Way way too much stock is placed in that study.

For one, their total sample size was only 323 events, only 3 of which met the "3.5%" level. So the statement that change is inevitable based on only 3 instances is really crazy.

Further, none of those three instances had participants thinking that 3.5% was some sort of goal, it was a correlation. So now you have a lot of protestors treating 3.5% as a goal rather than some organic emergent property of the broader movement. Even if there was something inevitable about having a 3.5% participation rate when no one is aware of that metric, simply knowing of the metric can change a lot.