R 3모수 모형

1 개요[ | ]

R IRT 3모수 모형
R 문항반응이론 3PLM
R 문항반응이론 3모수 모형

2 ltm 패키지[ | ]

R

CPU

4.1s

MEM

241M

4.9s

Reload

Copy

df <- read.csv("https://raw.githubusercontent.com/jmnote/ds/main/github.com/cran/ltm/data/LSAT.csv")
library(ltm)
model <- tpm(df)

# 로그우도
model$log.Lik

Loading required package: MASS

Loading required package: msm

Loading required package: polycor

-2466.65950899778

Copy

# 계수
coef(model)

A matrix: 5 × 3 of type dbl
	Gussng	Dffclt	Dscrmn
Item.1	0.03738668	-3.2964761	0.8286287
Item.2	0.07770994	-1.1451487	0.7603748
Item.3	0.01178206	-0.2490144	0.9015777
Item.4	0.03529306	-1.7657862	0.7006545
Item.5	0.05315665	-2.9902046	0.6657969

Copy

# 문항특성곡선
plot(model)

Copy

# 문항정보곡선
plot(model, "IIC")

3 irtplay 패키지[ | ]

R

Copy

df <- read.csv("https://raw.githubusercontent.com/jmnote/ds/main/github.com/cran/ltm/data/LSAT.csv")
library(irtplay)
model <- est_irt(data=df, model="3PLM", verbose=FALSE)
coef(model)

Loading

Copy

plotICC <- function(model) {
  p1 <- model$estimates$par.1
  p2 <- model$estimates$par.2
  p3 <- model$estimates$par.3
  p3[is.na(p3)] <- 0
  D <- model$scale.D
  z <- seq(-4, 4, length=100)
  len <- nrow(model$estimates)
  pos <- round(seq(10, 90, length=len))
  for( i in 1:len ) {
    f <- function(x) {p3[i]+(1-p3[i])*plogis(p1[i]*D*(x-p2[i]))}
    if( i == 1 ) plot(z, f(z), type='l', col=i, ylim=c(0,1),
                      main="Item Characteristic Curves",
                      xlab="Ability", ylab="Probability")
    else lines(z, f(z), type='l', col=i, ylim=c(0,1))
    text(z[pos[i]], f(z[pos[i]]), adj=c(0,2), labels=c(paste('Item',i)), col=i)
  }
}
plotICC(model)

Loading

4 같이 보기[ | ]