I'm sure the prediction rate would be high. higher = better in those in most ways. My question is really more about discrimination power given range restrictions all around. lmk if you ever wanna plot this study. it's super interesting and relevant to academic culture for sure

I wish. I remember seeing a poster early in process. It was by Buce Leise about 10 years ago. struggling to find.

mentors product resources.
resources predict outcomes.

it sounds more like a bias of "anyone can do it if you try hard enough ", which aligns to messaging from lots of deans even when on spite of fact🤣

totally. I was thinking about the model to test it and it gets lost to me in the contextual and situational factors (program opportunities, first gen, faculty success/career stage, data access, etc.). I saw some work from KU years back looking at student success and it was all mentor related

I agree with you on an empirical question but like, I wonder what the mean C of a competitive market applicant is (job or student even). High N + C seems to be the standard course by default.

Oh yeh 100%. I'm not even sure what that would mean. seems akin to GRE predicts academic ability approach to testing... so bad in other words (not that it has to be, I agree)

I don't disagree that assessments can be useful. I'm less than impressed with their actual behavioral prediction.

Not sure why that makes me have two heads.. Measuring C is likely not gonna give us much. We are dealing with a range restriction on the upper end no matter the level. my point is that our assessments don't target these things so the purpose isn't suited...snow tires are good, but not for everything

only weird if the measure isn't aligned to the outcome. I'm not sure most assessments are well suited for the type of prediction wanted, or nuanced enough when measuring broad traits to tap it.

also. most don't assess mentor success, which is a large predictor 🤷‍♂️

We need a section of the news letter reporting on porch personality of the board lol

according to the SSRI hypothesis,

no. likely not

totally. I also think the gate keeping at an editorial level is another very easy target to increase desire and ability to achieve a more open science.

and also,

if someone is 'taking shots' vs giving research criticism, the editors should be aggressive about that. I sign when my statements may be a conflict of interest or when I see it as a 'shot' in any way now that I'm tenured. wouldn't dream of it before.