In practice, most control systems are partially controlled. When a system is only partially controlled, the performance of the uncontrolled part of the system becomes worse than that of the corresponding part of the fully controlled system. To describe the performance degradation, a performance factorization is obtained. It is shown that the partial control performance is a product of the full control performance and a performance degradation factor. This factor is a submatrix of a left null-space projection determined by the measurement selection. The performance degradation factor provides an efficient tool for performance assessment and measurement selection. The theoretical results are applied to a highly integrated chemical plant-the HDA process.