Different strategies are used in manual assembly to assist assembly workers when selecting parts to install. This study measured the reliability in terms of the error rates of several selection strategies. Data from three car manufacturing plants, one in the USA and two in Mexico, were used in this study. For each selection strategy, data on errors were collected for 6–12 months in a minimum of 30 operations using samples of no less than 110,000 cars in each workstation. Only assembly workers with at least six months of experience performing the task were selected for the study. It was found that unassisted selection had higher error rates than assisted strategies (e.g., pick-to-light, pre-picked, scanner). The precise error range for each selection strategy should be considered in critical selection operations to increase the reliability of the operation and reduce quality issues. The quality improvement and error reduction approach shifted from corrective efforts and placing the responsibility mostly on the operators to enhancing the system’s reliability by implementing more capable selection strategies.