Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
mathstodon.xyz is one of the many independent Mastodon servers you can use to participate in the fediverse.
A Mastodon instance for maths people. We have LaTeX rendering in the web interface!

Administered by:

Server stats:

2.7K
active users

mathstodon.xyz: AboutStatusProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.7

#samplesize

0 posts0 participants0 posts today
Public
Joel Snyder

Application of JNDs to meta-science. Very sensible!

"For example, in clinical settings researchers may specify this smallest effect size of interest as the smallest difference in a health condition that patients themselves notice...this practice has been used at least since the advent of psychophysics in the second half of the 19th century"

#metascience #psychology #neuroscience #effectsize #psychophysics #samplesize

nature.com/articles/s41562-024

NatureEffect sizes and what to make of them - Nature Human BehaviourWe all care about effect sizes. Yet, traditional ways of evaluating them (P < 0.05 and generic benchmarks) are failing us. We propose two paths forward: setting better, contextualized benchmarks or — more radically — letting go of benchmarks altogether. Both paths point to adjusted expectations, more detailed reporting and slow science.
Public
katch wreck

`This review holds two main aims. The first aim is to explain the importance of sample size and its relationship to effect size (ES) and statistical significance. The second aim is to assist researchers planning to perform sample size estimations by suggesting and elucidating available alternative software, guidelines and references that will serve different scientific purposes.`

ncbi.nlm.nih.gov/pmc/articles/

PubMed Central (PMC)Sample size, power and effect size revisited: simplified and practical approaches in pre-clinical, clinical and laboratory studiesCalculating the sample size in scientific studies is one of the critical issues as regards the scientific contribution of the study. The sample size critically affects the hypothesis and the study design, and there is no straightforward way of calculating ...
#sampleSize#effectSize
Public
Jan R. Boehnke

I'll be offering an introduction to #simulation methods to determine #SampleSize-s for clustered / nested studies.

Apparently another popular session at #RMeF23 🙇

One of the classics that got me into this area is Ukoumunne's
onlinelibrary.wiley.com/doi/ab

I was always interested in how to straddle the overlap between observational #TherapistEffect studies* and #RCTs in this area.

#Rstats #StudyDesign

* eg., rdcu.be/dqi20

NCRM advert for the "How to" session on: How to use simulation methods for power analyses and sample size calculations in cluster-randomised designs Wednesday, 08.11.2023, UK-4pm
Public *
Jan R. Boehnke

One of the most problematic areas in the submission we get and papers I review are the sections on #SampleSize justifications.

See for example @lakens' excellent paper on the topic:
psyarxiv.com/9d3yf/
It now comes with a process guide in a #ShinyApp, which is an excellent support:
shiny.ieis.tue.nl/sample_size_
#Rstats

For #HRQL researchers:
rdcu.be/dnfO4

And many people teach this stuff, e.g., ☺️
researchgate.net/publication/3

#StudyDesign #NightshiftEditor
[edit: typo in hashtag]

Public
Jan R. Boehnke

Online #workshop:
Simulation-based power analyses in (generalized) linear mixed models
17.05.2023, 10-12h CEST

The workshop will cover basics of power analysis, linear mixed models, and why the combination of both requires a simulation-based approach.

In my experience, this is for many areas of #HealthSciences and #HRQL research a key problem when designing studies.

Maybe worth a read as well:
link.springer.com/article/10.3

Simulation-based power analyses in (generalized) linear mixed models Abstract The statistical power of a research design is closely linked to the reliability and replicability of empirical findings. Accounting for power while planning a study is therefore crucial and often a requirement for submissions in scientific journals. However, this can quickly become highly difficult in practice – especially for more complex, but very popular analysis procedures like linear mixed models (LMMs). In this workshop, we will briefly cover the basics of power analysis, linear mixed models, and why the combination of both requires a simulation-based approach. We will then focus on the R-package mixedpower and how to use it to estimate power in LMMs. The general aim of this workshop will be to help researchers build intuitions about simulation-based power analyses, and to empower them to set up highly powered research designs when they plan to use mixed-effect models to analyse the resultant data. A prerequisite for this workshop is a basic knowledge of R. Although we will briefly cover the basics of LMMs, familiarity with LMMs and the R-package lme4 is strongly recommended.
#PowerAnalysis#SampleSize#Rstats
Public
Harishjosev

My latest post - OC Curve and Reliability/Confidence Sample Sizes:

#SampleSize #Statistics #Reliability #Confidence #Minitab

I have had a lot of feedback on one of my earlier posts on OC curves and how one can use it to generate a reliability/confidence statement based on sample size, n and rejects, c. This post is mostly geared towards giving an overview of using OC curves to generate reliability/confidence values and using Minitab to do the same.

harishsnotebook.wordpress.com/

Harish&#039;s Notebook - My notes... Lean, Cybernetics, Quality &amp; Data Science.OC Curve and Reliability/Confidence Sample&nbsp;Sizes:By Harish
Public
Noémi K. Schuurman

(Interested in) Working with the random-intercept Cross-lagged Panel Model and unsure what sample size to use?

Jeroen Mulder fixed it for you:
Mulder, J.D., (2022). Power analysis for the random intercept cross-lagged panel model using the powRICLPM R-package. Structural Equation Modeling: A Multidisciplinary Journal.

jeroendmulder.github.io/powRIC

Other questions about the RI-CLPM?
jeroendmulder.github.io/RI-CLP

jeroendmulder.github.ioPower analysis for the RI-CLPMPerform user-friendly power analyses for the bivariate random intercept cross-lagged panel model (RI-CLPM). The strategy as proposed by Mulder (2022) <doi:10.1080/10705511.2022.2122467> is implemented. Extended power analysis options include the use of bounded estimation, inclusion of measurement error in the data generating model and estimation model (i.e., the stable trait autoregressive trait state, STARTS, model), imposing various constraints over time on the parameters of the estimation model, among others.
#SEM#longitudinal#PanelData
SearchLive feeds
About