AbstractThe modeling of the electrical conductivity of polymer composites reinforced with conductive fibers is investigated. Existing models generally can be divided into percolation theories and non‐percolation theories. The basis of the percolation theory is the fact that the conductivity of the composite increases dramatically at a certain fiber concentration called the percolation threshold. This theory can be used to model the behavior of the composite or to predict the percolation threshold itself. Non‐percolation theories include terms, which account for microstructural data such as fiber orientation, length, and packing arrangement. A comparison of experimental data with predictions from the various models reveals that only the percolation theory is able to accurately model the conductive behavior of an actual composite. Two alternative new models, which predict the volume resistivity of a composite using microstructural data, are evaluated. The first model relates resistivity to the concentration and orientation of the fibers, while assuming perfect fiber‐fiber contact. The relationship between resistivity and fiber concentration predicted by the model is in qualitative agreement with actual data, and predictions of the anisotropy in volume resistivity compare well with experimental results. The second model accounts for the effect of fiber‐fiber contact and fiber length on composite resistivity. Predictions are in excellent agreement with experimental data for polypropylene composites reinforced with nickel‐coated graphite fibers.