Factors that bring uncertainty into evaluation results should be allowed for: the organization's context; the precision of the measuring method; how accurate the monitoring and measuring equipment is; the competence of the observer or interviewer; the inherent variability of the process or characteristic; the reliability of data from third parties; the reliability of reporting, where low incident reports can reflect fear of reprisal; a high reporting threshold, such as logging only large spills, which can mislead; and the interpretation of data. Before judging whether performance improved, stayed level or deteriorated, the organization should consider the total level of uncertainty.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.