Myself and Rob Davies (Lancaster, UK) surveyed 163 researchers and 400 papers - we found widespread concerns about the use of linear mixed-effects models (LMMs) and huge diversity in how they were reported 2/
Prikaži ovu nit
Hooray! 5 years' work, some great reviewers and 'Best practice guidance for linear mixed-effects models in psychological science' is in press at Journal of Memory and Language. 50 free copies: https://authors.elsevier.com/c/1aUF4,H2pbYbqT … TLDR: FOR GOODNESS SAKE REPORT MODELS & RESULTS IN FULL. 1/
-
-
-
In response to that we summarise best practice for LMMs (model building, model selection, choosing random and fixed effects) and focus on how results can be reported transparently and clearly. 3/
Prikaži ovu nit -
This paper took 5 years, contains a piece of my heart and soul and was definitely worth it. Like all the best papers, we wrote it for ourselves as much as for anyone else. 4/ end
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
Available at https://eprints.lancs.ac.uk/id/eprint/140728/ … DOI link https://doi.org/10.1016/j.jml.2020.104092 … Pre print at https://psyarxiv.com/h3duq (but published paper is quite different, reviews improved it substantially)
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
This looks really useful. I like LMMs, but there is such a variety of ways to fit and report them. As a reviewer I have asked authors to report the full model and been rebuffed.
-
By far the most worrying thing we saw was 'under-reporting' (e.g. not reporting the model equation, betas/coefficients, or random effect variances). Almost impossible to understand the analysis without this information - let alone meta-analyse or replicate. Hope the paper helps!
Kraj razgovora
Novi razgovor -
-
-
-
Science twitter can be very helpful!
Kraj razgovora
Novi razgovor -
-
-
Those Elsevier links don't redirect for me — could you share the DOI? makes it easy to look up various ways
-
Same for me, here you go: https://doi.org/10.1016/j.jml.2020.104092 …
- Još 1 odgovor
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.