The results are also selected from a StackOverflow survey. It is hard to tell if SO survey respondents overrepresent higher-earning remote workers or not.
Yup, this comment should be at the very top. The first problem with this analysis of job salaries is it's based on a self-selected survey and not a random sample.