https://www.allsides.com/news-source/quillette-media-bias
Is it possible that the conservative behave poorly more often on Twitter and thus get more sanctions?
Is it possible that the content moderation on Twitter, which WAS algorithmic + reporting based, showed more reported acts for conservatives?
Is it possible that Trump supporters said more shitty things (racism, sexism, etc) than Clinton supporters did?
I think you'd need to look more closely at the survey construction to tell if any variable isolation was done at all.