Major aMNLFA bug fix! TL;DR: If you ran the old version (i.e., anytime before July 2021) without inspecting the Mplus inputs and outputs and altering them accordingly, you should rerun them using the new version now, and check your inputs and outputs! https://t.co/8YJABAj0w9
Don’t miss your chance to read new Cambridge Element Algorithms for Measurement Invariance Testing by @VeronicaTCole and Conor H. Lacey Free access available until 19 December.
https://t.co/ssS4Vc4UH7
#cambridgeelements#psychology
Free measurement invariance advice alert! @ConorHLacey and I just published a Cambridge Element on algorithms for testing measurement invariance. We tried to make it pretty comprehensive, and it’s available for free until 12/19! Check it out here: https://t.co/2xULuddxvw
Super proud of this systematic review and meta-analysis that just came out in Psychology of Addictive Behaviors that just came out in collaboration with @LRakerd, @LydiaBierce, and others https://t.co/3eRkYtgPpl
I'm not sure I've ever learned more from writing a paper -- both about the literature and about the process of getting our heads around the literature. It was the longest time I've ever spent writing a paper (1.5 years!!) and I'm really proud of what our team produced!
By contrast, the relationship between substance use and things like peer-reported friendship nominations or peer-reported likeability are weak or null, in the aggregate.
New simulation paper comparing performance of 2 newer SEM fit index methods (dynamic fit cutoffs, equivalence testing ) to traditional cutoffs (RMSEA < .06, CFI > .95).
Both methods were improvements, but dynamic fit were most consistent across models
https://t.co/2mRAssjXyO
Footnotes:
* OK, it may have come out 2 months ago. I don't know why I waited so long to share it, but I'm excited to hear your feedback!
** The average correlation among factor scores was ~.9 - so, not "close enough" for all applications (e.g., educational testing).
New-ish paper! I'm really excited about this multiverse analysis of integrative data analysis (IDA) we (including @NishaGottfredso, Andrea Hussong, Dan Bauer, and Patrick Curran) just* published in Prevention Science https://t.co/hWKMm00Zii 1/n
Getting the DIF model "right" may be less important than assessing the robustness of your focal findings (usually pertaining to the relationship between the factor scores and some other variable) to decisions made during the scoring process! /end