selection: Correcting Biased Estimates Under Selection

Description Usage Arguments Value Author(s) References See Also Examples

caseIV uses Hunter and Schmidt's (2004) correction procedure to estimate the correlation between t and p, assuming direct selection on a latent s.

1	caseIV(rtpi, rxxi, rxxa, ux)

`rtpi`	the correlation between t and p in the incumbent pool.
`rxxi`	the reliability of x in the incumbent pool.
`rxxa`	the reliability of x in the applicant pool.
`ux`	the ratio of selected to unselected variance in x.

a scalar that is the estimate of the correlation between t and p.

Dustin Fife

Hunter, J.E., and Schmidt, F.L. (2004). Methods of meta-analysis: Correcting error and bias in research findings. Thousand Oaks, CA: Sage.

See Also rel.correction to obtain estimates of rtpi from rxyi, caseIII, em, caseIIIR

	# correct an xy correlation for unreliability
	rtpi = rel.correction(rxyi=.3, rxxi=.8, ryyi=.6)

	# estimate Case IV
	caseIV(rtpi=rtpi, rxxi=.6, rxxa=.8, ux=.8)