Like me, do you constantly question the measurement assumptions underlying psychological research? Is your first thought in the morning "I'm pretty sure Likert scales aren't normally distributed variables?" If so, do we have a new paper for you: https://t.co/AUkr7r4PgV