This is a very interesting breakdown of developer preferences. Obviously it's from 2018, but perhaps that is more reliable than attempting a survey with 2019 not yet over. I love these kind of articles because it opens my eyes to libraries and tools that I have never used and/or heard of before.
It's no surprise that Bootstrap is still up there on the top, but it's a shame that Bulma is so low at less than 3%. I recently found out about Bulma from a post on CF and I found that it was very self explanatory and in my opinion, better than Bootstrap. Another surprise was seeing Materialize at only 2.6%. With all of the themes it's used in, I expected it to be much higher. However, I haven't really liked Materialize, so maybe other developers avoid it too.
One thing I do wonder though is how many people lied. It's possible that some developers want to appear better than they are, even if it was an anonymous survey (this is actually a common thing - people wanting to impress the person conducting the survey, even when done online)... It's interesting how many people say they do not use a framework. It makes me wonder if there should be considerations about what the developers are working on, or how much development they are doing. If a developer finishes the survey, but is only working one day per week, should their answers be equal to a developer working 6 days a week? All I am saying is that it would be really interesting to see the statistics broken down on a per-usage level instead of a per-developer statistic. So instead of seeing that 50% of developers said they lint their CSS, we could see a statistic along the lines of 29% of produced CSS is linted. I think that would be much more representative of framework and tool usage compared to having each developer's answers weighted equally.