Moderator veranderlikes en differensiele geldigheid van die aansoekvorm

Abstract

Empirical studies have shown the Weighted Application Blank (W.A.B.) to be an effective predictive device in the selection of employees. There are, however, studies which suggest that the predictive validity of the W.A.B. may be substantially overstated in the literature. Contrary to the principles of classic psychometric theory it is suggested that moderator variables are accountable for this paradox. This study demonstrates how biographical correlates of tenure function as moderators and empirical evidence of the moderating effects on the predictive validity of the W.A.B. are provided. It is suggested that a new approach to the development of the W.A.B. is necessary in order to utilise the effect of moderators.

Opsomming
Empiriese studies toon dat die Geweegde Aansoekvorm (G.A.V.) 'n doeltreffende voorspellingsinstrument in personeelseleksie is. Daar is egter studies wat voorspel dat die voorspellingsgeldigheid van die G.A.V. in die literatuur aansienlik oorbeklemtoon word. Teenstrydig met die beginsels van klassieke psigometriese teorie word voorgestel dat moderator veranderlikes vir hierdie paradoks verantwoordelik is. Hierdie studie toon hoe die biografiese korrelate van dienstyd as moderators fungeer en empiriese bewys van die modererende effek op die voorspellingsgeldigheid van die G.A.V. word gelewer. Daar word voorgestel dat ‘n nuwe benadering tot die ontwikkeling van die G.A.V. noodsaaklik is ten einde die effek van moderators te benut.