Do different national samples yield similar dimensions of national culture?
Purpose – Cross-national studies of employees’ values and beliefs have extracted dimensions of national culture from diverse samples of employees. The purpose of this paper is to find out if this sample diversity impacts the nature of the extracted dimensions: is a given dimension replicable across diverse samples (such as managers vs skilled workers?). Design/methodology/approach – The authors analyzed a set of values from the World Values Survey, comparing nation-level value structures from four types of samples in 46 countries: national representation, managers, experts without supervisory duties, and skilled workers. The authors analyzed the data with, and simultaneously compared, two data reduction methods: multidimensional scaling (MDS) plots (Shalom Schwartz’s preferred method) vs exploratory factor analysis (EFA). Findings – MDS plots suggested structural similarity across the four samples, whereas EFA suggests divergence. Research limitations/implications – Whether dimensions of national culture replicate across different samples or not depends on the data reduction method. There is no one best method in an abstract sense. Researchers’ choice of method should be contingent on their research philosophy: theory-driven vs empirical. Originality/value – No such study has been published previously.