The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
test.linear() for performing tests for
linearity in a linear model, i.e., RESET test, Harvey-Collier test, and
Rainbow test.test.homo() for performing tests for
homoscedasticity in a linear model, i.e., Breusch-Pagan test, White’s
test, Evans-King Generalized Least Squares test, and Verbyla’s
test.test.norm() for performing tests for
univariate normality (i.e., Lilliefors test, Shapiro-Wilk test,
Anderson–Darling, and Cramer-von Mises test), and multivariate normality
(i.e., Henze-Zirkler test, Henze-Wagner test, and Royston test).test.bm() for performing the two-sample
and paired (Permuted) Brunner-Munzel test for stochastic equality.check.outlier(), i.e., the
function performs outlier and influential observations diagnostics by
plotting statistical measures for leverage, distance, and influence for
liner models.test.t.formula and
test.welch can conduct multiple two-sample t-tests or Welch
tests with multiple outcome variables.plot, geom,
filename, width, height, and
dpi to the functions cor.matrix() and
multilevel.cor() to draw and save a correlation heat
map.horiz to the function
print.misty.object() after removing the argument in the
previous versionsort.var from the function
print.misty.object().cor and ref from the
function test.t() and test.welch().geom to the functions
aov.b(), aov.w(), test.levene(),
test.t(), test.welch(), and
test.z() and removed the arguments bar,
point, box, ci,
line, and jitter.data to the functions
cohens.d.default, ci.mean.diff.default,
ci.prop.diff.default, test.t.default, and
test.z.default.na.omit to the functions
test.t.formula and test.welch.sample in
the functions skewness and kurtosis to
FALSE.square, rotate,
color, tile.alpha, plot.margin,
legend.box.margin, legend.key.size,
legend.text.size, and units from the function
na.pattern, and added the argument geom to the
function na.pattern.multilevel.r2, and added the argument ... to
the function multilevel.r2.check.resid, and added the argument ... to the
function multilevel.r2.plot.misty.object and changed the default setting for all
arguments to NULL.multilevel.alpha() for computing point
estimate and Monte Carlo confidence interval for multilevel coefficient
alpha defined by Lai (2021).item.stats() for conducting item analysis
for dichotomous, polytomous, and continuous items.item.distract() for conducting distractor
analysis for multiple-choice items.data.items, Multiple-Choice,
Dichotomous, and Polytomous Item Data, to the package.opdyke to the functions
item.cfa(), item.invar(),
multilevel.cfa(), multilevel.invar()to compute
the discrepancy between model-implied and observed correlation expressed
in terms of the relative percentile of an Opdyke distribution (McNeish,
2025).test and se to the
functions item.cfa() and
multilevel.cfa().se to the function
multilevel.omega() to save computation time by specifying
se = "none".df.subset() can subset rows based on
variables that were not selected, e.g.,
df.subset(iris, Sepal.Length, subset = Species == "setosa").multilevel.cfa() and multilevel.omega().r.digits to the functions
item.alpha(), item.omega(), and
multilevel.omega()exact, ml,
color, and style to the function
cor.matrix().cor.matrix computes p-values also for the
tetrachoric and polychoric correlation coefficient.summary to the argument
print in the function multilevel.cor()item.cfa(),
item.invar(), modcomp(),
multilevel.cfa(), and multilevel.invar()"WLSMVS" and "ULSMV" to
the argument estimator in the function
item.cfa() item.invar(), and
item.noninvar()."mean.var.adjusted.corrected" and
"scaled.shifted.corrected" to the argument
test in the function item.cfa().print in
the function multilevel.omega to omega.print in
the function multilevel.cor to cor.Yuan-Bentler to
Asymptotic Yuan-Bentler in the summary table of the
functions item.cfa() and multilevel.cfa() when
specifying test = yuan.bentler.mplus, i.e., default when
specifying estimator = MLR.R in the functions
boot.bs, ci.cor, ci.mean,
ci.median, ci.prop, ci.var,
ci.sd, mplus.lca, and summa to
nrep.item.alpha,
item.omega, item.reverse, and
item.score.type and added the argument
ordered in the function item.alpha.sig from the function
multilevel.cor() and multilevel.cor() and
added the arguments color, and style.color, and style to
the function item.cfa(), item.invar()
multilevel.cfa(), and multilevel.invar().print, freq,
hypo, descript, epsilon,
effsize, posthoc, order,
split, tri, table, and
horiz from the function
print.misty.object().write and tri from
the function write.result().item.alpha() and
item.omega() function.modcomp() for model comparison by
providing a table with fit indices for lavaan model objects, information
criteria, and likelihood ratio or F-test for various model objects,
e.g., from the lm(), lme(),
nlme(), lmer(), and glmer()
function. .boot.bs() for perfoming Bollen-Stine
bootstrapping with incomplete data.sim.lavaan() for generating simulated data
from a lavaan model syntax with unstandardized and standardized
parameters.item.dfi() for computing simulation-based
dynamic fit index cutoffs for evaluating confirmatory factor models
based on multivariate normal, multivariate non-normal, likert-type, and
categorical data.write.result() supports result objects
from the functions aov.b(), test.levene(),
test.welch().aov from the method
argument in the multilevel.icc() and
multilevel.descript() function.data.table from Suggests
to Imports.hypo and
descript in the functions aov.b,
test.levene, test.welch, test.t,
test.z to FALSE.weighted in
the functions aov.b and test.t to
TRUE.epsilon in
the function aov.w to FALSE.aov.b, aov.w, ci.cor,
ci.mean, ci.median, ci.var,
ci.sd, test.levene, test.t,
test.welch, amd test.z.aov.b,
aov.w, test.levene from Df,
Sum Sq, Mean Sq, F value, and
Pr(>F) to df, SS,
MSS, F, and p.item.noninvar() for computing the effect
size measure dMACS by Nye and Drasgow (2011) and signed dMACS by Nye et
al. (2019) for evaluating the magnitude and the direction of
between-group and longitudinal measurement non-invariance for continuous
and ordered categorical items.difftest.chibarsq() for perfoming the
chi-bar-square difference test to compare the random intercept
cross-lagged panel model (RI-CLPM) and traditional cross-lagged panel
model (CLPM) as discussed in Hamaker et al. (2015).randeff and varcor to
the argument print of the function
summa().coef.robust() computes cluster-robust
standard errors for multilevel and linear mixed-effects models estimated
by using the lme() function from the nlme package.summa() prints a summary of the result
object returned by the function lme() from the nlme
package.write to the function
write.result().horiz to the function
print.misty.object().item.invar() evaluates between-group and
longitudinal measurement invariance for measurement models with ordered
categorical indicators utilizing the Wu and Estabrook (2016) approach to
model identification.center provides a two-step latent mean
centering approach.mplus.lca.summa also provides the approximate weight of
evidence criterion (AWE), approximate correct model probability (cmP),
approximate Bayes factors (aBF), average posterior class probability
(AvePP), odds of correct classification ratio (OCC), and Cohen’s d to
quantify class separation.mplus.lca.summa also provides
class-specific item response probabilities of the indicator variables
and plots for LCA with ordered or unordered categorical indicator
variables.partial in the function
item.invar was modified to enable freeing parameters in
specific groups when evaluating between-group measurement invariance
based on more than two groups.boot and R for
requesting (bias-corrected) bootstrap confidence intervals to the
function mplus.lca.constr.var to impose inequality
constraints for the variance parameters at the between level to the
function multilevel.cor.lavaan.run to the function
item.invar.center(), which caused an
error message when centering a L2 predictor in three-level data within
L3 clusters.center() function.multilevel.icc() rounds ICCs smaller than
.Machine$double.eps^0.5 to 0.digits in
the function mplus.lca.summa to 0.missing in
the function item.alpha to listwise.mplus.lca, changed the default setting of the argument
sort.p to FALSE, and the default setting of
the argument output to
c("SVALUES", "CINTERVAL", "TECH11").width in the function
mplus.lca.summa()and added the arguments
width.ind, width.nclass, and
height.categ.within in the function
multilevel.cor(), i.e., the function automatically
identifies variables that are measured at the within level and modeled
only at the within level.sig in the function
cor.matrix and multilevel.cor, and added the
arguments color and style.result.lca() to
mplus.lca.summa().print in
the function indirect to mc.between in the function
multilevel.cor(), i.e., the function automatically
identifies variables that are measured at the between level.item.cfa() and multilevel.cfa()
adapted to the recent changes in the lavaan package (requested by Yves
Rosseel).robust.lmer() to estimate a multilevel and
linear mixed-effects model based on a robust estimation method using the
rlmer() function from the robustlmm
package.summa supports result objects from the
rlmer() function from the robustlmm
package.summa computes confidence intervals based
on heteroscedasticity-consistent standard errors when specifying
robust = TRUE.summa computes cluster-robust standard
errors for multilevel and linear mixed-effects model when specifying
robust = TRUE.coeff.robust computes cluster-robust
standard errors for (generalized) linear models that are robust to the
violation of the observation independence due to clustering.coeff.robust computes cluster-robust
standard errors for multilevel and linear mixed-effects model that are
robust to the violation of the homoscedasticity assumption.coeff.std and coeff.robust from
Std. Error, z value, t value ,
Pr(>|z|), and Pr(>|t|) to
Std. Error, z, t ,
p, and p.digits in
the function coeff.robust to 2.adjust from the function
mplus.lca.summa.write.data() and
write.data1() to write data files in CSV, DAT, TXT, SPSS,
Excel, or Stata DTA format.df.long() and df.wide() to
reshape data frames between ‘wide’ and ‘long’ format.summa() for printing a summary output of
the object returned by the functions lm() and
lmer() from the lme4 package.descript prints the number of unique
elements nUQ after omitting missing values.write, append and
output to the function uniq.categ and adjust to
the function na.auxiliary that allows specifying
categorical variables to compute Phi coefficient and Cramer’s V.optim.switch to the function
multilevel.cor.multilevel.cor computes the within-group
correlation (or between-group correlation) in the presence of only one
between-group (or within-group) variable.check.resid performs residual diagnostics
for linear mixed-effects models estimated by using the
lmer() function.multilevel.icc(),
computation of the ICC(2) was wrong (thanks to Ammar Ansari and
Tetsuhiro Yamada).pNA to %NA in the
output of the functions ci.cor, ci.mean,
ci.median, ci.mean.w, ci.prop,
ci.var, ci.sd, descript,
item.alpha, item.cfa, item.omega,
multilevel.cfa, multilevel.omega, and
na.descript.nObs and %Obs to
nOb and %Ob in the frequency table of the
na.descript() function.descript(), item.alpha(),
item.cfa, item.omega(),
multilevel.cfa, and multilevel.omega print the
column %NA with zero digits if all variables have no
missing values.SD.x and
SD.y to SDx and SDyin the output
of the std.coef() function.digits in
the function std.coef to 2.pval to p in the
output of the functions aov.b, aov.w,
na.test, test.t, test.welch, and
test.z.robust.coef(),
std.coef(), and rwg.lindell() to
coeff.robust(), coeff.std() and
cluster.rwg().df.subset().descript prints the percentage of
observations at the minimum (%Min) and at the maximum (%Max).multilevel.cor estimates the model without
standard errors to speed up model estimation when specifying
sig = FALSE (default setting).plot.misty.object() for
plotting results of the na.pattern() function (thanks Teun
van den Brand).uniq for extracting unique elements in a
vector, matrix, or data frame and function uniq.n for
counting the number of unique elements in a vector or for each column in
a matrix or data frame.std.coef computes standardized
coefficients for multilevel and linear mixed-effects models estimated by
using the lmer or lme function from the lme4
or nlme package.item.alpha computes coefficient alpha by
estimating an essentially tau-equivalent measurement model allowing full
information maximum likelihood (FIML) method for missing data handling
by specifying missing = "fiml" along with
estimator = "ML".rescov to the function
item.alpha for specifying residual covariances when
computing coefficient alpha.item.invar(), functions
did not allow specifying more than two residual covariances (thanks to
Lydia Laninga-Wijnen).print in
the functions item.alpha and item.omega to
alpha and omega.na.omit in the function
item.omega and added the arguments estimator
and missing.plot.misty.object() for plotting a misty
object.Added the argument factor.labels to the function
df.head and df.tail.
Added the arguments filename, width,
height, units, and dpi to the
functions aov.b, aov.w,
multilevel.r2, multilevel.r2.manual,
na.pattern, test.levene, test.t,
test.welch, and test.z.
The function df.rename can rename columns in a
matrix or variables in a data frame using
old_name = new_name and using the functions
toupper, tolower, gsub, and
sub similar to the rename function in the
dplyr package.
multilevel.icc(),
computation of the ICC(1) at Level 2 was wrong in case of a three-level
data when specifying type = "1b" (thanks to David S.
DeGarmo).center(), within-cluster
centering of a Level-2 predictor variable in three-level was wrong
(thanks to Stefanos Mastrotheodoros).df.head() and
df.tail(), function could not handle date and times.subset in the function
df.subset is specified without quotation marks in line with
the argument subset in the function subset
function.data in the functions
as.na, na.as, center,
ci.cor, ci.cor, ci.mean,
ci.median, ci.mean.w, ci.var,
ci.sd, cluster.scores, coding,
cor.matrix, crosstab, descript,
df.duplicated, df.unique,
df.move, df.subset, effsize,
freq, item.alpha, item.cfa,
item.invar, item.omega,
item.reverse, item.scores,
lagged, multilevel.cfa,
multilevel.cor, multilevel.descript,
multilevel.icc, multilevel.invar,
multilevel.omega, na.auxiliary,
na.coverage, na.descript,
na.indicator, na.pattern,
na.prop, na.test, rec,
rwg.lindell, skewness to the first
position.x in the functions
df.check, df.head, df.rename,
df.sort to data.x in the function
multilevel.fit to model.alpha, ci.plot,
plot.point, saveplot in the functions
ci.cor, ci.mean, ci.median,
ci.prop, ci.var, and ci.sd to
hist.alpha, confint, point and
filename.fill.col in the functions
na.pattern to color.file in the functions
blimp.plot, mplus.plot, and
na.pattern to filename.saveplot from the functions
na.pattern.na in the
functions na.indicator(), to 1.ci.cor() for computing and potting Fisher
z’ confidence interval for the Pearson product-moment correlation
coefficient adjusted via sample joint moments method or via approximate
distribution method (Bishara et al., 2018), Spearman’s rank-order
correlation coefficient with Fieller et al. (1957) standard error,
Bonett and Wright (2000) standard error or rank-based inverse normal
(RIN) transformation, and Kendall’s Tau-b, and Kendall-Stuart’s
Tau-c. The function also supports five types of bootstrap confidence
intervalschr.trunc() for truncating a character
vector, so that the number of characters of each element of the
character vector is always less than or equal to the specified
width.df.head() and df.tail() for
printing the first or last rows of a data frame and displaying only as
many columns as fit on the console.df.check() which is a wrapper function
around the functions dim(), names(),
df.head(), and df.tail().skewness and kurtosis can
also compute Mardia’s multivariate skewness and kurtosis.ci.mean, ci.median,
ci.prop, ci.var, and ci.sd can
also compute and plot bootstrap confidence intervals.sample to the function
descript.append to the function
check.outlier.sep and dec to the
function read.data.gray1, gray2, and
gray3 to the argument color of the function
chr.color.blimp.print(), function
printed error messages.nrow, ncol and
scales into facet.nrow,
facet.ncol and facet.scales in the functions
blimp.plot() and mplus.plot().error.width into
errorbar.width in the functions aov.b(),
aov.w(), result.lca(), test.t()
and test.z().line.size and
line.type into linewidth and
linetype in the functions test.t() and
test.z().line.color1,
line.color2, line.type1,
line.type2, line.width1,
line.width2, bar.color,
axis.size, strip.size, xlimits,
and ylimits into line.col1,
line.col2, linetype1, linetype2,
linewidth1, linewidth2, bar.col,
axis.text.size, strip.text.size,
xlim, and ylim in the function
check.resid().line in the
functions aov.w(), to FALSE.size.mean,
size.prop and size.cor into one help
page.missing in
the functions multilevel.cfa(), to
"fiml".estimator
in the functions multilevel.cor(), to
"MLR".na in the
functions na.indicator(), to 1.na.satcor(), cfa.satcor(),
sem.satcor(), growth.satcor(), and
lavaan.satcor() to estimate a confirmatory factor analysis
model, structural equation model, growth curve model, or latent variable
model in the lavaan package using full information maximum
likelihood (FIML) method to missing data handling while automatically
specifying a saturated correlates model to incorporate auxiliary
variables into a substantive model.read.data() to read data files in CSV,
DAT, TXT, SPSS, Excel, or Stata DTA format.print in
the functions na.test(), to little.mplus.plot(), which caused
an error message when requesting a loop plot by specifying
plot = "loop".mplus.print(), which caused
an error message when printing a Mplus output for an automatic testing
of measurement invariance.mplus and blimp do not
require the ...; specification in the
VARIABLES section anymore when specifying variable names
with the argument data.labels in the function
blimp.plot() to show parameter labels in the facet
labels.na.auxiliary() does not print full
NA rows of the Cohen’s d matrix anymore.na.indicator() creates a missing data
indicator matrix with 0 = observed and
1 = missing.na, append and
name to the function na.indicator().mplus.print(), function did
not the print input result when specifying
print = "all".blimp(), which caused an
error message when specifying a posterior = TRUE and saving
the posterior distribution failed.blimp.print(), which caused
an error message when specifying a misty.object for the
argument x.blimp.plot(), function did
not save and plots regardless of the setting of the argument
saveplot.mplus.plot() to read a Mplus GH5 file to
display trace plots, posterior distribution plots, autocorrelation
plots, posterior predictive check plots, and loop plots.blimp.run() to run a group of Blimp models
located within a single directory or nested within subdirectories.blimp.print() to print a Blimp output file
on the R console.blimp.plot() to read the posterior
distribution for all parameters to display trace plots and posterior
distribution plots.blimp() to create and run a Blimp input to
print the output onblimp.update() to update specific input
command sections of a misty.object of type
blimp to create an updated Blimp input file, run the
updated input file, and print the updated Blimp output.mplus.bayes() to read a Mplus GH5 file and
blimp.bayes() to read the posterior distribution for all
parameters to compute point estimates, measures of dispersion, measures
of shape, credible intervals, convergence and efficiency diagnostics,
probability of direction, and probability of being in the ROPE for the
posterior distribution for each parameter.na.test function provides Jamshidian and Jalalꞌs
approach for testing the missing completely at random (MCAR)
assumption.clear() to clear the console equivalent to
Ctrl + L in RStudio.chr.color() to add color and style to
output texts on terminals that support ‘ANSI’ color and highlight
codes.default to the argument
print of the descript function.comment to the mplus
function.na.test performs Little’s MCAR test using
the mlest function from the mvnmle package
that can handle up to 50 variables instead of using the
prelim.norm function in the norm package that
can only handle about 30 variables.na.pattern plots the missing data pattern
when specifying plot = TRUE and runs faster.na.auxiliary computes semi-partial
correlations of an outcome variable conditional on the predictor
variables of a substantive model with a set of candidate auxiliary
variables to identify correlates of an incomplete outcome variable as
suggested by Raykov and West (2016)..print in
the functions mplus.print, to result.mplus.print does not print the section
MODEL FIT INFORMATION if the degrees of freedom is
zero.run.mplus in the function
mplus.lca() to mplus.run.ls.fit in
the function multilevel.cfa() to FALSE.mplus.update(), which
caused an error message when specifying
output = FALSE.mplus.lca(), which caused
an error message when checking the input for the argument
processors.item.cfa() and
multilevel.cfa(), functions did not allow specifying more
than two residual covariances (thanks to Lydia Laninga-Wijnen).multilevel.descript(),
average, minimum, and maximum cluster size at Level 3 were calculated
incorrectly.item.omega(), function did
not provide item statistics regardless of the print
argument setting (thanks to Ainhoa Coloma Carmona).mplus.run() according to the
latest version of the function runModels() in the
MplusAutomation package.mplus.print(), function did
not print result of a misty.object of type mplus.mplus.print() for printing a Mplus output
file on the R console.mplus() to create and run a Mplus input to
print the output on the console.update.mplus() to update specific Mplus
input command sections in the mplus object, run the updated
input file, and print the output on the console.chr.grep() and chr.grepl()
for multiple pattern matching, i.e., grep() and
grepl() functions for matching a vector of character
strings.write.mplus() is not restricted to
variable names with up to 8 characters anymore.run.mplus() to
mplus.run().posthoc in
the functions aov.b(), aov.w() and
test.welch() to FALSE.replace from
modifiedDate to modified in the functions
mplus.lca() and mplus.run().showOutput into
show.out and replaceOutfile into
replace.out in the function mplus.run().message to the function
mplus.run().test.welch(), function did
not print post hoc tests when specifying
posthoc = TRUE.result.lca(), function
excluded all outputs which involved the word ERROR even
though results were available (thanks to Michael Weber).multilevel.fit(), function
used the number of observations at the Within level instead of the
Between level for computing RMSEA at the Between Level (thanks to
Maurizio Sicorello).descript() which caused an
error message when specifying a split variable.robust.coef() which caused
an error message in the presence of missing data on predictor
variables.multilevel.icc() and
multilevel.descript() which caused an error message in when
specifying a tibble instead of a data frame (thanks to Tanja Held).mplus.lca(), the argument
processors allows to specify the number of processors and
threads separately.item.omega(), residual covariances can
be specified when type = "categ".., +, -, ~,
:, ::, functions which caused an warning
message.center(),
multilevel.icc(), and multilevel.descript()
which caused an error message in three-level data with ambiguously coded
cluster variables common in longitudinal data.multilevel.descript() to take into
account missing values, e.g., No. of cases and
No. of clusters show the number observations and clusters
after excluding missing values.write.sav() do not
require specifying all three columns label,
values, and missing anymore.na to the function
read.mplus().freq(), function did not
provide an output.df.subset() for subsetting data frames
using the operators ., +, -,
~, :, ::, and !
similar to functions from the R package tidyselect.lagged() to compute lagged values of
variables.df.move() to move variable(s) in a data
frame.read.dta() and write.dta()
to read and write Stata DTA files.coding() to code categorical variables,
i.e., dummy, simple, unweighted and weighted effect, repeated, forward
Helmert, reverse Helmert, and orthogonal polynomial coding.effsize() to compute effect sizes for
categorical variables, i.e., (adjusted) phi coefficient,
(bias-corrected) Cramer’s V, (bias-corrected) Tschuprow’s T, (adjusted)
Pearson’s contingency coefficient, Cohen’s w, and Fei.script.copy() to save a copy of the
current script in RStudio with the current date and time.as.na(), na.as()``center(),
ci.mean(), ci.mean.w(),
ci.median(), ci.prop(), ci.var(),
ci.sd(), cluster.scores(),
cor.matrix(), crosstab(),
descript(), freq(), item.alpha(),
item.cfa(), item.invar(),
item.omega(), item.reverse(),
item.scores(), multilevel.cfa(),
multilevel.cor(), multilevel.descript(),
multilevel.fit(), multilevel.icc(),
multilevel.invar(), multilevel.omega(),
na.auxiliary(), na.coverage(),
na.descript(), na.indicator(),
na.pattern(), na.prop(),
na.test() rec(), rwg.lindell(),
skewness(), and kurtosis() provide the
argument ... instead of the argument x to
specify variables from the data frame specified in data
using the operators ., +, -,
~, :, ::.multilevel.icc() computes intraclass
correlation coefficients in three-level data.multilevel.descript() computes multilevel
descriptive statistics in three-level data.center() centers predictor variables in
three-level data.na.descript() provides descriptive statistics
for missing data in two-level and three-level data.cor.matrix() computes tetrachoric and
polychoric correlation coefficients.write and append to
all functions providing a print function to save the print output into a
text file.names in
the function rec() to .e.label and
labels in the read.sav function to
FALSE.value in the function
na.as() to na to make it consistent with the
arguments of the function as.na().resid.cov in the function
item.omega() to resocv to make it consistent
with the arguments of the functions item.cfa() and
multilevel.cfa().names in the functions
center, cluster.scores,
item.reverse, and rec to name to
make it consistent with the arguments of the functions
item.scores(), na.prop(), and
lwg.lindell().x and ... in the
functions df.duplicated() and df.unique()to
... and data to make it consistent with all
other functions using the ... argument.as.na and
na.as into one help page.script.open,
script.close, and script.save into one help
page.skewness and
kurtosis into one help page.ci.mean and
ci.median into one help page.ci.var and
ci.sd into one help page.shift() and replaced it by the
function lagged().dummy.c() and replaced it by the
function coding()cor.phi(),
cor.cont(), cor.cramer(), and
eta.sq() and replaced them by the function
effsize().cor.poly() and integrated
polychoric correlation coefficient into the function
cor.matrix().multilevel.descript(),
function led to a node stack overflow.shift() to compute lagged or leading
values of a vector.libraries(), version of the
packages were not correctly displayed.test.welch(), to remove
errors for r-devel from a recent change in r-devel.group.ind to the function
result.lca() to specify. latent class indicators as
grouping variable in the bar charts.mplus.lca() can be used to conduct latent
class analysis with count, unordered categorical, and ordered
categorical indicator variables.result.lca() can be used to save bar charts
with error bars for confidence intervals for each of the latent class
solutions.dominance.manual(),
function provided the wrong rank ordering.mplus.lpa() and
results.lpa() to mplus.lca() and
results.lca().item.invar() for evaluating configural,
metric, scalar, and strict between-group or longitudinal (partial)
measurement invariance.robust.coef() for computing
heteroscedasticity-consistent standard errors and significance values
for linear models estimated by using the lm() function and
generalized linear models estimated by using the glm()
function.dominance() for linear models estimated by
using the lm() function and dominance.manual()
to conduct dominance analysis based on a (model-implied) correlation
matrix of the manifest or latent variables.check.resid() for performing residual
diagnostics to detect nonlinearity (partial residual or
component-plus-residual plots), nonconstant error variance (predicted
values vs. residuals plot), and non-normality of residuals (Q-Q plot and
histogram with density plot).mplus.lpa() for writing Mplus input files
for conducting latent profile analysis based on six different
variance-covariance structures.result.lpa() for creating a summary result
table for latent profile analysis from multiple Mplus output files
within subfolders.order to the function
multilevel.cor() to order variables in the output table so
that variables specified in the argument between are shown
first.multilevel.cfa() and multilevel.invar().item.cfa(), multilevel.cfa(), and
multilevel.invar().write.result() can also write results based on
the return object of the std.coef function.min.value in the function
item.cfa(), multilevel.cfa(), and
multilevel.invar() to mod.minval and changed
the default setting to 6.63.r2mlm from the
Imports field in the DESCRIPTION due to
dependencies issues.multilevel.descript() can also deal with
between-cluster variables by reporting means and standard deviations at
the cluster level.print to the function
multilevel.descript() to request standard deviation of the
variance components.multilevel.fit() for computing
simultaneous and level-specific model fit information for a fitted
multilevel model containing no cross-level constraints from the R
package lavaan.multilevel.cfa() for conducting multilevel
confirmatory factor analysis using the R package lavaan to investigate
four types of constructs, i.e., within-cluster, shared, configural, and
simultaneous shared and configural cluster constructs.multilevel.invar() for evaluating
configural, metric, and scalar cross-level measurement invariance using
multilevel confirmatory factor analysis.multilevel.omega() for computing point
estimate and Monte Carlo confidence interval for the multilevel
composite reliability defined by Lai (2021) for a within-cluster
construct, shared cluster-level construct, and configural cluster
construct.multilevel.cor(), e.g., warning message is printed when
absolute correlations are greater than 1.cluster in the function
multilevel.cor(), multilevel.descript(), and
multilevel.icc() can also be specified using the variable
name of the cluster variable in x.item.cfa() function, e.g.,
loglikelihood and information criteria are shown above chi-square test
of model fit and label Ad Hoc changed to
Scaled.multilevel.cor(), which
caused an error message (thanks to Richard Janzen).libraries() to load and attach multiple
add-on packages at once.check.outlier() computes statistical
measures for leverage, distance, and influence for linear models
estimated by using the lm() functionwrite.result(), result tables are
in line with the arguments print, tri,
digits, p.digits, and icc.digits
specified in the object x (thanks to Stefan Kulakow).crosstab() displays marginal row-wise,
column-wise, and total percentages in the output (thanks to Joachim
Fritz Punter and Lisa Bucher). Note that the function now also returns
the crosstable in the list element result$crosstab of the
return object .Value sections in the documentation of the
functions.weighted in
the test.t and the na.auxiliary function to
FALSE in line with the recommendation by Delacre et
al. (2021).collin.diag() to
check.collin().read.mplus(), an error
message was printed if comments in the Mplus input file contains special
characters (e.g., ä, ü, ö).std.coef(), the function
was not applicable to predictors specified as character vector or
factor.script.close(),
script.new(), script.open(), and
script.save() to close, open, and save R scripts in
RStudio.setsource() to set the working directory
to the source file location in RStudio equivalent to using the menu item
Session - Set Working Directory - To Source File Location.restart() to restart the RStudio session
equivalent to using the menu item Session - Restart R.multilevel.r2.manual() to compute
R-squared measures by Rights and Sterba (2019) for multilevel and linear
mixed effects models by manually inputting parameter estimates.center(), cluster.scores(),
rec(), and item.reverse() can be applied to
more than one variable at once.aov.w() for performing repeated measures
analysis of variance (within-subject ANOVA) including paired-samples
t-tests for multiple comparison, descriptive statistics, effect size
measures, and a plot showing error bars for within-subject confidence
intervals.ci.mean.w() for computing
difference-adjusted Cousineau-Morey within-subject confidence
intervals.ci.mean.diff() computes the confidence
interval for the difference for an arithmetic mean in a one-sample
design.aov.b(), test.t(),
test.welch(), and test.z() plot
difference-adjusted confidence intervals in two-sample design by
default.jitter.height to the functions
aov.b(), test.levene(), test.t(),
aov.welch(), and test.z().adjust to the function
ci.mean(), to apply difference-adjustment for the
confidence interval.test.t() displays the confidence interval for
the mean difference in the one-sample t-test.test.t(), result table
provided by the function did not display the confidence interval
correctly.aov.b() for performing between-subject
analysis of variance including Tukey HSD post hoc test for multiple
comparison.as.na() is also applicable to arraysplot and arguments for various
graphical parameters for plotting results to the functions
test.levene(), test.t(),
test.welch(), and test.z().write for writing results into an
Excel file to the functions cor.matrix(),
crosstab(), descript(), freq(),
item.alpha(), item.cfa(),
item.omega(), multilevel.cor(),
multilevel.descript(), na.coverage(),
na.descript(), and na.pattern()posthoc for conducting Games-Howell
post hoc test for multiple comparison to the functions
test.welch().item.cfa() for conducting confirmatory
factor analysis using the R package lavaan.write.result() can also write results based on
the return object of the item.cfa() function.exclude of the function freq()
can also be set to FALSE.multilevel.cor() to
make it consistent with the output of the function
item.cfa().na.omit in the function
multilevel.cor() to missing to make it
consistent with the arguments of the function
item.cfa().estimator
in the function multilevel.cor() to ML, so
that full information maximum likelihood method is used for missing data
handling.multilevel.cor(), function
did not use Huber-White robust standard errors, but conventional
standard errors when specifying estimator = "MLR".multilevel.r2() for computing R-squared
measures for multilevel and linear mixed effects models.write.xlsx() for writing Excel files
(.xlsx).write.result() for writing results of a
misty object into an Excel file.multilevel.descript().round to the function
freq() for rounding numeric variables.na.test() function when
running into numerical problems.sig in the
functions cor.matrix() and multilevel.cor() to
FALSE.collin.diag() function.print.misty.object(),
function did not print the result object of the the function
crosstab() correctly when requesting percentages.multilevel.cor() for computing the
within-group and between-group correlation matrix using the lavaan
package.na.test() for performing Little’s missing
completely at random (MCAR) test.indirect() for computing confidence
intervals for the indirect effect using the asymptotic normal method,
the distribution of the product method, and the Monte Carlo method.multilevel.indirect() for computing
confidence intervals for the indirect effect in a 1-1-1 multilevel
mediation model using the Monte Carlo method.cor.matrix() highlights statistically
significant correlation coefficients in boldface.cor.matrix() shows the results in a table when
computing a correlation coefficient for two variables.stat) and degrees of freedom
(df) to the argument print in the function
cor.matrix().continuity for continuity correction
to the function cor.matrix() for testing Spearman’s
rank-order correlation coefficient and Kendall’s Tau-b correlation.cor.matrix() when computing Spearman’s rank-order
correlation coefficient or Kendall’s Tau-b correlation.group in the functions
center(), group.scores(),
multilevel.descript(), multilevel.icc(), and
rwg.lindell() to cluster.group.scores() to
cluster.scores().cor.matrix(), function did
not print sample sizes when specifying a grouping variable and using
listwise deletion.write.mplus() writes a Mplus input template
with variables names specified in the DATA command along with the
tab-delimited data file by default.print() in the
write.mplus() function.weighted in
the test.welch() function into FALSE following
the recommendation by Delacre et al. (2021).cohens.d(), function
printed warning messages of the pt() function.cohens.d(), function could
not deal with more than one variable in a one-sample design.test.t() for performing one-sample,
two-sample, and paired-sample t-tests including Cohen’s d effect size
measure.test.welch() for performing Welch’s t-test
including Cohen’s d effect size measure and Welch’s ANOVA including
\(\eta^2\) and \(\omega^2\) effect size measures.print
in the function descript().format, label,
labels, missing to the function
read.sav() to remove variable formats, variable labels,
value labels, value labels for user-defined missings, and widths from
attributes of the variable.item.reverse() can also be applied to to items
with non-integer values.cor.matrix() when
specifying a grouping variable comprises the combined results of both
groups in the matrices.read.mplus() can also deal with consecutive
variables (e.g., x1-x5).group and split arguments to the
function cohens.d().test.z function.cohens.d() computes various kinds of Cohen’s
d, Hedges’ d, and Glass’s \(\Delta\)
including confidence intervals, e.g., weighted and unweighted pooled
standard deviation in a two-sample design, with and without controlling
for the correlation between the two sets of measurement in a
paired-sample design, or with and without the small-sample correction
factor.alpha.coef() to
item.alpha(), cont.coef() to
cor.cont(), cramers.v() to
cor.cramer(), levenes.test() to
test.levene(), mgsub() to
chr.gsub(), omega.coef() to
item.omega(), reverse.item() to
item.reverse(), phi.coef() to
cor.phi(), poly.cor() to
cor.poly(), scores() to
item.scores(), stromit() to
chr.omit(), trim() to chr.trim(),
z.test() to test.z(),use in the
cor.matrix() function into na.omit.method in
the functions multilevel.descript() and
multilevel.icc() to "lme4"; if the lme4
package is not installed, "aov" will be used.ci.mean.diff() and
ci.mean.prop() when computing confidence intervals in
two-sample designs, i.e., results are divided in two rows according to
the grouping variable.ci.mean.diff() and
ci.mean.prop() when computing confidence intervals in
paired-sample designs, i.e., output reports the number of missing data
pairs (nNA), instead of number of missing values for each
variable separately (nNA1 and nNA2).descript() when
specifying the argument levenes.test(), i.e., duplicated
labels in the column group or variable are not
shown.cohens.d() into a generic
function with the methods cohens.d.default() and
cohens.d.formula().hypo and descript to the
functions test.levene() and test.z().freq,
descript, and crosstab function.as.na in the as.na()
function into na.center() which caused an
error message in case of groups with only one observation when trying to
apply group mean centering.center() which caused an
error message when trying to apply grand mean centering of a Level 1
predictor.cohens.d(), an error
message was printed in the between subject design whenever specifying a
grouping variable with missing values.cor.matrix(), which caused
an error when using listwise deletion for missing data while specifying
a grouping variable.descript(), which caused an
error message when selection only one or two argument statistical
measures using the argument print.freq(), where the argument
split was broken.test.zz(), where the
alternative hypothesis was displayed wrong when specifying
alternative = "greater" or
alternative = "less".collin.diag() for collinearity diagnostics
including tolerance, (generalized) standard error inflation factor,
(generalized) variance inflation factor, eigenvalues, conditional
indices, and variance proportions for linear, generalized linear, and
mixed-effects models.std.coef() for computing standardized
coefficients (StdX, StdY, and StdYX) for linear models estimated by
using the lm() function.mgsub() for multiple pattern matching and
replacements, i.e., gsub() function for matching and
replacing a vector of character strings.df.duplicated() and
df.unique() extracting duplicated or unique rows of a
matrix or data frame.read.xlsx(), default
setting of the argument progress was wrong.print.misty.object().z.test() for performing one sample, two
sample, and paired sample z-test.omega.coef() does not access internal slots of
a fitted lavaan object anymore (requested by Yves Rosseel).levenes.test().size.mean(),
size.prop(), and size.cor() to include Greek
letters.theta in the
size.mean() function into delta.ci.mean(), ci.mean.diff(),
ci.median(), ci.prop(),
ci.prop.diff(), ci.sd(), ci.var()
for computing confidence interval for the arithmetic mean, the
difference in arithmetic means, the median, the proportion, the
difference in proportions, the variance, and the standard
deviation.levenes.test() for conducting Levene’s
test for homogeneity of variance.omega.coef() for computing coefficient
omega (McDonald, 1978), hierarchical omega (Kelley &
Pornprasertmanit, 2016), and categorical omega (Green & Yang,
2009).read.xlsx() for reading Excel files
(.xlsx).coef.alpha().cor.matrix().as.na() can also replace user-specified values
with missing values in lists.use in the
alpha.coef() function into a logical argument
na.omit.pval.digits in the
cor.matrix() function into p.digits.print.cont.coef(),
print.cramers.v(), print.na.auxiliary(),
print.na.coverage(), print.phi.coef(), and
print.poly.cor() into
print.square.matrix()is.vector()
function was used to test if an object is a vector. Instead
is.atomic() function is used to test if an object is a
vector.as.na(), function converted
strings in data frames to factors.trim() for removing whitespace from start
and/or end of a string. Note that this function is equivalent to the
function trimws() in the base package.
However, the trimws() function fails to remove whitespace
in some instances.cohens.d(), function
returned NA for Cohen’s d in within-subject design in the
presence of missing valuesalpha.coef(), function did
not provide any item statistics irrespective of the argument
printas.na(), function always
generated a warning message irrespective of the argument
as.na.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.