Level 5 is this
(From a paper a colleague reviewed; N = 11 btw)

I also lay out an approach for estimating any nonlinear function as a linearized SEM through the use of Taylor series expansions. And so you don't have to do the math, I did it for you! With automatic {lavaan} syntax generation using the {leni} package function leni_sem().

Which let's you do all sorts of cool things, like test whether boys really are developmentally delayed compared with girls. Or using a multiphase version, to model puberty or cortical thinning.

General approach is to define a set of transformations for a linear function to a corresponding set of nonlinear parameters that more closely match developmental features of interest. For instance, the timing of developmental transitions or the height of a quadratic peak. See a cubic example below.

New paper+package combination for linear estimation with nonlinear inference (LENI) - methods to obtain the results of a nonlinear model without the hassle of nonlinear estimation. Presented this at Flux this fall and really excited to see it out!

osf.io/preprints/ps...
github.com/E-M-McCormic...

What a wonderful thing to come back from vacation to read! First publication at Psych Methods - feeling really proud of this one.

Reviewers can’t critique my paper if they can’t stand to look at it. Thanks to @rogierk.bsky.social for the “inspiration”?

PubMed now comes with a “the Greek letters YELL at you” feature. 😂