I'm a doctoral student at Umeå University attached to Umeå School of Business, Economics, and Statistics. My main supervisor: Norbert Steigenberger (Professor at Umeå University). Cryptocurrency market offers a valuable empirical setting to examine how technology-driven nascent markets are created and allow entrepreneurial activity to thrive. At the inception of the initial coin offering (ICO) market, there were no regulations governing market exchanges. However, start-ups continue to raise funds and remain operational. The mechanism that guided market exchanges in a formal institutional void is called informal institutions. In other words, these constitute rule systems created by the community and are not based on law. Currently, we have limited knowledge of how these informal institutions get created in nascent markets and influence entrepreneurial activity. Therefore, my dissertation project attempts to answer the research question, "How do institutional arrangements led by informal institutions get created in nascent markets, govern the market, and influence entrepreneurial activity?" The allocation requested here supports the measurement and analysis work with regard to my first paper of the dissertation. As I'm deriving the latent construct "community governance" by analysing community discourse in Bitcointalk ANN threads, most of the analysis will include NLP. I have already derived the data corpus, which includes 3.9 million forum posts. The material is publicly accessible and contains no personal or otherwise sensitive data; contributors post under self-selected pseudonyms and no attempt is made to identify individuals. The analytical strategy proceeds in two stages. The community governance construct is built by applying supervised text classification across the full corpus. The classifier is trained on a manually annotated sample (ca 1,800 posts) which were coded against an explicit codebook, and further annotation proceeds through an iterative sampling procedure that prioritises informative cases. Classifier performance is assessed against a randomly drawn evaluation sample (600 posts) so that population-level quantities are estimated without bias from targeted sampling. The resulting monthly series is analysed alongside other variables that are already constructed using time-series analytical methods (i.e., VAR and Granger causality). The intended contribution is to see how informal institutions, which include community governance, came to develop in the ICO market and whether it had an impact on market outcomes.