Abstract:Student start-ups are a significant part of overall university entrepreneurship. Yet, we know little about the determinants of this type of start-ups and, specifically, the relevance of context effects. Drawing on organizational and regional context literature, we develop and test a model that aims to explain student entrepreneurship in a contextual perspective. Based on unique micro data and using multi-level techniques, we analyse nascent and new entrepreneurial activities of business and economics students at 41 European universities. Our analysis reveals that individual and contextual determinants influence students' propensity to start a business. While peoples' individual characteristics are most important, the organizational and regional contexts also play a role and have a differentiated effect, depending on the source of the venture idea and the stage of its development. Organisational characteristics, like the prevalence of fellow students who have attended entrepreneurship education, influence whether students take action to start a new firm (nascent entrepreneurship) but do not seem to support the actual establishment of a new firm. In contrast, the latter is less dependent on the university context but more strongly influenced by regional characteristics. Overall, our study contributes to our understanding of the emergence of start-ups in the organizational context of universities and has implications for initiatives and programs that aim at encouraging students to become entrepreneurs.

While the importance of a supportive context for entrepreneurship is widely acknowledged, its antecedents are rarely investigated. We apply the concept of organizational climate to higher education institutions and examine the drivers of students' perceptions of the entrepreneurial climate in their university. Combining data from two unique datasets and using multilevel techniques, we analyze the relationship between university characteristics and such climate perceptions of 8009 students at public universities in Germany. We find university entrepreneurship measures to have a positive effect on students' climate perceptions, which also depend on students' background and gender. In addition, we find evidence for different peer effects, depending on students' affinity for entrepreneurship. For the general student population, including entrepreneurship content in their normal studies seems to be required to initiate a social process of sensemaking. However, students' perception of the entrepreneurial climate only depend to a certain degree on intentional entrepreneurship measures. In our study, general university characteristics have the strongest influence on climate perceptions. Overall, our study adds to our understanding of which parameters are important for establishing a more favorable and inspiring climate for becoming an entrepreneur at higher education institutions.

This paper seeks to better understand the significance of spatial context conditions and personal attributes for early-stage entrepreneurship. We combine individual with regional and national level data using multilevel analysis to test our hypotheses. We differentiate between two phases in the entrepreneurial process as well as between general and ambitious entrepreneurship. First, we show that both the national and the regional context significantly impact individual entrepreneurial activities. Second, individual level characteristics exert the greatest overall influence, but the direction of this influence is not stable. Third, the impact of the three levels varies across the different phases in the entrepreneurial process as well as between different types of start-ups. Fourth, we demonstrate that cross-level interactions between individual characteristics and spatial context factors are important in explaining entrepreneurial activities.JEL classification: L26, M13, J23 New firm creation from a multidimensional perspectiveThere is broad consensus within entrepreneurship research that individual entrepreneurial activities cannot be explained without considering contextual factors (e.g., Hindle 2010; Nielsen

No abstract