As the author mentioned, “there is inherent selection bias in these results since it only surveys people who chose to use React Native.”

These are employees that already chose to use X, have successfully worked with X for months or years, and who are financially incentivized to stay with X based on the knowledge/skills developed over time.

If you were going to do a similar survey of native development it would have to be with participants that are equally invested in that skill set.

And of course none of this speaks to the context of the survey. Did they know the data would be used to make a strategic choice? Did they have certainty that the responses were anonymous? Might they have thought a negative response would affect the projects they worked on? What about other social or career implications? This kind of survey can easily be manipulated, so I wouldn’t put much weight in the numbers without more info.