Effective training can provide the military with the edge it needs to operate successfully, win wars, and save lives. Yet, properly assessing the degree to which training is effective is a complex process, involving measures of learner performance, measures of organizational impact as well as determinations of what tradeoffs must be made to develop, acquire, and maintain any training solution. Most training effectiveness assessments are conducted to determine the utility of adopting new, or at least different, instructional capabilities. In these cases ‘utility' is measured in terms of metrics like reducing the time learners need to reach criterion performance or increasing learner performance while holding time constant. While these measures quantify whether or not learners have increased their knowledge, skills, and abilities beyond those of others trained differently and/or have done so more quickly, they do not quantify the impact such training will have on their performance in real world settings nor how such performance will impact the organization of which they are a part. Consequently, these measures do not provide decision makers with the information they need to adequately assess tradeoffs in committing long-term funding, personnel, and infrastructure resources to a particular training capability or to training itself. Their decisions require more comprehensive analyses that assess the long-term impact of potential training capabilities, combined with a wider set of metrics that account for research and development investments, equipment purchases, equipment maintenance costs, learner and instructor time, follow-on sustainment, and on the job training. In this paper, we address this gap by developing an approach based on learning and economic theory for assessing return on investment and thereby obtaining measurement information needed for these analyses. Frameworks and examples applying this approach to evaluations of training systems are provided.