Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social)

PhD Student in Virtual reality | Media Psychology & Neuroscience | Psychometrics and statistical models in CMC | Gender, sex and sexuality studies 🏳️‍🌈

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

A special thanks to the co-author, @AHildePsych, whose expertise as one of the developers of this method greatly enhanced this tutorial.

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

I'm not great at promoting work on social media, but I wanted to share a tutorial I wrote on Local Structural Equation Modeling (LSEM). Hope this can be helpful for those interested in exploring moderation effects (including measurement invariance) in SEM. https://t.co/EKorT4VWZW

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

This reduces the chances of the model misspecification of the moderation effect, which can happen when relying strongly on inappropriate theoretical assumptions (such as misusing a cubic polynomial curve fit what should be a bounded S-shaped age development effect).

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

Just had a frustrating experience: My colleague and I were teaching a stats course and couldn’t replicate a published paper’s multinomial logistic regression using their OSF dataset. After we contacted the authors, they made the entire OSF private! Good job!🤮 #openscience

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

Even with the challenging weather, it was a blast to run our workshop on nonlinear moderation! @franikowsp @AHildePsych Huge thanks to DGPs2024 for the great organization @univienna. #DGPs2024

[Xinkai Du @XinkaiDu]

Day 1 in beautiful Prague at #IMPS2024. Come to New Building Room 4C at 2:30pm for a symposium on network psychometrics! Except for the fit measure evaluation study, the symposium covers diverse topic on CAT, comparison, missingness and non-normality in network analysis.

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

To share more This study data was also open-accessed used in another paper to re-examine the reliability of the perceptual matching task. @hcp4715 Having more collaborative research practices that I am proud to be a part of. https://t.co/Rju4MAkeWq

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

I am excited to share that after two years, a non-significant result of my PhD experiments has finally been published. #OpenScience #PhDLife https://t.co/fAXFEyUcyA

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

However, our results did not provide evidence for a modulation of self-prioritization on facial emotion recognition using Bayesian analysis, regardless of emotion type.

[Dr Cassie Short @cassieannshort]

Had such an amazing time at #dppd2023 with my wonderful colleagues. Here are some snaps of them being incredible 👏 @angonesegiulia @kristantodan12 @LiuThorsten and Jing 🌟

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

@sachaa01 Dear Selena, are you still interested in switching? I have two tickets for Gelsenkirchen (17.07), and I really want to have tickets for Hamburg.

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

Just wrapped up an incredible conference on psychological methods. #EAM2023 Only downside? My trip home, thanks to a nearly five-hour delay from the “pünktlich” Deutsche Bahn, saw me reaching home at midnight.

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

Summer well spent at the SMiP summer school at Uni Mannheim. Presented my poster and connected with a community of cool researchers. Excited for what’s next!

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

@KouMurayama @KlintKanopka @Ed_Donnellan Yes, In theoretical level, I really understand this paper want to invent a measurement model. however, when you use random effects, it implicitly assumed a latent variable (not theta, but a latent method factor). in the statistical level, as you said it is only small difference.

[Tuo(Thorsten) Liu (🦋@liuthorsten.bsky.social) @LiuThorsten]

@KouMurayama @KlintKanopka @Ed_Donnellan I am only a tiny new newbie in psychometrics, but I personally(!) think this study is trying to reinvent the wheel. There is already random IRT (https://t.co/fSFBvDLdza…), and the model in this paper looked like is the 2PL + normal continuous response (De Boeck, 2008)...

[Nico Andersen @andersen_ni]

The true reason why Italy banned #ChatGPT #OpenAI