Background: The effects of users, glove types, and tourniquet devices on the performance of limb tourniquet use in simulated first aid were measured. Materials and Methods: Four users conducted 180 tests of tourniquet performance in eight glove groups compared with bare hands as a control. Results: Among tests, 99% (n = 179) had favorable results for each of the following: effectiveness (i.e., bleeding control), distal pulse stoppage, and tourniquet placement at the correct site. However, only 90% of tests ended with a satisfactory result, which is a composite outcome of aggregated metrics if all (patient status is stable, tourniquet placement is good, and pressure is good) are satisfactory. Of 18 unsatisfactory results, 17 (94%) were due to pressure problems. Most of the variance of the majority of continuous metrics (time to determination of bleeding control, trial time, overall time, pressure, and blood loss) could be attributed to the users (62%, 55%, 61%, 8%, and 68%, respectively). Glove effects impaired and slowed performance; three groups (cold gloves layered under mittens, mittens, and cold gloves) consistently had significant effects and five groups (examination gloves, flight gloves, leather gloves, glove liners, and glove liners layered under leather gloves) did not. For time to bleeding control and blood loss, performance using these same three glove groups had worse results compared with bare hands by 26, 18, and 17 seconds and by 188, 116, and 124mL, respectively. Device effects occurred only with continuous metrics and were often dominated by user effects. Conclusion: In simulated first aid with tourniquets used to control bleeding, users had major effects on most performance metrics. Glove effects were significant for three of eight glove types. Tourniquet device effects occurred only with continuous metrics and were often dominated by user effects.