FIONA MacPHAIL
MOVING BEYOND STATISTICAL VALIDITY
IN ECONOMICS
*
1. INTRODUCTION
Despite the wide use of economic indicators, general methodologi-
cal discussions of their validity, or systematic discussions of validity
theory in which types of validity are defined and evaluated, do not
occur within Economics. For example, a literature search conducted
on the Journal of Economic Literature CD Rom data base, using the
key word ‘validity’, generated about 800 articles where validity was
present in the title or abstract.
1
Based upon a review of the titles,
however, there was not a single article providing a comprehensive
analysis of the meaning of validity and its role in economic analysis.
Further, the term validity is not an entry in the New Palgrave Dic-
tionary of Economics, nor is it present in the index of a sample of
standard econometrics textbooks.
Although methodological analyses of validity are lacking in Eco-
nomics, the term ‘validity’ is used in the applied and theoretical
economics literature. In applied economics, validity tends to be
equated with the accuracy or ‘truthfulness’ of results. In assessing
the empirical validity of results, economists give high priority to
statistical analysis.
2
In other social sciences, in contrast to Economics, validity is care-
fully defined and the historical evolution of the notion of validity
has been analyzed (see, for example, Hubley and Zumbo, 1996).
An illustrative definition of validity is provided by Messick (1993:
p. 13), who in the context of educational measurement, states that
validity is “an integrated evaluative judgment of the degree to
which empirical evidence and theoretical rationales support the ade-
quacy and appropriateness of inferences and actions based upon test
scores or other modes of assessment.”
3
Validity of inferences drawn
Social Indicators Research 45: 119–149, 1998.
© 1998 Kluwer Academic Publishers. Printed in the Netherlands.