The choice of technology to transport passengers in large metropolitan areas is an important issue everywhere. There are many factors involved in this choice. This paper deals with the possibility of the objective use of available information in the analysis of the suitability of a rail public transport system for a city. A database has been made from publications on public city transportation and country level information. Logit models of choice have been calibrated by the maximum likelihood and nonlinear least square methods based on the acquired information. Each city is treated as an "individual, choosing rail or non-rail modes for its trips. Only cities with a population of more than one million have been included in the analysis to ensure the instigation of mode diversification in these cities. Selected models have been validated and then used to suggest the desirability of a rail public transport mode in some sample cities, according to world practice.