dummy_expand {lazy.tools} | R Documentation |
Dummmy Expantion of Categorical Data
dummy_expand(Uc, ncat = NULL, type = NULL, zero = 1)
Uc |
Item Response matrix in Compressed Format |
ncat |
a vector consisting of # of categories of each item |
type |
NOT used. |
zero |
= 0 to exclude the zero-th category from output |
In compressed format, Uc[i,j]=k means that subject i answers to item j
by selecting option k, k=0,1,...,ncat[j].
This function expands this compressed format to uncompressed
dummy expanded format:
That is, each item takes ncat[j] colums of U[, fromP[j]:toP[j]],
and the k+fromP[j]-th colum will be 1, and 0 elsewhere.
A list of
U, ncat, type, zero, fromP, toP
# generate item response
# in uncompressed format
# set.seed(1701)
# res <- gendataIRT( 1, paramS1, npoints=7, thmin=-3, thmax=3 )
# U=res$U
# in compressed format
# set.seed(1701)
# res1 <- gendataIRT( 1, paramS1, npoints=7, thmin=-3, thmax=3, compress=1 )
# Uc=res1$U
#
# uncompressed data: ncat=c(2,2,4)
U <- "
Q1_0 Q1_1 Q2_0 Q2_1 Q3_0 Q3_1 Q3_2 Q3_3
1 0 1 0 1 0 0 0
1 0 1 0 0 1 0 0
1 0 1 0 1 0 0 0
1 0 1 0 0 1 0 0
1 0 0 1 0 0 1 0
0 1 0 1 0 0 0 1
0 1 0 1 0 0 0 1
"; U=cards( U, header=1 )
#
# compressed data: ncat=c(2,2,4)
Uc <- "
Q1 Q2 Q3
0 0 0
0 0 1
0 0 0
0 0 1
0 1 2
1 1 3
1 1 3
"; Uc=cards( Uc, header=1 )
# from compressed to uncompressed
U1=dummy_expand(Uc)$U
# back to compressed
Uc1=dummy_compress(U)$Uc
Print(U-U1)
Print(Uc-Uc1)
# zero=0 option
U2=dummy_expand(Uc, zero=0)$U
Uc2=dummy_compress(U2, zero=0, ncat=c(2,2,4))$Uc
Print(Uc-Uc2)