategy-use questionnaire, this study examines the nature of language strategies reported by test takers of the Michigan English Language Assessment Battery (MELAB). It further investigates the relationships between test takers’ reported strategy use and language test performance on the MELAB in the context of English as a second language (ESL). The results show that MELAB test takers’ perceptions of cognitive strategy use primarily fall into six dimensions: repeating/confirming information strategies, writing strategies, practicing strategies, generating strategies, applying rules strategies, and linking with prior knowledge strategies. MELAB test takers’ perceptions of metacognitive strategy use fall into three dimensions: evaluating, monitoring, and assessing. The results also reveal that some strategies had a significant, positive effect on language performance and some had a significant, negative effect on language performance, whereas others seemed to have no effect with this group of participants. Language testing researchers have been concerned with the identification of individual characteristics that influence variation in performance on language tests since the 1970s. One important variable that may account for the differences on language performance, according to Dreyer and Oxford (1996), is the use of language strategies, which are thought to be used by students at all instructional levels with various outcomes. The present study examines the nature of learner strategies reported by test-takers of the Michigan English Language Assessment Battery (MELAB). This study also investigates the relationships between reported learner strategy use and language test performance on the MELAB in the context of English as a second language (ESL). Factors Affecting Second Language Performance Language researchers have long held an interest in factors that may affect performance and scores on language tests. Bachman (1990) proposed a model to investigate the effects of three types of systematic sources of variability on test scores: communicative language ability, the personal characteristics of test takers, and the characteristics of the test method or test tasks. Among the three types of systematic sources of variability, communicative language ability was considered the central factor accounting for the variation of test scores in second language learning. It consists of three components: language competence, strategic competence, and psycho-physiological mechanisms. Bachman also argued that the second factor that influences test performance—test-taker characteristics—includes a variety of personal attributes such as age, gender, native language, educational background, attitudes, motivation, anxiety, learning strategies, and cognitive style. Bachman’s third factor—test