Lane changing is an important component of microscopic traffic simulation models. In this paper a systematic approach for modeling lane changing behavior using a discrete choice framework is presented. Lane change is modeled as a sequence of three steps: decision to consider a lane change, choice of left or right lane, and search for an acceptable gap to execute the decision. Results from the estimation of the parameters of the gap acceptance model (using on-ramp merging data) show that in addition to the gap length, other important factors that affect drivers gap acceptance behavior are relative speed, distance remaining to the point at which lane change must be complete, and delay in completing merging. (A) For the covering abstract see IRRD 886400.