Comparability Through Shared Standards and Procedures: Why Observations Need Compatible Standards Before They Can Be Compared

This conceptual paper introduces Comparability Through Shared Standards and Procedures, a mechanism in which shared units and compatible methods make observations from different positions comparable. A common unit label or matching numerical result does not, by itself, establish that the observations measure the same thing. The paper assesses comparisons through the definitions, procedures, and calibration used to produce each observation. Different procedures can remain comparable when a reliable conversion connects them. Comparability also remains distinct from accuracy: aligned measurements may still contain error, and the intended comparison determines which conditions must be shared. The discussion connects the mechanism to Porter’s account of quantification and trust and examines Egyptian cubit standards through Schaffer’s and Yasuoka’s historical studies. These materials describe different lengths and, in Greco-Roman records, different systems of subdivision under the same unit name. The example illustrates why identifying definitions and applicable conversions matters before comparing figures. Part of the Shared Capacity and Learning series within On Human Modeling, the paper examines who establishes and maintains standards and who relies on the resulting comparisons. Its framework can inform human–AI research on whether evaluation scores and measurements produced by different systems support the comparisons made from them.