Understanding what determines entrepreneurial activity in countries and regions has motivated scholars for decades. In recent years, researchers began to understand the various actors and factors that influence entrepreneurship as a spatially embedded ecosystem. This cumulative dissertation examines the relevance of individual elements of such entrepreneurial ecosystems (EEs) for entrepreneurial activity, with a specific focus on the role of start-up competitions as policy instruments within EEs. The dissertation includes four scientific articles, each of which addresses one element of EEs in detail. First, a meta-analysis synthesizing existing knowledge on determinants of entrepreneurial activity is presented, and the statistical effect sizes of the relationships between each element of EEs and entrepreneurial activity on different spatial levels (local, regional, country) are provided. Subsequently, start-up competitions are examined in order to determine what role they play in EEs. A conceptual article illustrates the manifold interdependent relationships between start-up competitions and different ecosystem elements. In the next article, two start-up competitions in the contrasting EEs of Berlin and Hannover are analyzed based on 45 qualitative interviews. This is followed by an article in which a subset of these interviews is analyzed to further explore the learning processes of participants in the competitions.Findings show that a) the EE approach has strong explanatory power regarding entrepreneurial activity, b) the relevance of its elements depends on the spatial level of observation, c) start-up competitions anchor public entrepreneurship support in the two analyzed regions and provide networking opportunities, d) this both influences the ecosystem and is influenced by the ecosystem and the region, and e) entrepreneurial learning in the competitions is not affected by the region or ecosystem but rather by participants’ prior experiences. Overall, this dissertation contributes to knowledge of EEs and indicates that there are generic functions underlying the mechanisms of action of each element of the ecosystems (such as providing knowledge or capital), but that the form of these functions is influenced by (regional) context. The dissertation closes with a summary of the main findings, and implications for policy and further research are presented.