The paper gives an alternative statistical modeling approach for developing modal correction factors for hydrocarbon (HC) emissions from motor vehicles. 'Modal' refers to operating modes of vehicle activity; e.g., cruise, idle, deceleration, acceleration, power, and positive kinetic energy. The modeling method used to estimate HC modal emission factors is called binary recursive partitioning, or hierarchical tree-based regression (HTBR). The paper explains the statistical theory and gives specific modeling results for modal emission factors for HCs. A data set containing 4800 vehicle tests representing 29 different laboratory testing cycles was developed and used in the analysis. HTBR methods provide statistical features that are problematic for classical ordinary least squares (OLS) regression methods, a commonly applied statistical technique for analyzing emissions data (both the California Air Resources Board (CARB) and the U.S. EPA have applied OLS regression techniques to derive emission factors in the EMFAC and MOBILE models, respectively).