The aim of present study was to investigate the role of specific dimensions of perceived social support from family, gender, and geographic area of residence in predicting hope levels of high school students. Additionally, the factor structure of Turkish Version of the Hope Scale was reexamined due to controversial findings in Turkish literature. The sample was composed of 737 students (407 female, 330 male) from two high schools in rural and urban areas of Ankara, Turkey. Data were collected by administering the Hope Scale (Snyder et al. in J Pers Soc Psychol 60:570-585, 1991), the Perceived Social Support from Family Questionnaire (Güngör 1996), and a demographic information form. The role of independent variables in prediction of hope was investigated via Multiple Regression Analysis. Due to presence of some suppressor variables in the full model, two separate regression analyses were conducted for rural and urban areas. In the regression analysis for rural area students, informational support was the only predictor of hope levels. On the other hand, urban area students' hope levels were predicted by love-and esteem-related support, instrumental support, and gender. Lastly, the factor structure of Turkish Version of the Hope Scale was examined through conducting a Confirmatory Factor Analysis (CFA) to test how well two-factor model fit to current data. Results of the CFA indicated that two-factor model was confirmed.