Kevin Murphy

Who they are

Murphy's affiliation is given here as it appears on the 1993 paper, which is the work this site draws on.

This page records only what the paper and its published record state. Biography beyond the cited work is not given here.

Key papers

Nature and consequences of halo error (1993), with Jako and Anhalt. The paper reviewed the definition of halo error that had dominated research since Edward Thorndike named it in 1920. That definition held that halo error was common, that it was an error of the rater with true and illusory parts, that it inflated correlations between rating dimensions because a general impression shaped specific judgements, and that it was harmful and should be removed. The authors reviewed research showing that each major element of this view was either wrong or problematic.

Their central concern was that true halo, where qualities really do go together, and illusory halo, where a rater's impression spreads, are confounded in most studies. Because of that confound and the unclear consequences of halo errors, they proposed a moratorium on using halo indices as dependent measures in applied research, and called for work that separates actual halo errors from the apparent halo effect.

Key ideas

The paper is the main counterweight to treating every correlation between ratings as bias. If capable people really are capable in several ways, ratings that agree may be accurate.

The critique is strongest against correlational evidence such as Thorndike's ratings of officers. It has less force against experiments that hold the person or product constant and still find judgements moving, such as Nisbett and Wilson's 1977 study of a lecturer filmed in a warm and a cold version.

---