Original Research

The discriminant validity of the culture assessment instrument: A comparison of company sub-cultures

Leo Petkoon, Gert Roodt
SA Journal of Industrial Psychology | Vol 30, No 1 | a141 | DOI: https://doi.org/10.4102/sajip.v30i1.141 | © 2004 Leo Petkoon, Gert Roodt | This work is licensed under CC Attribution 4.0
Submitted: 26 October 2004 | Published: 26 October 2004

About the author(s)

Leo Petkoon, Rand Afrikaans University, South Africa
Gert Roodt, Rand Afrikaans University, South Africa

Full Text:

PDF (286KB)

Share this article

Bookmark and Share


The primary purpose of this study was to assess the discriminant validity of the Culture Assessment Instrument (CAI); that is to distinguish between company mean sub-culture scores and between mean scores of a target company and that of a norm group. The primary data was obtained by a sample of convenience (N = 593) from a transport organisation. The secondary data of the norm group was constituted by convenience samples (N = 4066) from various companies originating from different industries. The 56 item scores of the CAI were factor analysed on two levels followed by iterative item analyses. Although significant differences were detected between mean culture scores, only a small proportion of the variance in these scores could be attributed to culture differences. On these grounds, the CAI does not possess discriminant validity. Suggestions for improving the CAI were made.

Die primêre doel van die studie was om die diskriminante geldigheid van die ‘Culture Assessment Instrument’ (CAI) te beoordeel; dit is om tussen ondernemings se gemiddelde kultuur-subtelling te onderskei en tussen die gemiddelde tellings van ‘n teiken onderneming en ’n normgroep. Die primêre data is verkry van ’n geleentheidsteekproef (N = 593) uit ’n transport-onderneming. Die sekondêre data van die normgroep is saamgestel uit geleentheidsteekproewe (N = 4066) van verskillende ondernemings afkomstig uit verskeie industrieë. Die 56 itemtellings van die CAI is op twee vlakke gefaktoranaliseer, gevolg deur iteratiewe itemontledings. Ofskoon beduidende verskille tussen gemiddelde kultuurtellings gevind is, kon slegs ’n klein proporsie van die variansie in die tellings aan kultuurverskille toegeskryf word. Op hierdie gronde beskik die CAI nie oor diskriminante geldigheid nie. Voorstelle ter verbetering van die CAI is gemaak.


Company sub-cultures


Total abstract views: 2427
Total article views: 3297

Crossref Citations

No related citations found.