Abstract

This chapter explains the concepts and to consider the numerous assumptions involved in these procedures. A weight equal to unity corresponds to the assumption that the observed measure and the construct have the same unit of measurement. Psychological test theorists and econometricians usually make this assumption, whereas path analysts and factor analysts commonly assign the factor a variance of unity. A stochastic component of behavior would seem an essential in economics." This comment probably applies to all the social sciences in which it is generally necessary in linear structural models to assume that all the other unmeasured variables influencing a variable of interest are independent of the measured influences. The emphasis on different methods of measurement represents an attempt to insure that the correlations among variables as much as possible represent commonality with the underlying trait rather than consistencies due to similarities of testing methods.

Keywords

Computer science

Related Publications

Publication Info

Year
2017
Type
book-chapter
Pages
270-292
Citations
22
Access
Closed

External Links

View on DOI.org

Social Impact

Altmetric
PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

22
OpenAlex

Cite This

Charles E. Werts, Robert L. Linn, Karl G. Jöreskog (2017). Quantifying Unmeasured Variables. , 270-292. https://doi.org/10.4324/9781351329088-11

Identifiers

DOI
10.4324/9781351329088-11