The construction of an ordinal scale Y , to be associated to a performance index in evaluating social-agents, is outlined which is quanti ed, adopting an intrinsic worthiness criterion, standardized on a chosen reference-agent, eg that which is intended as representing a (actual or perhaps hypothetical) best practice . The usual practice of using the equispaced scale is re-interpreted. The ordinal levels of Y are identified by design through scheduling a hierarchical sequence of increasingly stringent goals to be achieved . The index, to be associated to Y , borrows the structure of the Yaari-Quiggin functional, from the RDEU theory. But, the concept of value increases , in advancing on the scheduled goals, is meaningfully re-interpreted herein, besides utility-based meaning, as social worthiness . These value increases may be extracted, fully normalized on the data-behavior of the chosen reference social-agent , upon a probabilistic formal setting using data- analysis (perhaps pseudo-Bayesian) tools. Thus, the ordinal levels of Y remain quantified, alternatively to various other approaches.
Worthiness-based Scale Quantifying Re-interpretare scale ordinali equi-distanziate
D'EPIFANIO GIULIO
2020
Abstract
The construction of an ordinal scale Y , to be associated to a performance index in evaluating social-agents, is outlined which is quanti ed, adopting an intrinsic worthiness criterion, standardized on a chosen reference-agent, eg that which is intended as representing a (actual or perhaps hypothetical) best practice . The usual practice of using the equispaced scale is re-interpreted. The ordinal levels of Y are identified by design through scheduling a hierarchical sequence of increasingly stringent goals to be achieved . The index, to be associated to Y , borrows the structure of the Yaari-Quiggin functional, from the RDEU theory. But, the concept of value increases , in advancing on the scheduled goals, is meaningfully re-interpreted herein, besides utility-based meaning, as social worthiness . These value increases may be extracted, fully normalized on the data-behavior of the chosen reference social-agent , upon a probabilistic formal setting using data- analysis (perhaps pseudo-Bayesian) tools. Thus, the ordinal levels of Y remain quantified, alternatively to various other approaches.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.