Recently, different hybrid GNSS/cellular methods combining GNSS measurements with cellular network measurements have been proposed. These methods are designed to improve the availability (and accuracy) of position determination in situations where few satellite signals can be received, such as in urban canyons or even indoor environments. In order to get some interesting insights into the performance of these hybrid GNSS/cellular methods under various conditions and system geometry configurations, we present in this paper a detailed analytical and numerical performance analysis. Our analysis is based on the Cramer-Rao lower bound (CRLB) theory for deterministic and random parameters, as well as an analytical asymptotic expression for the location mean-square error (MSE). Based on this theoretical framework, we investigate analytically and using Monte-Carlo simulations how the positioning accuracy of different hybrid systems is affected by the relative geometry and the type and number of measurements between the mobile station (MS), the cellular base stations (BSs), and the satellites. The effect of important biased cellular network measurement errors (such as due to multipath and non line of sight (NLOS) propagation) are also considered.