Most induction motor analyses that allow for the presence of inter-bar currents are based on a rotor model in which the inter-bar currents are assumed to flow between adjacent pairs of bars. An alternative model is proposed in which the inter-bar currents flow into and around the rotor core. The validity of the new model is verified by means of measurements made on 54 rotors of identical design, which have been subjected to a range of treatments during manufacture to produce bar-to-bar resistances that differ by up to three orders of magnitude. It is shown that the new rotor model may readily be incorporated into existing analyses by redefining a single inter-bar resistance parameter. An alternative viewpoint of this is that the new rotor model redefines the relationship between the resistance parameter used in analytical models and the bar-to-bar resistance measured under test. © IEE, 2005.