New all.mirt.objects slot in jrt objects to store all
mirt objects created during the automatic estimation process, to
facilitate further analysis without needing to re-fit the models.
Argument to change the x and y axis titles in the
info.plot() function.
Removed the irr dependency.
additional.stats = TRUE now reports selected
psych-based reliability indices and no longer reports
Fleiss kappa or ICC values.
Updated line geoms to use linewidth, avoiding the
ggplot2 3.4.0 deprecation warning for the size
aesthetic.
Vignette and warning/note improvements, including clearer warning
messages when judges have unobserved response categories.
jrt 1.1.2
Hannan–Quinn (HQ) Criterion is now in outputs, and can now be used
for automatic selection criterion.
Vignette improvements for clarity.
jrt 1.1.1
Fixes issue in jrt() with slope-constrained models when
ordinal scales start with 0 rather than 1.
Added a new example dataset with missing data.
Patched issues with displaying of amount of missing data.
jrt 1.1.0
Fixes for compatibility with recent versions of the
mirt package.
Fixes for compatibility with recent versions of the
ggplot2 package.
Update package citation.
Added documentation of S4 classes/methods.
Added the possibility to override the automatic facet naming in
jcc.plot() with argument facet.names.
Added the possibility to change the opacity of the different
category lines in jcc.plot() with
line.opacity.
Simplified messages for situations with judges/items with unobserved
categories.
Made Akaike Information Criterion (AIC) the default option in model
comparisons.
jrt 1.0.1
Simulated dataset is now directly included as
jrt::ratings.
Model weights (often called Akaike weights) are now presented in the
model selection output (per reviewer suggestion).
Calls to mirt can now be showed with
show.calls = TRUE argument in the
jrt()function.
References were updated.
jrt 1.0.0
Initial release with functions jrt(), jcc.plot() and
info.plot()