9.6 Attribute Estimation in GDINA R package

To obtain the mastery probability \(E(\alpha_{ik})\), we can use personparm():

Code

head(personparm(est,what = "mp"))

##         A1     A2    A3
## [1,] 0.996 0.0928 0.979
## [2,] 0.983 0.9887 0.996
## [3,] 0.086 0.9683 0.989
## [4,] 0.947 0.7269 0.995
## [5,] 0.045 0.0008 0.992
## [6,] 0.519 0.3937 0.026

The EAP estimates can be obtained by comparing the mastery probabilities with 0.5 as in

Code

head(personparm(est,what = "mp")>0.5)*1

##      A1 A2 A3
## [1,]  1  0  1
## [2,]  1  1  1
## [3,]  0  1  1
## [4,]  1  1  1
## [5,]  0  0  1
## [6,]  1  0  0

or directly setting what = “EAP”,

Code

head(personparm(est,what = "EAP"))

##      A1 A2 A3
## [1,]  1  0  1
## [2,]  1  1  1
## [3,]  0  1  1
## [4,]  1  1  1
## [5,]  0  0  1
## [6,]  1  0  0

To obtain MLE and MAP estiamtes, set what = “MLE” or what = “MAP”:

Code

head(personparm(est,what = "MLE"))

##   A1 A2 A3 multimodes
## 1  1  0  1      FALSE
## 2  1  1  1      FALSE
## 3  0  1  1      FALSE
## 4  1  1  1      FALSE
## 5  0  0  1      FALSE
## 6  0  0  0      FALSE

Code

head(personparm(est,what = "MAP"))

##   A1 A2 A3 multimodes
## 1  1  0  1      FALSE
## 2  1  1  1      FALSE
## 3  0  1  1      FALSE
## 4  1  1  1      FALSE
## 5  0  0  1      FALSE
## 6  0  0  0      FALSE