Just to add a little context, one issue she ignores entirely is the relationship between needs based entrepreneurship and the total entrepreneurial activity (TE

Just to add a little context, one issue she ignores entirely is the relationship between needs based entrepreneurship and the total entrepreneurial activity (TEA) rate. So, one of the effects you're going to run into are that the countries with the highest TEA rate are going to be 1.) developing economies (which have higher needs based entrepreneurship) that 2.) have reached a high enough development level to have functioning university systems and 3.) are in the process of adopting western-culture norms. As a result, all of the traditional determinants of entrepreneurial activity, such as capital availability, education infrastructure, &c... aren't going to matter that much because her DV only measures total entrepreneurial activity (e.g., person A opens a street stand because they don't have any other way of generating income) or variables that are heavily correlated with TEA, which isn't what most folks are really interested in when they talk about entrepreneurship. However, her sample is also going to be biased in favor of developing countries that have higher levels of normative western-culture beliefs (e.g., support for gender equality), because developing countries with university systems sophisticated enough to know about & participate in a project such as GEM (and generate good data) are also more likely to have already started the process of incorporating western norms into their culture. Gender equality is a massive value added, and I know a lot of people who have done actual research which proves it. But, you can't run a couple of regression models on GEM data and declare that capital access, geography, education, and all of the other traditional determinants of entrepreneurship aren't influential.