Some lightweight, likely to remain free models (e.g. the ones you can download right now), are quite good. I think a coder needs decent fundamentals, but that gen AI moves the balance point where higher coding expertise is no longer worth the opportunity cost of learning something else.

Non-invariance

Ah, but have you considered the hermeneutics of positionality statements through a post-modern ontological lens? A constructivist epistemology challenges the positivist Weltanschauung implicit when quantitative methods are imported to psychology !?!

Grit : persistence and passion

Paper's a bit sparse on "why". Do we figure this applies to interpreting any single correlation in a large matrix or it's something to do with regularization?

Student of mine got access earlier this year. I don't think they send a confirmation email, you just get access. Revelle is on bluesy, you could reach out directly to him

Soooo egg-plant emoji for quantifying power is out?!?

Scale development is viewed as iterative. If you have something to measure and your first attempt fails, you just keep trying. If you were to argue that that sounds more like engineering than science, I'd probably have to agree.

Self monitors probably do?

Yes! This is a debate I want to see. What if all alt causal directions would be stupid ? What if you fit a RICLPM on longitudinal data? What if you fit a well-powered mediation and results are null? Enough to disprove a theory? Do criticisms of cross-sect med apply to all cross-sect analyses?

Super impressive paper, can't wait to discuss it in class. Do you have any indications as to whether measurement non-invariance could explain some results? For example, could "I love flowers" refer to differing degrees of appreciating flowers in different languages ? Or was all data English?