Semantics and normative goals, not data distribution, should drive standardization

Evidence

As this brief survey has suggested, there are a variety of ways to use the empirical distribution of data to do standardization and which are also semantic transformations. … The position here is that the empirical distribution of data can help inform but it is semantics, normative concerns, and the like that should determine the semantic transformations. — Gary Goertz, Social Science Concepts and Measurement, 885

Categorization

• SEED: Angle: how supposedly objective global rankings smuggle normative judgments into what looks like neutral statistical procedure.
• XREF: Connects to the reader’s interest in how measurement categories carry politics, e.g. colonial census classifications in Along the Archival Grain.