This study was conducted in order to evaluate sensitivity of two levels of delay discounting tasks and to determine whether levels of task reward could discriminate people with diverse status of health-related behavior. Reward magnitude was used as an independent variable. Participants included 202 undergraduate students (87 males and 115 females, mean age 20.03(SD=1.88)). Each participant performed two levels of delay discounting tasks and completed the Barratt Impulsive Scale-11(BIS-11). They were classified into groups according to their status on two health-related behaviors, smoking and drinking alcohol. According to the results, the discounting rate was significantly higher when the value of reward in the discounting task was lower. In addition, the group of people who smoke or drink alcohol reported a significantly higher discounting rate in the task with lower value of reward. However, delay discounting task with higher value of reward and BIS-11 did not explain the differences between the groups. Results implicated that reward magnitude in delay discounting task may influence sensitivity of the task. This suggested the need for careful selection of the amount of reward of delay discounting task when evaluating impulsivity. Additional information and limitation for future research were discussed.