Definition of The Fisher Transformation: Hypotheses about the value of r, the correlation coefficient between variables x and y of the underlying population, can be tested using the Fisher transformation of a sample's correlation coefficient r. Let N be the sample's size.

This transformation is defined by:

z = 0.5 * ln ( (1+r)/(1-r) )

z is approximately normally distributed with mean r, and standard error 1/((N-3)^0.5).

This is a common way of testing whether a correlation coefficient is significantly different from 0, and hence ascribing a p-value. (Econterms)

Terms related to The Fisher Transformation:

• p-value

Writing a Term Paper? Here are a few starting points for research on The Fisher Transformation:

Books on The Fisher Transformation:
None

Journal Articles on The Fisher Transformation:
None

<Return to Main Page>

Discuss in my forum