Doctoral programs in the sciences are structured to train primarily future academic professionals. However, it is well-documented that only about a third of all PhD's in science and engineering eventually find permanent employment in academia. At the same time, however, the average employment rate for science and engineering PhD's has been consistently above the national average, indicating that such graduates can successfully seek employment outside their immediate field of expertise. This is doubtless due to their general aptitude for tackling complex tasks, which is in turn a direct outcome of their training in research. Concurrently, the last decade has witnessed profound changes and restructuring of traditional industrial research and development. Many industrial and corporate laboratories that had been at one time heavily engaged in cutting edge basic research were refocused to have a greater and more immediate impact on production, allowing firms to compete on several fronts at the same time, make more efficient use of internal resources, and keep up with rapidly changing technology. This process has naturally favored skilled but broadly educated scientists, who are capable of working beyond the traditional boundary of their own field, typically within multidisciplinary teams. These qualities are at odds with the narrow focus that characterizes graduate student research in science doctoral programs across the nation. Graduate students are under pressure to produce individual original contributions within a limited field. They are practically never encouraged to explore the possible relevance of what they are learning to other areas of research, or even to familiarize themselves with research themes or terminology from other disciplines. These issues are particularly important in Computational Science, as the great generality of its methods and techniques makes them relevant to virtually any research work. Yet, very seldom are PhD's in scientific disciplines or engineering, even those who have performed computational work for their thesis project, capable of quickly exporting methods and applications to other fields without substantial retraining. The PhD programs in Computational Science and Computational Science with concentration in Statistics provide an alternative to these traditional approaches, accomplishing its goal by pursuing an interdisciplinary approach to graduate training.