Studies have indicated significant linkages between Language Learning Strategies (LLS) use and language performance level (O’Malley et al. 1985a; Manghubai 1991; Wharton 2000). There are studies that explore whether good language learners use LLS frequently and in large quantity (Green & Oxford 1995). Therefore, this paper describes a study designed to explicate the relationship between LLS use and Arabic language proficiency level. Data for the study were collected using self-report questionnaires, including the adaptation Strategy Inventory for Language Learning (Oxford 1990) version 7.0 with additional of one new construct. The sample was comprised of 460 Form 4 students from thirteen religious secondary schools in Terengganu, Malaysia. The MANOVA (significance level p < 0.05) and ANOVA (significance level p < 0.01) statistical tests were applied on the questionnaire data to examine the relationship between language performance and LLS use. The results indicate that students in higher proficiency level used a significantly greater number of strategies more frequently than did their counterparts in lower proficiency levels in all seven strategy constructs. The medium proficiency level students used a greater number of strategies than those in the low proficiency level in all seven strategy constructs except in the affective strategy.