The development and utilization of commensurable measurement scales is driven by the desire of decision-makers to compare apples with apples. Comparable measures serve the need to understand the resources required to achieve an output or an outcome and hence value for money. Commensurability rests at the heart of efficiency and effectiveness measures and ensures that discussion regarding different approaches to achieve the same outcome can be accomplished in a technical and unemotional assessment which utilizes the full benefit of a rational, objective approach. When such a scale is available an accountable decision-maker (agent) may use the scale to select from among the available alternatives and to defend their choice to their supervisor(s) (principal).
However, in practice decision-makers are often faced with incommensurable measures and are still required to make choices to optimize the expected utility created by their allocation of scarce resources. The Value Sieve, 1 is a resource allocation methodology developed to assist managers and administrators of organizations who must make choices using incommensurable, apples vs. oranges, measures. Further the Value Sieve creates both procedural and consequential accountabilities to better manage the problems which are associated with decision-making under conditions of complexity, risk, ambiguity and ignorance. 2
At its core, commensurability encompasses traditional measurement validity and reliability concerns. Validity issues ask if the variable measured is indeed a reasonable representation of the dimension of interest within the applied setting. Reliability asks if the measure of a variable can be taken consistently. While validity and reliability are most commonly associated with experimental procedures they must also be considered within the context of quasi-experimental procedures and programme evaluation. 3
The rigour of experimental process is often unobtainable within an applied setting. This fact will surely frustrate those individuals who wish