The Nature of the Data, or How to Choose a Correlation Coefficient  

Oleh:

Carroll, John B.

Jenis: Article from Journal - e-Journal
Dalam koleksi: Psychometrika: A Journal of Quantitative Psychology vol. 26 no. 4 (Dec. 1961), page 347-372.
Topik: Errors of Scaling; Errors of Scale-Dependent Selection; Errors of Measurement;
Fulltext: 2604_1004.pdf (1.34MB)

Isi artikel

Some of our more self-critical psychometric brethren have reminded us (e.g., Dunlap [8]) that the purpose for which our society was founded--" the development of psychology as a quantitative rational science"--implies that our primary attention should be focussed on the formulation and testing of mathematical models for behavior, not on the elaboration of statistical methodology. I would very much like to have devoted this address to a new mathematical formulation of some aspect of behavior, something that would clearly fall within the purview of the Psychometric Society. Unfortunately, I have found myself encumbered with a good deal of unfinished business, both in and out of psychometrics, and time for mathematically formulating behavior has not been plentiful. But after reflection I have decided that some of my unfinished business does indeed have a solid connection with the rationales by which we quantify behavior, and it is about this that I wish to speak.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali