It is important to look at the background and qualifications of the coding bootcamp graduates who then (allegedly) land a high paying job. I have seen a number of cases where the graduate landed a so-called "data science" job but had a strong relevant background even without the boot camp. For example, an actuary might plausibly take a 12-week data science/coding bootcamp to learn or strengthen their knowledge of certain coding or other skills and then land a job as a "data scientist" as actuaries already have most of the statistical knowledge and intuitions from their work.

Data science in particular is drawing heavily on highly qualified applicants with backgrounds in statistics, mathematics, various sciences such as physics where there is extensive data analysis but not formal CS degrees. Many can benefit however from a crash course in some aspects of coding or certain hot data science tools. The job placement prospects for applicants without the strong background, relying solely on the coding bootcamp training, will probably be much poorer.