The main aim of this study was to investigate whether certain variables namely school attachment, perceived social support, cognitive flexibility and gender can predict the resilience of senior high school students that are economically disadvantaged. Another aim of the research is to reveal the relation between the level of academic resilience and gender variable. This study also seeks to examine the differences between academically resilient students with different levels of resilience in terms of the stated variables. For these purposes, a quantitative research method, a correlational research model, was used. The participants consisted of 304 senior students enrolled in 18 Anatolian High Schools during the 2014-2015 academic year in the three districts of the Ankara province, Turkey. Data was collected using 'the Resilience Scale for Adults', 'School Attachment Scale for Children and Adolescents', 'the Multidimensional Scale of Perceived Social Support', 'Cognitive Flexibility Inventory' and the student information form which was designed by the authors. To analyze the collected data, a Standard Multi Regression Analysis, a Multivariate Analysis of Variance (MANOVA) and a CM-square Test were employed. The results of the Multi Regression Analysis revealed that cognitive flexibility and perceived social support significantly predicted the level of academic resilience of economically disadvantaged high school students. However, school attachment was not found to be a significant predictor. The three variables given above accounted for approximately 41% of variance in the resilience scores of the participants. Furthermore, the results obtained from MANOVA showed no difference between students with high and low academic resilience in terms of cognitive flexibility, school attachment and perceived social support. Moreover, there was a significant relationship between the two groups in terms of gender according to the results of the Chi-square Test.