Gas Metal Arc Welding (GMAW) is widely employed for joining relatively thick sheet steels in automotive body-in-white structures and frames. The GMAW process is very flexible for various joint geometries and has relatively high welding speed. However, fatigue failures can occur at welded joints subjected to various types of loads. Thus, vehicle design engineers need to understand the fatigue characteristics of welded joints produced by GMAW. Currently, automotive structures employ various advanced high strength steels (AHSS) such as dual-phase (DP) and transformation-induced plasticity (TRIP) steels to produce lighter vehicle structures with improved safety performance and fuel economy, and reduced harmful emissions. Relatively thick gages of AHSS are commonly joined to conventional high strength steels and/or mild steels using GMAW in current body-in-white structures and frames. Therefore, the Sheet Steel Fatigue Committee of Auto/Steel Partnership (A/S-P) has completed fatigue tests of GMAW joints for DP590 GA, SAE 1008, HSLA HR 420, DP600 HR, Boron, DQSK, TRIP780 GI, and DP780 GI sheet steels. Dissimilar metal welded joints were tested for DP590GA and SAE1008, DP600 and SAE1008, TRIP780 and SAE1008, DP780 and SAE1008, and Boron and HSLA. The specimen configurations included single lap-shear, double lap-shear, butt weld, start-stop, and perch mount. The fatigue test results obtained from various sheet steels and specimen types showed that the strength of the base metal was not an important parameter to determine the GMAW fatigue life. For the same specimen type and sheet thickness, the test results for the different base materials collapsed nicely into a well defined curve when plotted in log-log scales. However, some specimen types showed greater scatter at long fatigue lives than others. This study investigated the sources of the scatter in terms of the dimensional variability of the weld geometry, such as weld toe radius, weld gap, horizontal weld leg length, and vertical weld leg length.