Experimental and analytical investigations were performed to evaluate the comparative performance of pavements with and without geosynthetic stabilization. This was accomplished by the testing of a total of 18 pavement sections which could be classified into four different types: one which was constructed without geosynthetics and which served as a control, and three which were stabilized with one of two geotextiles or a geogrid. The pavement sections were constructed to model a typical secondary road in Virginia which is constructed over a silty sand subgrade material. Loading of the pavement sections was accomplished through the use of a computer-controlled pneumatic system which delivered 80 lb/in2 (552 kPa) through a rigid plate at a frequency of approximately one-half Hertz. The resulting displacement of the pavement surface was monitored by an array of linear variable displacement transducers (LVDTs). The performance of each pavement section was evaluated using the American Association of State Highway and Transportation Officials (AASHTO) flexible pavement design method. Models based on empirical and mechanistic relationships were considered. A theoretical pavement section was also analyzed to assess the influence of the stiffness of a wearing course layer. An economic study was performed to assess the potential cost benefit of geosynthetic stabilization and recommendations have been made for additional research.