Four different laboratory compaction methods, i.e., the Marshall, vibration, static, and gyratory methods, were used to study the effects of type of compaction method on the triaxial stress-strain behavior of asphalt concrete. The behavior was compared to that of asphalt concrete compacted in the field by a vibratory roller. Although the asphalt concrete specimens were all made of the same mix and compacted to approximately the same density (air voids), the resulting stress-strain curves were very different. The secant modulus up to 1 % axial strain for the gyratory compacted specimens was six times that of the field compacted specimen, and the axial strain of the field specimen at failure was six times that of the gyratory specimen. The reason is that, although the asphalt concrete density is the same for the different specimens, the aggregate particle arrangement and interlocking (skeleton structure) after compaction are very dependent upon the method of compaction used. For field quality control, common practice is to compare the behavior of field compacted specimens to specifications based on laboratory specimens. Therefore, it is important to use laboratory compaction methods that yield specimens with behavior similar to that resulting from roller compaction in the field. Two such methods are proposed, a modified Marshall method and a special compaction method.