Multilevel Analysis of Entrepreneurial Activity: Exploring Individual-level Determinants and the Moderating Role of National Culture

Abstract

In this paper, we argue that national culture is important for explaining variations in entrepreneurial activity across countries. We examine both direct and interaction (moderation) effects of culture on entrepreneurial activity by applying a hierarchical logistic regression model for multilevel analysis. We use individual-level data for 84 countries that participated in the Global Entrepreneurship Monitor (GEM) from 2009 to 2013 (969,246 observations) merged with national-level data on the cultural context from GEM's National Expert Survey (NES). The results of our empirical analysis imply that a supportive culture towards entrepreneurship not only positively affects the number of entrepreneurs but also the allocation of entrepreneurs towards higher shares of women and lower educated individuals, arguably two demographic groups in the labour market with lower confidence levels. These findings are of particular relevance for the future development of entrepreneurship and understanding of existing relationships between individual and cultural factors. This focus also provides insight into how entrepreneurship can be designed according to the cultural context, serving future policy development.