Reliability is defined as the ability to adhere to a preplanned schedule or to defined intervals and a fixed travel time. Reliability is expected to be high particularly in metropolitan cities. To achieve this, high-cost investments such as rail systems that are operated independently of private vehicle traffic and simultaneous passenger information systems are needed. The social benefit to be made from these investments depends largely on how it is perceived, rather than the magnitude of the reliability. In this study, it is aimed to develop a parametric support data for benefit analysis of investments in terms of reliability. It is investigated whether the purpose of travel and personal characteristics are effective in the sense of reliability or not. Parametric models are developed by using multiple logistic regressions with the collected data by the help of the stated preference survey. When the model findings are examined, it is found that parameters such as gender, presence at the university for 1-3 years, use of public transportation, etc. are effective in tending to direct transportation modes that constitute a higher sense of reliability. In the reliability scenario model testing the high risk taking behavior, it is found that disability for using metro as a transfer alternative and living for a long time in the city generate a tendency to shorten the travel time by taking more risks in reliability. It is concluded that reliability is perceived differently, even on different trip purposes of the same person, and therefore it should be considered as a socio-economic parameter like age, gender, income, etc. on its own, rather than a cost weighting component in transportation demand models.