The halo effect: It really isn't unitary: A rejoinder to Nathan (1986)

This paper is in response to Nathan's (1986) critique of Fox, Bizman & Herrmann's (1983) paper ‘The halo effect: Is it a unitary effect’. Specific points raised in Nathan's critique are rebutted. Recent numerical analyses are then employed to demonstrate the non‐equivalence of two extant measures of halo‐SD indices and D × D correlations. The SD indices, challenged by Nathan, are shown to conform with at least one commonly accepted definition of halo and to bear upon important psychological phenomena in the rating process. The SD and D × D measures are shown to reflect two different types of halo ‐ co‐occurrent and covariant. Conceptual analyses are offered to further elucidate the qualitatively differing foci of these types of halo. Summary sections address the implications of this diversity in halo for programmes aimed at reducing rating error. 1989 The British Psychological Society