NEWS

---

CDM 5.4

FIXED * fixed warnings in calculation of standard errors in

• 'gdina' and 'gdina.wald' (thanks to Andres Alberto Burga Leon)

DATA * included/modified datasets:

EXAMP * included/modified examples:

CDM 5.3

ADDED * included function CDM::IRT.RMSD which computed group-wise

• item fit. The RMSD item fit statistic was denoted as RMSEA in previous publications and is internally computed by the CDM::itemfit.rmsea function in the CDM package.

ADDED * included chi square statistic and MAD statistic (mean

• absolute deviation) of item fit in CDM::IRT.RMSD

NOTE * cleaned code in function 'WaldTest' a bit

• (thanks to suggestions of Michael Chirico)

NOTE * added links to recently released GDINA package in R.

• Included Example 1d for fitting a GDINA model with GDINA::GDINA.

DATA * included/modified datasets:

EXAMP * included/modified examples: IRT.RMSD (1), gdina (1d

CDM 5.2

NOTE * extended 'item_by_group' function which can now handle

• partial set of group-specific items

ADDED * included option for a subset of invariant items in

• multiple group estimation of GDINA model in 'gdina' (requested by Hamid Bahrami)

FIXED * fixed a bug in setting boundaries for guessing and slipping

• parameters in 'din' (thanks to Jill-Jen Vie)

NOTE * reference to new Journal of Statistical Software George et al

• article in the CITATION file and the CDM manual

DATA * included/modified datasets:

EXAMP * included/modified examples: item_by_group (1),

• gdina (5)

CDM 5.1

FIXED * changed the method of standard error calculation in

• 'gdina' function. It seems that previous CDM version produced overestimated standard errors (thanks to Xue-Lan Qiu). The improved calculation is now the default specified as the argument 'se_version=1'. The behavior of older CDM versions can be obtained with 'se_version=0'.

DATA * included/modified datasets:

EXAMP * included/modified examples:

CDM 5.0

NOTE * changed the way of calculating numerical derivatives in

• 'numerical_Hessian'

NOTE * included utility function 'abs_approx' for a differentiable

• function approximation (and its derivative 'abs_approx_D1') of absolute value function

NOTE * added argument 'append' in function 'osink

NOTE * included a utility function 'summary_sink' which sinks

• summary output of a method into a file

NOTE * included utility function 'deltaMethod' for nonlinear

• transformations of parameters

NOTE * included utility function 'item_by_group' which extends

• a dataset of item responses and a grouping variable to an extended dataset of group-specific item responses

ADDED * added sequential GDINA model for polytomous item responses

• (Ma & de la Torre, 2016, BJMSP) in 'gdina' function.

NOTE * included argument 'invariance' in 'gdina' which allows

• specification of non-invariant item parameters across multiple groups (see ?gdina, Example 5)

NOTE * changed number of M-steps in 'slca' in default to 10

• iterations (msteps=10)

DATA * included/modified datasets: data.cdm04

EXAMP * included/modified examples: abs_approx (1),

• summary_sink (1), deltaMethod (1), item_by_group (1), gdina (5,11)

CDM 4.991

NOTE * changed argument structure of 'numerical_hessian

NOTE * included maximum likelihood estimation method for

• GDINA models (method='ML') which seems to be more stable when method='ULS' (or 'WLS') has a bad convergence behavior. Now, the GDINA function switches to ML estimation if the RRUM model is used (thanks to a correspondence with Thomas Eckes).

NOTE * added an example of fitting (data.cdm, Example 1) the

• reduced RUM model in the CDM package (see the recent Chiu et al., IJT paper)

NOTE * changed default value of 'max.increment' argument in

• 'gdina' function to .3

FIXED * fixed a bug in 'gdina.dif' (thanks to Sinem Senfarah

DATA * included/modified datasets: data.cdm03

EXAMP * included/modified examples: data.cdm (1

CDM 4.8

FIXED * fixed a bug in 'itemfit.sx2' for fitted objects in TAM

NOTE * slightly changed output of 'print' methods

NOTE * changed argument 'rdigits' to 'digits' in 'summary

• method if necessary

NOTE * included utility functions 'osink' and 'csink' for

• writing output of 'summary' into a file

NOTE * allowed argument 'file' in 'summary' method for objects

• of class 'gdina', 'gdm', 'mcdina' and 'slca'

ADDED * included function 'numerical_Hessian' for a numerical

• evaluation of the Hessian matrix

DATA * included/modified datasets:

EXAMP * included/modified examples: osink (1

• numerical_Hessian (1)

CDM 4.7

NOTE * extended Wald test for item fit 'gdina.wald' for DINO

• models. In addition, an effect size of approximation ('uwgtdist' and 'wgtdist') is included

NOTE * cleaned namespace file and defined imported functions

• in a more explicit way

ADDED * added a S3 method 'IRT.parameterTable

ADDED * added a utility function 'WaldTest' for performing

• Wald tests

DATA * included/modified datasets:

EXAMP * included/modified examples: gdina.wald (1), data.ecpe (1

CDM 4.6

NOTE * included optional lower and upper bounds for estimation

• of guessing and slipping parameters in 'din' (arguments 'guess.min', 'guess.max', 'slip.min', 'slip.max') (requested by Kevin Carl Santos)

DATA * included/modified datasets:

EXAMP * included/modified examples:

CDM 4.5

NOTE * extended 'din' and 'gdina' functions for items with more

• than five measured attributes per item (requested by Alvaro Artavia Medrano)

NOTE * included 'converged' value for all models indicating

• whether maximum number of iterations was reached (suggested by an anonymous reviewer)

DATA * included/modified datasets:

EXAMP * included/modified examples:

CDM 4.4

FIXED * fixed a bug in function 'sequential.items

DATA * included/modified datasets:

EXAMP * included/modified examples:

CDM 4.3

NOTE * included t and p values in 'IRT.jackknife.gdina

FIXED * fixed bugs in 'summary.gdina' and 'gdina' due to

• computational problems in standard error calculation

DATA * included/modified datasets:

EXAMP * included/modified examples:

CDM 4.2

FIXED * fixed the ignorance of the 'centered.latent' argument

• in the 'gdm' function

ADDED * included S3 methods 'IRT.expectedCounts' for extracting

• expected counts and 'IRT.data' for extracting the used item response dataset

ADDED * included general plotting function for item response

• functions 'IRT.irfprobPlot'. This function can be used with both lattice and graphics packages.

ADDED * included S3 method 'predict' (and function 'IRT.predict

• for computing expected values and residuals for fitted models in CDM

NOTE * added attributes group and weights to 'IRT.data' value

NOTE * added a summary method for 'IRT.compareModels

NOTE * changed output of posterior in 'gdina

ADDED * included function 'IRT.repDesign' for constructing

• replicate designs

ADDED * added arguments for fixing and initial parameters in

• 'gdina' function

NOTE * added parameter table and matrix of polychoric correlations

• among skills in 'gdina' output

ADDED * added S3 method 'IRT.jackknife' and an initial version

• for models of class 'gdina'

ADDED * added function 'IRT.derivedParameter' for replication methods

• for derived parameters based on objects of class 'IRT.jackknife'

DATA * included/modified datasets: data.timss11.G4.AUT

• data.timss11.G4.AUT.part

EXAMP * included/modified examples: IRT.expectedCounts (1,2

• IRT.irfprobPlot (1,2), predict (1,2), IRT.data (1), IRT.repDesign (1,2), gdina (1), IRT.jackknife (1,2)

CDM 4.1 (2014-12-17)

NOTE * included confidence intervals S3 method for din objects

FIXED * fixed a bug in printing BIC for 'gdina' objects

NOTE * included S3 method 'IRT.modelfit' for assessment of

• absolute model fit

NOTE * included S3 method 'IRT.se' for computation of standard

• errors

NOTE * included S3 method 'IRT.IC' for extracting information

• criteria and 'IRT.compareModels' for model comparisons

NOTE * included object names in 'anova' methods for CDM objects

NOTE * included S3 method 'IRT.anova

NOTE * removed MX2 statistic from 'modelfit.cor' function

DATA * included/modified datasets:

EXAMP * included/modified examples: vcov.din (1), IRT.modelfit (1

• IRT.IC (1), IRT.compareModels (1), anova (1,2)

CDM 4.0 (2014-11-22)

NOTE * included argument for iteration progress in 'gdm

• and 'slca'

NOTE * included value call in 'din', 'gdina', 'gdm

• 'mcdina' and 'slca'

NOTE * changed print method in 'din' and included print method

• for 'gdina', 'gdm' and 'slca'

NOTE * added anova and coef method for 'slca' function

ADDED * added asymptotic covariance matrix estimation S3 method

• 'vcov' for objects of class din

ADDED * added computation of entropy of whole test and items in

• function 'entropy.lca' (Asparouhov & Muthen, 2014)

NOTE * included a summary function for 'cdi.kli

NOTE * included a function 'skillspace.full' for creating a

• skill space for dichotomous skills

ADDED * included a plot S3 method for 'gdm' and 'slca'

NOTE * corrected a bug in person value of 'gdm': missing standard

• errors of EAPs of multiple dimensions are now included

ADDED * included person fit appropriateness statistic (Liu et al., 2009

ADDED * included S3 methods 'IRT.likelihood', 'IRT.personfit

• 'IRT.irfprob' and 'IRT.factor.scores' for objects of classes din, gdina, gdm, slca and mcdina

NOTE * included person parameter estimation (classification) for

• 'slca' and 'mcdina' functions

DATA * included/modified datasets:

EXAMP * included/modified examples: data.dcm (1), modelfit.cor (2

• vcov-CDM (1), entropy.lca (1,2,3), personfit.appropriateness (1,2), slca (1), IRT.likelihood (1), IRT.irfprob (1), IRT.factor.scores (1)

CDM 3.4

NOTE * speeded itemfit.sx2 a bit

NOTE * included output value 'pattern' in multiple group GDINA models

• (requested by Sung Eun Kim)

NOTE * included adjusted mean absolute deviation of Q3 statistics (MADaQ3

• in modelfit.cor.din

EXAMP * included examples for running CDMs in mirt package

• DINA model ('data.dcm', Model 1)
• additive GDINA model ('data.dcm', Model 6)

DATA * included/modified datasets:

EXAMP * included/modified examples: modelfit.cor (1), din.deterministic

• itemfit.sx2 (1), data.dcm (1,6)

CDM 3.3

ADDED * included structured latent class analysis for polytomous data ('slca

DATA * included/modified datasets:

EXAMP * included/modified examples: slca (1,2,3,4,5

CDM 3.2

NOTE * included an argument save.devmin in 'gdina' for more parsimonious

• savings of output values to avoid working memory overflow for large datasets

NOTE * included effect size of differential item functioning in

• gdina model (included in 'gdina.dif')

FIXED * fixed a bug in itemfit.sx2

DATA * included/modified datasets:

EXAMP * included/modified examples: gdina.dif (1

CDM 3.1

FIXED * fixed a bug when printing objects of class 'din

NOTE * allowed vector input of item response pattern in 'plot.din

NOTE * disabled argument log.file in 'summary.din

NOTE * included random initial parameter values in 'din' and 'gdina

NOTE * changed some settings in plot.din

DATA * included/modified datasets:

EXAMP * included/modified examples: din (1), gdina (1), plot.din

CDM 3.0

NOTE * changed output in summary.gdina a bit (fixed a bug for

• the output of skill probabilities)

NOTE * included the argument 'avoid.zeroprobs' in gdina function

• to avoid zero probabilities of skill classes and inversion of singular matrices

NOTE * fixed problems in itemfit.sx2 for empty cells (thanks to

• Balal Izanloo)

ADDED * included a Wald test for the GDINA model (gdina.wald

ADDED * included assessment of item-wise differential item functioning

• in the GDINA model (gdina.dif)

FIXED * removed minor bug in summary.gdina

FIXED * fixed a bug with singular design matrix in skill space reduction

ADDED * included calculation of polychoric correlation in summary.gdina

• and an additional function skill.polychor

NOTE * removed parallel plot form plot.din

FIXED * fixed a crucial bug for in the multidimensional GDMs with item slope

• parameter estimation

ADDED * included function 'mcdina' for estimating the multiple group multiple

• choice DINA (MCDINA) model with polytomous attributes

DATA * included/modified datasets: data.cdm01, data.cdm02

EXAMP * included/modified examples: data.ecpe, gdina.wald (1),

• gdina.dif (1), din (7), mcdina (1,2,3)

CDM 2.7

ADDED * included cognitive diagnostic indices based on Kullback-Leibler

• information ('cdi.kli')

NOTE * changed NAMESPACE and DESCRIPTION files due to a change in Rcpp

NOTE * included non-exported Rcpp files for calculating individual

• likelihood and posterior distribution for polytomous items and multiple groups

DATA * included/modified datasets:

EXAMP * included/modified examples: cdi.kli (1

CDM 2.6

FIXED * fixed a bug in polytomous GDINA models with only one skill

NOTE * included value 'zeroprob.skillclasses' in skillspace.hierarchy

NOTE * included a more efficient skill space reduction method in

• gdina: using the argument 'reduced.skillspace.method', the new implemented method is now the default

ADDED * included data transformation function 'sequential.items' for

• constructing dichtomous pseudo items from polytomous items for estimation sequential item response models

DATA * included/modified datasets: data.sda6, data.timss03.G8.su

• data.timss07.G4.lee

EXAMP * included/modified examples: gdm (10

CDM 2.5

FIXED * fixed a bug when inverting a singular matrix in standard error

• calculation in gdina function (thanks to Hsin-Hao Chen)

NOTE * included conditional item probabilities in gdina output

NOTE * included gdina plot function (thanks to a discussion with

• Hamdollah Ravand)

FIXED * fixed a bug in summary.dina if data without column names

• is provided (thanks to Hu Jinxiang)

FIXED * removed a bug in gdina function which occured sometimes

• due to division by zero (thanks to Hu Jinxiang)

FIXED * fixed a bug in cdm.est.class.accuracy for GDINA models

• (thanks to Hamdollah Ravand)

FIXED * fixed some instable numerical behavior for higher order

• GDINA models (thanks to Ed Merkle)

FIXED * fixed a bug in calculating number of estimated parameters for

• 'din' function in case of skill classes with zero probability (thanks to Juergen Gross)

NOTE * added some explanations in the manual of 'modelfit.cor

ADDED * included S-X2 item fit statistic ('itemfit.sx2') accompanied

• with a summary and a plot method

ADDED * included marginal skill classification reliability in

• 'cdm.est.class.accuracy' function

ADDED * implemented significance tests in modelfit.cor and modelfit.cor.din

ADDED * included jackknife estimation of standard

• errors and bias corrected estimators in modelfit.cor.din

NOTE * speeded function 'modelfit.cor' by moving some parts of the code

• to Rcpp

NOTE * included warnings in case of nonconvergence in gdina function

• (thanks to a dataset provided by Hamdollah Ravand)

NOTE * slight change of algorithm for skill space reduction which

• seems to improve convergence behavior

ADDED * included a function for skill space approximation

ADDED * included a function for calculation reduced skill spaces

• if hierarchies among skills exist ('skillspace.hierarchy')

ADDED * included generalized distance discriminating method and a function

• for computing ideal response patterns

DATA * included/modified datasets: data.hr, data.melab, data.jang,

• data.dtmr

EXAMP * included/modified examples: gdm (4,5,6), gdina (8,9), sim.din (3),

• sim.gdina (2,3), itemfit.sx2 (1,2), din(6)

CDM 2.4

DATA * included two datasets 'data.dcm' and 'data.ecpe

DATA * eliminated a recoding error in data.Students

CDM 2.3

NOTE * worked on slight improvements of computation time in din

• and gdina

DATA * included multiple group dataset 'data.mg

NOTE * cleaned CDM manual (thanks to Michaela Mayr

CDM 2.2

NOTE * accelerated rowMaxs function

ADDED * included Q-matrix validation function (de la Torre, 2008

ADDED * included joint maximum likelihood estimation and

• classification function for the mixed DINA/DINO model -> function din.deterministic

EXAMP * included another example (based on attribute vectors)

• in sim.din function

ADDED * included polytomous GDINA model (Chen & de la Torre, 2013)

CDM 2.01

FIXED * fixed a bug in gdm in multiple group estimation: group

• labeling was incorrect and is now included in output group.stat and displayed in summary.gdm

NOTE * improved convergence behavior for multiple group analysis

NOTE * accelerated calculation in gdm: skillspace reduction

CDM 1.5

NOTE * allowed for centering of intercepts and slopes in gdm

ADDED * included satured skill distribution estimation in gdm

FIXED * fixed a bug in the calculation of RMSEA itemfit, although

• is not clear if the correct formula is used

FIXED * fixed a bug in gdm with varying number of item categories

• (thanks to Artur Pokropek)

ADDED * included the multidimensional latent class IRT model

• (Bartolucci, 2007) in gdm (skillspace="est")

NOTE * speeded calculation of modelfit.cor

CDM 1.4

FIXED * eliminated a bug in GDINA function for link="log

ADDED * included the parametrization of the reduced RUM model in gdina

ADDED * worked on higher order GDINA model which has the argument

• HOGDINA in gdina. Now there is dependency on the 'fa' factor analysis function in the psych package.

ADDED * included log-likelihood extractor functions

ADDED * included group-wise calculation of RMSEA statistics

• see itemfit.rmsea and the gdm output (thanks to Artur Pokropek)

FIXED * (hopefully) fixed a bug in gdm function with differing number of

• item categories (thanks to Artur Pokropek)

NOTE * use frequency patterns in gdm to speed up calculations

• for large data sets (argument 'use.freqpatt' in gdm function)

NOTE * included more than one step in gdm function

• arguments msteps (set to 8 M steps as the default) and convM

CDM 1.3

ADDED * included the general diagnostic model (function gdm

ADDED * included a function for calculating equivalent skill classes

• for the DINA/DINO model

ADDED * modelfit and anova functions are also accessible for the GDM

FIXED * fixed a bug in the function for classification accuracy

ADDED * included the RMSEA item fit index

FIXED * fixed a bug in gdina estimation

ADDED * included coef extractor function

CDM 1.2-3

ADDED * included a model fit function modelfit.cor(.din) for the

• comparison of observed and expected correlations

FIXED * fixed a bug in the gdina function when applying the model

• with one skill

CDM 1.2-1

ADDED * allow for user specified reduced skill space designs

ADDED * included tetrachoric correlations in summary.gdina function

ADDED * handled zero probability classes in gdina function

ADDED * allowed for output and input of attribute patterns in sim.din

• function

NOTE * included relative deviance change as convergence criterion in din

• function

ADDED * included function for determining a statistically equivalent

• DINA model (von Davier, 2013)

NOTE * changed structure of din summary

ADDED * included an anova function

ADDED * included a function for estimating classification reliability

FIXED * fixed a bug in gdina function when creating design matrices

CDM 1.2.0

ADDED * implemented the latent skill space reduction according

• to Xu and von Davier (2008)

FIXED * fixed a bug in calculation of estimated parameters

FIXED * fixed a bug in IDI calculation

FIXED * corrected MLE based classification in case of latent

• classes with a priori spceified zero probability in the din function (yet to be done in gdina)

FIXED * fixed a bug in standard error calculation in gdina

NOTE * speed improvements in din and gdina function

• included a C function for calculating the posterior distribution

NOTE * check of admissible rule in gdina

ADDED * include rule "GDINA2" and "GDINA1" ("GDINA1" is

• equivalent to "ACDM")

ADDED * Simulation function for the GDINA model included

ADDED * multiple group gdina function for estimation of

• generalized DINA models (de la Torre, 2011)

ADDED * included new function summary.gdina

CDM 1.1

NOTE * highest argument deleted from print.din method

NOTE * new argument for summary.din n.skill.classes

NOTE * skill pattern probabilities deleted from print.din

NOTE * skill pattern summary replaced by n most frequent skill

• classes in summary.din

NOTE * (argument ask=TRUE in summary.din placed by overwrite=FALSE

NOTE * use a different item discrimination index (Lee et al., 2012

NOTE * remove GDINA estimation from this package version because

• it is in an experimental status -> gdina will be included in CDM >= 1.1-1

NOTE * new optional attribute for q-matrix specification list of skill

• labels

NOTE * changed ouput in din function

• -> pattern, like and posterior are sorted according to the original input data frame

FIXED * remove a bug for models with only attribute

ADDED * new argument for din method zeroprob.skillclasses

ADDED * new argument for din method equal.itempars

NOTE * new arguments for summary.din method log.file and ask

FIXED * minor bug fixes in print.din method

NOTE * replace internal method rowProds with faster rowProds2

NOTE * option for writing an output file of model results included in

• function summary.din

CDM 1.0

• [— FIRST CRAN RELEASE —]