The sensitivity of the wheel/rail contact problem to the approximations made in some of the creepage expressions is examined in this investigation. It is known that railroad vehicle models that employ kinematic linearization can predict, particularly at high speeds, significantly different dynamic response as compared to models that are based on fully nonlinear kinematic and dynamic equations. In order to analytically examine this problem and numerically quantify the effect of the approximations used in the linearized railroad vehicle models, the fully nonlinear kinematic and dynamic equations of a wheel set are presented. The linearized kinematic and dynamic equations used in some railroad vehicle models are obtained from the fully nonlinear model in order to shed light on the assumptions and approximations used in the linearized models. The assumptions of small angles that are often made in developing railroad vehicle models and their effect on the angular velocity, angular acceleration, and the inertia forces are investigated. The velocity creepage expressions that result from the use of the assumptions of small angles are obtained and compared with the fully nonlinear expressions. Newton-Euler equations for the wheel set are presented and their dependence on Euler angles and their time derivatives is discussed. The effect of the linearization assumptions on the form of Newton-Euler equations is examined. A suspended wheel set model is used as an example to obtain the numerical results required to quantify the effect of the linearization. The results obtained in this investigation show that linearization of the creepages can lead to significant errors in the values predicted for the longitudinal and tangential forces as well as the spin moment. There are also significant differences between the two models in the prediction of the lateral and vertical forces used to evaluate the ratios as demonstrated by the results presented in this investigation.