In a previous question I asked about the history of auto-correlation

In this question I'm interested in the discovery process behind cross-correlation.

  • When was it invented?
  • Was it independently invented multiple times?
  • What motivated its invention?

In my opinion correlation is synonymous with cross-correlation. If you accept that then Galton is your guy circa 1877. If your want the first time the word cross-correlation was used then Thorndike is your guy circa 1920.

In a psychology study published in 1920, Thorndike asked commanding officers to rate their soldiers; he found high cross-correlation between all positive and all negative traits. People seem not to think of other individuals in mixed terms; instead we seem to see each person as roughly good or roughly bad across all categories of measurement.

(From Wikipedia on the "Halo effect.")

More on Thorndike

Way back in 1920, just after the first burst of enthusiasm about then-new IQ tests, psychologist Edward Thorndike created the original formulation of "social intelligence." One way he defined it was as "the ability to understand and manage men and women," skills we all need to live well in the world.

But that definition by itself also allows pure manipulation to be considered a mark of interpersonal talent.

Daniel Goleman, Social Intelligence: The New Science of Human Relationships, p. 11.

| cite | improve this answer | |
  • 2
    $\begingroup$ Almost the entire reply was taken wholesale from two sources without any attribution. (I have edited it accordingly.) The lack of attribution is plagiarism, no matter in what medium it occurs. Do not do that again on this site. $\endgroup$ – whuber Jul 21 '11 at 17:29
  • $\begingroup$ OK didn't know and now I do know to attribute "stuff" $\endgroup$ – IrishStat Jul 21 '11 at 20:36
  • $\begingroup$ The book is copyrighted (of course) and the Wikipedia article links directly to a Creative Commons license and a terms of use page. $\endgroup$ – whuber Jul 21 '11 at 20:58

Hard to date an intuitive geometric concept. Maybe starting with the Gramian matrix? The use of the covariance matrix in statistics is very much related to the concept of Mahalanobis distances.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.