dirf_p {lazy.irt}

R Documentation

Calculation of the Derivatives of the Item Category Response Function with respect to Item Parameters

Description

Calculation of the Derivatives of the Item Category Response Function with respect to Item Parameters

Usage

dirf_p(
  paramj,
  theta = NULL,
  weight = NULL,
  smallP = 0,
  DinP = 1,
  thmin = -4,
  thmax = 4,
  npoints = 21,
  Pj = NULL,
  PPj = NULL,
  zero = 0,
  cat.first = 0,
  log = 0,
  print = 0
)

Arguments

paramj	item parameters data frame for ONE item
theta	Discrete theta values
weight	Weight data frame: NOT used.
smallP	Minimum value of probability
DinP	= 1 to include D=1.7 in logistic function
thmin	Minimum value of discrete thata value
thmax	Maximum value of discrete thata value
npoints	# of discrete points for theta
Pj	icrf: npoints x (ncatj-1) (no zero category) or NULL
PPj	icbrf of the Graded Response Model or NULL
zero	= 1 to include the zero-th category in output
cat.first	= 1 to chage the category fist in the rows of Jack.
log	= 1 to obtain the log Jacobian: d log(ICRF) d param
print	= 1 to print result

Value

list of (Jack, Pj, PPj)
Jack (length(theta) x (ncatj-1)) x ncatj
derivative of vec(Pj) with respect to (a, b1, b2, ...)
If cat.first = 0
theta changes first, then k changes from 1 to ncatj
If cat.first = 1, category(k) changes first, then theta.
Pj Item Category Response Function (icrf) (length(theta) x (ncatj-1))
PPj will be output when type="G" or "Gn".

Examples

res=dirf_p( paramS1[3,], npoints=5, print=1 )
res=dirf_p( paramS1[3,], npoints=5, print=1, zero=1 )
res=dirf_p( paramS1[3,], npoints=5, print=1, cat.first=1 )

---