`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`

1 2 3 4 5 | ```
# 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)
``` |

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.