PCMRS {PCMRS}

Model Response Styles in Partial Credit Models

Description

Performs PCMRS, a method to model response styles in Partial Credit Models

Usage

PCMRS(
  Y,
  Q = 10,
  scaled = TRUE,
  method = c("L-BFGS-B", "nlminb"),
  cores = 30,
  lambda = 0
)

Arguments

Y

Data frame containing the ordinal item response data (as ordered factors), one row per obeservation, one column per item.

Q

Number of nodes to be used (per dimension) in two-dimensional Gauss-Hermite-Quadrature.

scaled

Should the scaled version of the response style parameterization be used? Default is TRUE.

method

Specifies optimization algorithm used by optim, either L-BFGS-B or nlminb.

cores

Number of cores to be used in parallelized computation.

lambda

Tuning parameter for optional L2 penalty on coefficient vector (for stabilized estimation)

Value

delta

Matrix containing all item parameters for the PCMRS model, one row per item, one column per category.

Sigma

2*2 covariance matrix for both random effects, namely the ability parameters theta and the response style parameters gamma.

delta.PCM

Matrix containing all item parameters for the simple PCM model, one row per item, one column per category.

sigma.PCM

Estimate for variance of ability parameters theta in the simple PCM model.

Y

Data frame containing the ordinal item response data, one row per obeservation, one column per item.

scaled

Logical, TRUE if scaled version of the response style parameterization is used.

neg.loglik

Negative marginal log-likelihood

Author(s)

Gunther Schauberger
[e-mail]
https://www.sg.tum.de/epidemiologie/team/schauberger/

References

Tutz, Gerhard, Schauberger, Gunther and Berger, Moritz (2018): Response Styles in the Partial Credit Model, Applied Psychological Measurement, https://journals.sagepub.com/doi/10.1177/0146621617748322

See Also

person.posterior PCMRS-package

Examples

## Not run: 
################################################
## Small example to illustrate model and person estimation
################################################

data(tenseness)

set.seed(5)
samples <- sample(1:nrow(tenseness), 100)
tense_small <- tenseness[samples,1:4]

m_small <- PCMRS(tense_small, cores = 2)
m_small
plot(m_small)

persons <- person.posterior(m_small, cores = 2)
plot(jitter(persons, 100))

################################################
## Example from Tutz et al. 2017:
################################################

data(emotion)
m.emotion <- PCMRS(emotion)
m.emotion

plot(m.emotion)

## End(Not run)