Lots of questions of method here. Cohort comparisons require large and demographically similar samples, identical assessments, and appropriate covariates.
I was particularly concerned that response items were lumped together into one category of "violence" indicating any support for violence. I don't agree that supporting violence should label someone as "violent," which is misleading even when you justify it. I find the apparently greater support for violence among conservative students now, when compared to 2021, difficult to explain.
This needs to report Ns, much more detail on the method, more sophisticated stats, and p values. Were the same colleges sampled? The same areas of the country? What happens when you don't collapse categories for the support for violence question, which is rarely psychometrically justified? The Charlie Kirk murder makes it very difficult to draw time-based conclusions from 2025.
Agreed re: lumping together rarely with sometimes and always.
Rarely is bad but respondents may be signaling something a bit more nuanced there than those saying sometimes and always. It feels like FIRE wants to tout the biggest, scariest numbers.
That's good observation. Violence is how nearly all differences were solved in ye olden times and after Kirk's murder and the reaction, I'm all out of words.
Lots of questions of method here. Cohort comparisons require large and demographically similar samples, identical assessments, and appropriate covariates.
I was particularly concerned that response items were lumped together into one category of "violence" indicating any support for violence. I don't agree that supporting violence should label someone as "violent," which is misleading even when you justify it. I find the apparently greater support for violence among conservative students now, when compared to 2021, difficult to explain.
This needs to report Ns, much more detail on the method, more sophisticated stats, and p values. Were the same colleges sampled? The same areas of the country? What happens when you don't collapse categories for the support for violence question, which is rarely psychometrically justified? The Charlie Kirk murder makes it very difficult to draw time-based conclusions from 2025.
Agreed re: lumping together rarely with sometimes and always.
Rarely is bad but respondents may be signaling something a bit more nuanced there than those saying sometimes and always. It feels like FIRE wants to tout the biggest, scariest numbers.
yes, rarely may be bad but it is a long way from always.
Brutally escalates when war escalates, wow big news.
After Charlie Kirk I no longer believe in conversations.
we’re heading towards violence. Its ok, it’s how difference were solved in most parts of history.
That's good observation. Violence is how nearly all differences were solved in ye olden times and after Kirk's murder and the reaction, I'm all out of words.
The endgame of cancel culture is not subdued conservatives forced into submission. It's Mutually Assured Destruction.
A strange game, the only winning move is not to play...