Application of a nomological model of selection validity

Bartmann, Tanya Justine (2001-04)

Thesis (MA) -- University of Stellenbosch, 2001.

Thesis

ENGLISH ABSTRACT: The purpose of this study was to determine the applicability of a nomological model for the assessment of psychological measurement validity with reference to personnel selection. The model, which can be traced to Thorndike (1949), Campbell and Fiske (1959), Nunnally (1967, cited in Duvenage, 1990), Guion (1974, cited in Duvenage, 1990), Binning and Barrett (1989) and Duvenage (1990), provides the evidential bases for two empirically founded validation strategies, namely criterion- and constructrelated validity. Theoretically, the Nomological Model for Psychological Measurement Validity Assessment proved to be applicable for validating that a measure is able to predict job performance. An easily understandable empirical investigation was required to verify the scientific utility of the model for selection validation. The model was used to validate the procedure for selecting machine operators in a large food manufacturer. The investigation illustrated that by utilising Route 1 of the nomological model, evidence of criterion-related validity, as described by legal and professional standards, is generated. Furthermore, the study indicated that by utilising Route 2 of the model, evidential requirements of construct-related validity as set by legal and professional standards are met. The results of the study proved that the Nomological Model for Psychological Measurement Validity Assessment is theoretically, professionally, legally and practically applicable for the validation of personnel selection procedures.

AFRIKAANSE OPSOMMING: Die doel van hierdie studie was om die toepasbaarheid van 'n nomologiese model vir die bepaling van die geldigheid van sielkundige meting ten opsigte van personeelkeuring vas te stel. Die model, wat sy oorsprong in die werk van Thorndike (1949), Campbell en Fiske (1959), Nunnally (1967, aangehaal uit Duvenage, 1990), Guion (1974, aangehaal uit Duvenage, 1990), Binning en Barrett (1989) en Duvenage (1990) het, verskaf getuienis basisse vir twee ernpirles gefundeerde valideringsstrategiee, naamlik kriterlum- en konstruk-verwante geldigheid. Daar is teoretiese bewyse gevind vir die toepasbaarheid van die Nomologiese Model vir Sielkundige Meting Geldigheids Raming vir die validering van metingsvoorspelbaarheid van werksprestasie. 'n Maklik verstaanbare empmese ondersoek was nodig om die model se wetenskaplike nut vir keuringsgeldigheid te bepaal. Die model was gebruik vir die validering van keuringsprosedure-geldigheid van masjienoperateurs in 'n groot voedselverwerkingsmaatskappy. Die ondersoek het ge"illustreer dat deur die toepassing van Roete 1 van die nomologiese model, bewyse gegenereer is vir kriterium-verwante geldigheid, 5005 voorgeskryf deur relevante wetgewing en professionele standaarde. Verder, deur die toepassing van Roete 2 van hierdie model, word voldoende bewyse vir konstruk-verwante geldigheid, 5005 voorgeskryf deur wetlike en professionele standaarde, verkry. Die resultate van die studie bewys dat die Nomologiese Model vir Sielkundige Meting Geldigheids Raming teoreties, professioneel, wettig en prakties toepasbaar is vir die validering van personeel keu ri ngsprosed ures.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/52122
This item appears in the following collections: