Learn R Programming

dcmr (version 1.0)

Attribute profile estimation using Diagnostic Classification Models and MCMC

Description

Analysis of dichotomous response data to obtain attribute profile estimates for respondents using Diagnostic Classification Model (DCM) and Markov Chain Monte Carlo (MCMC) method. The estimation procedure uses a loglinear cognitive diagnostic modeling (LDCM) framework that allows for the estimation of a host of DCMs such as NIDO, NIDA, NC-RUM etc.

Copy Link

Version

Install

install.packages('dcmr')

Monthly Downloads

3

Version

1.0

License

GPL (>= 3)

Maintainer

Diane Losardo

Last Published

July 26th, 2014

Functions in dcmr (1.0)

CalculatePiIStarPrime

Calculate value for pi.i.star.prime for one item
GetClassProbsFromMus

Obtain attribute profile probabilities from latent variable means
GetLambdaName

Obtain name of a particular lambda parameter
GetClassProbabilitiesMCMC

Calculate attribute profile probabilities
GetRequiredAttributesLambdaFullDCM

Calculate required attributes for lamdbas fully specified DCM
GetProbCorrectNcrum

Calculate item probability of correct response for NCRUM parameterization
GetRequiredAttributesLambdaDina

Calculate required attributes for lambdas for a DINA model.
attribute.profile.class-class

attribute.profile.class
GetAttributesProbabilitiesMCMC

Calculate attribute probabilities
GetThresholdValuesKernelPiR

Calculate item threshold values for NCRUM parameterization
GetRequiredAttributesLambdaCrum

Calculate required attributes for lambdas for a CRUM model.
iterate

Perform one iteration of MCMC procedure
GetThresholdValues

GetThresholdValues
all.results.class-class

all.results.class
DrawClasses

Draw latent variable values
GetGammaName

Obtain name of a particular gamma parameter
DrawAlphas

Draw values from multinomial distribution
GetAttributeProfiles

Attribute Profiles
GetParameterNames

Parameter Names for All DCM Models
ScoreDCM

Score dichotomous response data using DCM and MCMC
head,attribute.class-method

headattribute.class
class.probabilities.interaction.test

Class probabilities for Q-matrix containing interaction between attributes
InitializeParameters

Initialize parameter estimates
parameter.means.DCM.Mplus.test

Parameter estimates calibrated using Mplus for fully saturated model.
dcm.scorer.class-class

dcm.scorer.class
GetThresholdValuesKernel

Calculate item threshold values for kernel parameterization
GetAllProbsCorrectNcrum

Calculate all item probabilities of correct response for NCRUM parameterization
parameter.means.DCM.kernel.Mplus.interaction.test

Kernel parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with interaction.
SampleParameterEstimates

Randomly sample parameter estimates
parameter.acov.DCM.Mplus.interaction.test

Covariance matrix of parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with interaction.
GetKernelParameterNames

Kernel Parameter Names for all DCM Models Given a Q-matrix and model type it generates item and structural parameter names. These are kernel parameters: item thresholds (lambdas) and latent variable thresholds (gammas)
attribute.class-class

attribute.class
GetLambdaNamesForItem

Gets lambda names for a given item
GetRequiredAttributesLambdaDino

Calculate required attributes for lambdas for a DINO model.
parameter.acov.DCM.Mplus.test

Covariance matrix of parameter estimates calibrated using Mplus for fully saturated model.
head,attribute.profile.class-method

head attribute.profile.class
mcmc

Performs MCMC routine for DCM
LongFormatResults

Tranform dataframe to long format
parameter.class-class

parameter.class
parameter.means.DCM.Mplus.interaction.test

Parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with interaction.
class.probabilities.test

Class probabilities for Q-matrix containing no interaction between attributes
GetMusFromGammas

Calcualte attribute means from gamma parameters
observations.test

Obervations
GetGammaNames

Obtain names of all gamma parameters
parameter.means.names.NCRUM.interaction.test

Names of parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with interaction.
summary,attribute.class-method

summary attribute.class
print,attribute.class-method

print attribute.class
GetRequiredAttributesLambdaNido

Calculate required attributes for lambdas for a NIDO model.
parameter.means.names.DCM.Mplus.test

Names of parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with no interaction.
GetLambdaNames

Gets names of lambda parameters
summary,attribute.profile.class-method

summary attribute.profile.class
GetRequiredAttributes

Generate required attributes
parameter.means.NCRUM.interaction.test

Parameter estimates calibrated using Mplus for NC-RUM model and Q-matrix with interaction.
qmatrix.test

Q-matrix
plot,dcm.scorer.class,missing-method

plot of dcm.scorer.class
print,attribute.profile.class-method

print attribute.profile.class
plot,attribute.profile.class,missing-method

plot attribute.profile.class
GetThresholdLabels

Obtain item threshold labels
GetParameterResultsMCMC

Calculate results for attributes and attribute profiles
summary,dcm.scorer.class-method

summary of dcm.scorer.class
qmatrix.test.interaction

Q-matrix with interaction
parameter.means.names.DCM.Mplus.interaction.test

Names of parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with interaction.
plot,attribute.class,missing-method

plot attribute.class
parameter.means.DCM.kernel.Mplus.test

Kernel parameter estimates calibrated using Mplus for fully saturated model.
parameter.means.names.DCM.kernel.Mplus.interaction.test

Names of kernel parameter estimates calibrated using Mplus for fully saturated model and Q-matrix with interaction.