check.plot.mcmc.3pnob: Plot of the discrimination marginal posterior means against...
In bairt: Bayesian Analysis of Item Response Theory Models

Description Usage Arguments Details Value Author(s) References See Also Examples

Marginal Posterior means of b_j plotted against the marginal posterior means of a_j. Each point is labeled with the number of the corresponding Item.

For the Three-Parameter Normal Ogive Item Response Model (3pno), the size of the numbers refers to the marginal posterior means of c_j.

The Potential Scale Reduction Factor (Rhat) is calculated for each chain, bairt generates a single MCMC and evaluates convergence by breaking the chain in three sub chains and comparing the between- and within-subchain variance.

The black color suggests convergence and red items indicate convergence problems (Rhat greater than 1.1).

1
2
3

## S3 method for class 'mcmc.3pnob'
check.plot(mcmclist, converg.test = T, c.probs = c(0,
  0.2, 0.5, 1), legen = "topleft", ...)

`mcmclist`	A mcmc.2pnob or mcmc.3pnob class object.
`converg.test`	Checking if Rhat is major that 1.1.
`c.probs`	Vector for assignment of intervals the Guessing (c).
`legen`	Coordinates to be used to position the Guessing (c) legend.
`...`	Further arguments.

If converg.test = TRUE the items with Rhat menor that 1.1 are print in red color. It is useful for quick check of the convergence.

A plot of the discrimination marginal posterior means against difficulty marginal posterior means. For the Three-parameter model the guessing marginal posterior means are represented by the number size of the item.

Javier Mart<c3><ad>nez

Johnson, V. E. & Albert, J. H. (1999). Ordinal Data Modeling. New York: Springer.

Gelman, A., Carlin, J. B., Stern, H. S. & Rubin, B. (2004). Bayesian Data Analysis.New York: Chapman & Hall/CRC.

mcmc.2pnob, mcmc.3pnob and continue.mcmc.bairt.

# data for model
data("MathTest")

# Only for the first 500 examinees of the data MathTest
# Two-Parameter Normal Ogive Model
model2 <- mcmc.2pnob(MathTest[1:500,], iter = 400, burning = 100)
check.plot(model2)
chain.study(model2, parameter = "b", chain = 12)
chain.study(model2, parameter = "theta", chain = 10)


# For all examinees of the data
# Two-Parameter Normal Ogive Model
modelAll2 <- mcmc.2pnob(MathTest, iter = 3500, burning = 500, thin = 10)
check.plot(modelAll2)
chain.study(modelAll2, parameter = "b", chain = 14)
chain.study(modelAll2, parameter = "theta", chain = 10)

# Three-Parameter Normal Ogive Model
modelAll3 <- mcmc.3pnob(MathTest, iter = 3500, burning = 500, thin = 10)
check.plot(modelAll3)
chain.study(modelAll3, parameter = "b", chain = 12)
chain.study(modelAll3, parameter = "c", chain = 10)


## End(Not run)