Nonunique solutions to the likelihood equation for the three-parameter logistic model

Yen, Wendy M.; Burket, George R.; Sykes, Robert C.

doi:10.1007/bf02294584

Cited by 30 publications

(22 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These evaluations found that PARMATE produced item parameter estimates that were almost always slightly more accurate than those of both B1LOG and LOGIST. For a given set of item parameters, and in the absence of multiple maxima in the likelihood function (Yen, Burket, & Sykes, 1991), maximum likelihood trait estimates produced by PARMATE and LOGIST are identical.…”

Section: Scaling Proceduresmentioning

confidence: 99%

Comparison of Item Response Theory and Thurstone Methods of Vertical Scaling

Yen¹,

Burket²

1997

J Educational Measurement

Self Cite

View full text Add to dashboard Cite

Vertical achievement scales, which range from the lower elementary grades to high school, are used pervasively in educational assessment. Using simulated data modeled after real tests, the present article examines two procedures available for vertical scaling: a Thurstone method and three‐parameter item response theory. Neither procedure produced artifactual scale shrinkage; both procedures produced modest scale expansion for one simulated condition.

show abstract

Section: Scaling Proceduresmentioning

confidence: 99%

Comparison of Item Response Theory and Thurstone Methods of Vertical Scaling

Yen¹,

Burket²

1997

J Educational Measurement

Self Cite

View full text Add to dashboard Cite

show abstract

“…Then, for any values of , we could expect the converged values, ( , , , , and ). We cannot always guarantee that the converged values are uniquely determined ( [13], [14]). However, we have experienced that the converged values are uniquely determined in many cases.…”

Section: Matrix Completion By Using the Em-type Irtmentioning

confidence: 99%

Automatically growing dually adaptive online IRT testing

Hirose

Aizawa

2014

2014 IEEE International Conference on Teaching, Assessment and Learning for Engineering (TALE)

View full text Add to dashboard Cite

The item response theory (IRT) is widely used in many fields recently because it provides us the abilities of examinees and the difficulties of items simultaneously. Online IRT test systems are often equipped with the adaptive testing. Such a testing method selects the most appropriate items to examinees automatically. However, the item difficulty values remain the same even if the features of the examinees and the size of the items change. Calibration of the difficulty values is required in such cases, and the dually adaptive online IRT testing system works, where "dually adaptive" means that one is targeted to the adequate item selection and the other is targeted to the adjustment of the item difficulty values. Using this, the database in the system will automatically be growing and be updated. In this paper, we introduce such a novel system with applications.

show abstract

“…We cannot always guarantee that the converged values are uniquely determined ( [15], [16]). However, we have experienced that the converged values are uniquely determined in many cases.…”

Section: A Matrix Completion By Using the Em-type Irtmentioning

confidence: 99%

A simulation study to the dually adaptive online IRT testing system

Hirose

Tokusada

2014

2014 IEEE International Conference on Teaching, Assessment and Learning for Engineering (TALE)

View full text Add to dashboard Cite

The item response theory (IRT) gives us valuable information about the difficulties of items as well as the abilities of examinees. Adaptive testing using the IRT provides the most appropriate items to examinees automatically. However, the item parameters remain the same even if the features of the examinees and the number of examinees change. We propose to use the dually adaptive online IRT testing system, where "dually adaptive" means that one is targeted to the adequate item selection and the other is to the adjustment of the difficulty values for items. The key idea for this is to use the matrix completion method to the incomplete matrix. By using the simulation study, we found that the proposed method works well.

show abstract

Nonunique solutions to the likelihood equation for the three-parameter logistic model

Abstract: item response theory, maximum likelihood estimation, tests,

Cited by 30 publications

References 6 publications

Comparison of Item Response Theory and Thurstone Methods of Vertical Scaling

Comparison of Item Response Theory and Thurstone Methods of Vertical Scaling

Automatically growing dually adaptive online IRT testing

A simulation study to the dually adaptive online IRT testing system

Contact Info

Product

Resources

About