mirt (version 1.46.1)

SAT12: Description of SAT12 data

Description

Data obtained from the TESTFACT (Woods et al., 2003) manual, with 32 response pattern scored items for a grade 12 science assessment test (SAT) measuring topics of chemistry, biology, and physics. The scoring key for these data is [1, 4, 5, 2, 3, 1, 2, 1, 3, 1, 2, 4, 2, 1, 5, 3, 4, 4, 1, 4, 3, 3, 4, 1, 3, 5, 1, 3, 1, 5, 4, 5], respectively. However, careful analysis using the nominal response model suggests that the scoring key for item 32 may be incorrect, and should be changed from 5 to 3.

Author

Phil Chalmers

References

Wood, R., Wilson, D. T., Gibbons, R. D., Schilling, S. G., Muraki, E., & Bock, R. D. (2003). TESTFACT 4 for Windows: Test Scoring, Item Statistics, and Full-information Item Factor Analysis [Computer software]. Lincolnwood, IL: Scientific Software International.

Examples

# \donttest{

itemstats(SAT12, use_ts = FALSE)

# score the data (missing scored as 0)
head(SAT12)
dat <- key2binary(SAT12,
    key = c(1,4,5,2,3,1,2,1,3,1,2,4,2,1,5,3,4,4,1,4,3,3,4,1,3,5,1,3,1,5,4,5))
head(dat)
itemstats(dat)

# score the data, missing (value of 8) treated as NA
SAT12missing <- SAT12
SAT12missing[SAT12missing == 8] <- NA
dat <- key2binary(SAT12missing,
    key = c(1,4,5,2,3,1,2,1,3,1,2,4,2,1,5,3,4,4,1,4,3,3,4,1,3,5,1,3,1,5,4,5))
head(dat)

# potentially better scoring for item 32 (based on nominal model finding)
dat <- key2binary(SAT12,
    key = c(1,4,5,2,3,1,2,1,3,1,2,4,2,1,5,3,4,4,1,4,3,3,4,1,3,5,1,3,1,5,4,3))

# fit 2PL model to each item
mod <- mirt(dat)
plot(mod)
itemfit(mod)

# fit monotonic spline model to items 4 through 6 to improve fit
itemtype <- rep('2PL', 32)
itemtype[4:6] <- 'monospline'
mod_mspline <- mirt(dat, itemtype = itemtype)
anova(mod, mod_mspline)

plot(mod, which.items=4:6, type = 'trace')
plot(mod_mspline, which.items=4:6, type = 'trace')

# }