irf {lazy.irt}

R Documentation

Calculation of Item Response Function

Description

This function calculates icrf, irf and trf.
Japanese help file: (irf_JPH)

Usage

irf(
  param,
  theta = NULL,
  weight = NULL,
  zero = 1,
  smallP = 1e-09,
  thmin = -4,
  thmax = 4,
  npoints = 21,
  DinP = 1,
  print = 1,
  debug = 0,
  plot = 0,
  colors = "black",
  plotchar = NULL,
  linetype = NULL
)

Arguments

param	Item Parameter Data Frame
theta	Discrete theta values
weight	Weight data frame
zero	= 0 to exclude the zero-th category from output
smallP	Minimum value of probability
thmin	Minimum value of discrete thata valuepackage
thmax	Maximum value of discrete thata value
npoints	# of discrete points for theta
DinP	= 1 to include D=1.7 in logistic function
print	= 1 to print result
debug	= 1 to print intemediate result
plot	= 1 to plot result
colors	= color of the lines in plot
plotchar	= pch vector to be used in the plot
linetype	= lty vector to be used in the plot

Details

It is assumed that the data frame has the following columns in the following order.

name, type, ncat, p1, p2, ..., p[ncat+1]


When item_type == "B" or "B2" or "B3" or "Bn", ( ncat == 2 )
p1 is the discrimination parameter
p2 is the difficulty parameter
p3 is the gussing parameter


When item_type == "G" or "Gn",
p1 is the discrimination parameter
p2 is the category threshold parameter b_j1
p3 is the category threshold parameter b_j2

p_ncat is the category threshold parameter b_j,[ncat-1]


When item_type == "P",
p1 is the discrimination parameter
p2 is the step parameter b_j1
p3 is the step parameter b_j2

p_ncat is the step parameter b_j,[ncat-1]


When item_type == "PN,
p1 is the slope parameter
p2 is the intercept parameter b_j1
p3 is the intercept parameter b_j2

p_ncat is the intercept parameter b_j,[ncat-1]


When item_type == "N,
p1 - p_(ncat[j]-1) are the slope parameters
p_ncat[j] - p_2*(ncat[j]-1) are the intercept parameters

Regardless of the item_type, there will be ncat[j]
item paramters for item j, except for the Binary Items
which have the gusseing param as the ncat[j]+1st,
and the Nominal Items which have 2*(ncat[j]-1) item parameters.

Value

list( ICRF, IRF, TRF, fromP, toP )

where
ICRF npoints x sum(ncat)
IRF npoints x nitems weighted by item category weight
TRF npoints x 1 weighted by item category weight
and item weight
fromP, toP location of each item category in ICRF

Needs:
icrfB, icrfBN, icrfG, idrfGn, icrfPN, icrfP, icrfN, checkparam

Examples

Cards="
name type ncat   p1   p2   p3   p4
  Q1    B    2  0.9  0.0,,
  Q2    B3   2  1.0  0.0  0.2,
  Q3    G    4  1.0 -2.0  0.0  2.0
  Q4    P    4  0.8 -2.0  0.0  2.0
"; param <- cards( Cards, header=1 )
irf( param, plot=1 )

---