The aim of this study was to determine the relationship between language learning strategies of secondary school students and the concept of academic self. The study was carried out with 522 students enrolled in an Anatolian high school in the central district of Ankara. In the study, Personal and Academic Self-Concept Inventory (PASCI), which was developed by Fleming and Whalen (1984), and the Turkish version of Strategy Inventory of Language Learning (SILL), which was developed by Oxford (1990) and adapted into Turkish by Cesur & Fer (2007), were used. Canonical correlation analysis was used in the study. As a result of the study, it was concluded that the scores obtained in certain sub-dimensions of language learning strategies were predicted by the scores obtained in some of the academic self-concept sub-dimensions; however, the effect of the scores concerning some sub-dimensions remained at low levels. In the study, the relationship between the strategies in the language learning strategies sub-dimension and the academic self-concept subdimensions was explained through two canonical variables for the 9th, 10th, and 11th Grades. Based on this result, it could be argued that the increase in scores obtained from academic self-concept led to an increase in scores obtained from language learning strategies. In addition, it was concluded that the best predictors in explaining memory strategies were academic and verbal skills; post-cognitive strategies had low power to predict language learning strategies; and, social anxiety and family acceptance dimensions had low powers in predicting academic self-concept perception.