Concrete pavement distress is often associated with the effectiveness of load transfer at joints and contributing factors such as pumping. Current analytical methods attempt to simulate load transfer, curling, and load effects in the modeling of pavement response. In general these analytical models do not accurately define load transfer and thermal effects. Therefore, research has been initiated to develop improved methods for analysis and design of concrete pavements. The initial results of tests conducted on a Florida Department of Transportation test pavement indicate that pavement and joint response can be effectively modeled using a three-slab, two-joint, finite-element computer program (FEACONS III). Besides conventional layer parameter input, the program requires spring constants for pavement-edge friction, joint shear, and joint moment. The analysis of plain concrete pavement was performed using the falling weight deflectometer (FWD). Data were collected during different seasons, when the differential and average slab temperatures varied substantially. Generally four different load levels were used in the FWD to assess load-deflection linearity. Temperature-curling and contraction-expansion effects were also monitored independently. Spring stiffnesses were varied in the FEACONS III analyses until the predicted deflection basins matched those measured for different temperature and loading conditions. The results obtained with a downward curling (differential approximately 9 deg F or 5 deg C) indicated that spring stiffnesses representing edge friction, joint shear, and moment at the joint remained constant regardless of loading position. This suggests that differential drying shrinkage or a moisture differential had produced upward warping, which was offset by the 9 deg F (5 deg C) downward curling. At other differential temperatures, the spring stiffness varied according to slab lift-off and load position. The average slab temperature (seasonal) was found to have a pronounced effect on joint stiffness. At high temperatures, the shear and moment stiffnesses were very high, providing close deflections for loaded and unloaded sides of the joint. When mean slab temperature was lowered, the analyses indicated a significant reduction in joint stiffness.