Advancements in parallel computing have brought partitioned analysis to the forefront of numerical modeling and simulation practices. The strongly coupled models obtained through partitioned analysis are instrumental in making critical decisions that may affect public safety such as failure analysis of structures in extreme conditions, performance of ballistic missile systems, and reliability of nuclear reactor cladding. Scientists and engineers who develop coupled models as well as decision makers who depend on these models' predictions need an understanding of the credibility of such complex modeling efforts. To ensure satisfactory predictive performance, the development of frameworks and methods for verification and validation of coupled models must complement the progress that is currently being made in multi-scale, multi-physics modeling and simulation. A unique aspect of partitioned analysis is that the iterative nature of coupling operations causes uncertainties and errors in constituent models to be passed back and forth through the interface, which may lead to the accumulation of these uncertainties and errors in the coupled model. Understanding and quantifying the sources of uncertainties and errors in the coupled models would therefore allow model developers to strategically improve the model's predictive capability. Such understanding would also open the door for optimization of the allocation of resources for conducting calibration experiments and for further developing the computer code. With advancements in verification and validation frameworks and methods specifically tailored for partitioned analysis, strongly coupled models can be used with confidence to garner previously unavailable knowledge for critical engineering applications.