MBS 95-02
The Ongoing Dialog Between Empirical Science and Measurement Theory
R. Duncan Luce
This article attempts to highlight some of the major developments in
the representational theory of measurement during the past 50 years with,
perhaps, a somewhat personal slant on what is included. Some emphasis is
placed on the ongoing interplay between abstract theory development and
attempted empirical applications. The article has three major sections.
The first concerns classical representational measurement, which was the
successful attempt to formulate the major measurement methods of classical
physics: extensive and additive conjoint structures, their distributive
interlock in dimensional analysis, and intensive (averaging)structures.
The second, which is called contemporary representational measurement,
somewhat overlaps the classical one but includes new findings: representations
of non-additive concatenation and conjoint structures, the theory of scale
types, results for general homogeneous structures, structures with singular
points, generalized distributive triples and the possible enlargement of
dimensional analysis, and the concept of meaningfulness. The last section
concerns two areas of applications of these ideas: psychophysical scaling
and decision making under risk.