jasp-stats / jaspreliability Goto Github PK
View Code? Open in Web Editor NEWThe Reliability Module
License: GNU General Public License v3.0
The Reliability Module
License: GNU General Public License v3.0
needs new Bayesrel
version (0.7.0.6) because there are no priors saved for only two items
While making the verified unit test I see that that the "mean" results from the new module differ from the previous results.
In particular, the verified results show a mean of
https://jasp-stats.github.io/jasp-verification-project/reliability-module.html
whereas the new mean is 17.55. Julius, is this related to the mean of the sum scores (17.55) vs mean of the (column)means? See also the remark
JASP calculates the overall mean and SD across all questions, whereas other software calculates the overall mean and SD across subjects.
options <- analysisOptions("reliabilityUniDimFrequentist")
options$omegaScale <- TRUE
options$alphaScale <- TRUE
options$lambda2Scale <- TRUE
options$lambda6Scale <- TRUE
options$meanScale <- TRUE
options$omegaItem <- TRUE
options$alphaItem <- TRUE
options$lambda2Item <- TRUE
options$lambda6Item <- TRUE
options$itemRestCor <- TRUE
options$itemMean <- TRUE
options$itemSd <- TRUE
options$omegaAnalytic <- TRUE
options$variables <- c(paste("Question", c(1, 4:8), sep="_0"), paste("Question", 10, sep="_"))
set.seed(1)
results <- jaspTools::runAnalysis("reliabilityUniDimFrequentist",
paste0(packagePath, "/tests/testthat/Reliability.csv"), options)
For the Bayesian part of the module it makes sense to also have the ESS as an indication of autocorrelation. How should this be calculated? If we have 3 chains with 1000 iterations each, the ESS is computed for each chain anyways, but how should we report the results:
How do other programs do that, if at all... Do you have any input @vandenman?
Jasp file:
https://github.com/jasp-stats/jasp-verification-project/blob/main/Datasets/Reliability.jasp
Note
Sería interesante que puedan incluir en la seccion de información de jasp , el contenido y los procedimientos de la confiabilidad bayesiana, así sería mas facil usarlo y poder recomendarlo.
Recently I calculated Cohen's kappa with SPSS and with JASP. Unfortunately, JASP sometimes produces strange results that do not correspond at all to the results of SPSS. E.g.: a test with N=59. Cohen's kappa in SPSS is 0.873, and kappa in JASP is -0.031.
For comparison, I also calculated the percentage agreement: 93.3%. Higher than Cohen's kappa from SPSS, but that is what you expect.
So it seems that de kappa from SPSS is right, but what happens with the kappa from JASP? There were 14 missings, can this be the reason for the result in JASP?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.