This is a very poorly done analysis. At a minimum she needs to define top school/top company. Also I'd like to see the confidence intervals around the effect si

This is a very poorly done analysis. At a minimum she needs to define top school/top company. Also I'd like to see the confidence intervals around the effect sizes. In addition, looking up MOOC information from LinkedIn may result in a lot of false negatives. (She doesn't mention if MOOC courses in non-CS subjects count.) Did all the interviewees have CS degrees? What about the Masters degrees, is she including non-CS ones? Is the sample of interviewees representative or there's any selection bias that we should be aware of? A study which doesn't answer so many basic methodological questions is garbage.