Software reliability can be improved by tolerating software faults, such as using N-version programming technique. Reliability evaluation is focused on the modeling and analysis techniques for fault prediction purpose. In this paper, a straightforward analysis method for evaluating reliability of software system established by N-version programming is proposed. The dependent failure parameters are assumed as random variables instead of constant. A case study is presented of the analysis of failure data from two software projects; the effectiveness of proposed evaluation methodology is demonstrated.