Show Summary Details

Quick Reference

Constructing a linear model of a judge's predictions of a criterion variable and then using the linear model instead of the judge to make further judgements or predictions. Such linear models consistently outperform the judges from whose judgements they are derived, because the coefficient weights are distillations of the judges' underlying principles, free of the extraneous variables that inevitably influence each of their specific judgements. This technique became widely known in the early 1970s under the leadership of the US psychologists Lewis Robert Goldberg (born 1932) and Robyn Mason Dawes (born 1936). In one typical study, a group of judges studied ten background, aptitude, and personality measures taken from graduate students and predicted the students' grade point averages in the first year of graduate study. Linear models of every one of the judges performed better than the judges themselves in predicting the students' grade point averages. See also actuarial prediction. bootstrap adj. vb. [So called because it is reminiscent of lifting oneself up by one's own bootstraps]

Subjects: Psychology.

Reference entries

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.