Big Data Analysis of Entrepreneurial Intention and Success: The Role of Individual and Social Network Factors with Edu-cation and Experience as Moderators — Evidence from GEM APS Survey

Abstract

Entrepreneurship has become a key driver of economic growth and employment in developed and developing economies. However, it depends on socio-economic status, educational level, experience-based learning, and the quality of institutional backup. Although policymakers and scholars have continued to analyze this terrain, there remains considerable uncertainty about determinants interrelates to either nurture or hinder entrepreneurial mobilization. To bridge these gaps, the current study aims to contribute a systematic examination of entrepreneurial activity by investigating big-data analytic techniques. Using the Global Entrepreneurship Monitor (GEM) Analysis of the Pulse Survey (APS) data on Bahrain, the study formulates eight hypotheses. The systematic analysis process involves variable selection, data cleaning, and preparation using statistical and computational methods like correlation analysis, Chi-Square tests, ANOVA, and logistic regression with p-value testing. SPSS was used to run the statistical tests in the study. Python scikit-learn library was used to build machine-learning models. Among the analytic suite, there were logistic regression, decision-tree, and random-forest classifiers. Empirical evidence on the multifaceted drivers is evident from findings, highlighing personal characteristics, social networks, and contextual moderators. The study contributes to entrepreneurship theory and policy development, offering insights for fostering entrepreneurial ecosystems in Bahrain.