A Generalized Partial Credit Model

Muraki, Eiji

doi:10.1007/978-1-4757-2691-6_9

Cited by 180 publications

(147 citation statements)

References 12 publications

Supporting

Mentioning

143

Contrasting

Unclassified

Order By: Relevance

“…The approach of Stocking and Lord (Stocking & Lord, 1983) is used with a generalized partial credit model (Muraki, 1997). This approach reflects practices of the particular testing program under study.…”

Section: Irt True-score Equatingmentioning

confidence: 99%

See 1 more Smart Citation

Jackknifing Techniques for Evaluation of Equating Accuracy

Haberman

Lee

Qian

2009

ETS Research Report Series

View full text Add to dashboard Cite

Grouped jackknifing may be used to evaluate the stability of equating procedures with respect to sampling error and with respect to changes in anchor selection. Properties of grouped jackknifing are reviewed for simple-random and stratified sampling, and its use is described for comparisons of anchor sets. Application is made to examples of item response theory (IRT) true-score equating in which two-parameter logistic and general partial credit models are employed.

show abstract

Section: Irt True-score Equatingmentioning

confidence: 99%

“…where item discrimination a m is an unknown positive real number, item difficulty b m is an unknown real number, and the category coefficients d mk , 1 ≤ k ≤ r m − 1, are real numbers unknown save for the constraint that their sum is 0 (Muraki, 1997). Thus d uk = 0 if r m = 2.…”

Section: Irt True-score Equatingmentioning

confidence: 99%

Jackknifing Techniques for Evaluation of Equating Accuracy

Haberman

Lee

Qian

2009

ETS Research Report Series

View full text Add to dashboard Cite

show abstract

“…The first consideration when choosing the right model involves the number of item response categories. For polytomous items, variations of the Partial Credit Model (Masters, 1982); Rating Scale Model (Andrich, 1978); Generalized Partial Credit Model (Muraki, 1992(Muraki, , 1997 as well as the GRM (Samejima 1969(Samejima , 1997 are available for ordered responses, and the Nominal Model (Bock, 1972) is appropriate for items with a non-specified response order. We consider the GRM (Samejima, 1997) because it is appropriate to use when item responses are ordered categorical responses.…”

Section: Graded Response Modelmentioning

confidence: 99%

The effects of scanning position on evaluation of cerebral atrophy level: assessed by item response theory

Mahsin

Zhao

2016

CSAM

View full text Add to dashboard Cite

Cerebral atrophy affects the brain and is a common feature of patients with mild cognitive impairment or Alzheimer's diseases. It is evaluated by the radiologist or reader based on patient's history, age and the space between the brain and the skull as indicated by magnetic resonance (MR) images. A total of 70 patients were scanned in the supine and prone positions before three radiologist assessed their atrophy level. This study examined the radiologist's assessment of the cerebral atrophy level using a graded response model of item response theory (IRT). A graded response model (GRM) is fitted to our data and then item-fit and person-fit statistics are evaluated to assess the fitted model. Our analysis found that the cerebral atrophy level is better discriminated by readers in the prone position because all item slopes were greater than 2 at this position, versus the supine position where all the slope parameters were less than 1. However, the thresholds are very similar for the first reader and are quite different for the second and third readers because the scanning position affects readers differently as the category threshold estimates vary considerably between the readers.

show abstract

“…Calibration of the IRT model and evaluation of item fit. We used the generalized partial credit model (GPCM) (Muraki, 1997) as the IRT model forming the basis for the CAT.…”

Section: Field-testing and Psychometric Analysesmentioning

confidence: 99%

“…In the GPCM, each item has a slope parameter describing the item's ability to discriminate between subjects with different levels of pain, and a set of threshold parameters describing how likely it is to report problems on the item. Item fit was examined using Muraki's test (Muraki, 1997), bias estimates (average difference between expected and observed item responses) and the infit and outfit statistics, which are mean square residuals often used in Rasch fit analysis (Bond and Fox, 2007;Wright and Linacre, 1994;Wright and Masters, 1982). Infit and outfit values between 0.7 and 1.3 are often regarded as acceptable (Wright and Linacre, 1994).…”

Section: Field-testing and Psychometric Analysesmentioning

confidence: 99%

Development of an item bank for computerized adaptive test (CAT) measurement of pain

et al. 2015

View full text Add to dashboard Cite

Abstract:Background: Patient-reported outcomes should ideally be adapted to the individual patient while maintaining comparability of scores across patients. This is achievable using computerized adaptive testing (CAT). The aim here was to develop an item bank for CAT measurement of the pain domain as measured by the EORTC QLQ-C30 questionnaire. Methods: The development process consisted of four steps. 1) Literature search. 2) Formulation of new items. 3) Pre-testing. 4) Field-testing and psychometric analyses for the final selection of items. Results: In step 1) we identified 337 pain items from the literature. 2) Twenty-nine new items fitting the QLQ-C30 item style were formulated. Expert evaluations reduced this to 26 items. 3) Based on patient interviews (N=31) the list was further reduced to 21 items. 4) We obtained responses from 1,103 cancer patients from five countries. Psychometric evaluations showed that 16 items could be retained in a unidimensional item bank. Evaluations indicated that use of the CAT measure may reduce sample size requirements with 15-25% compared to using the QLQ-C30 pain scale. Conclusions: We have established an item bank of 16 items suitable for CAT measurement of pain. We recommend initiating CAT measurement by screening for pain using the two original QLQ-C30 pain items.

show abstract

A Generalized Partial Credit Model

Cited by 180 publications

References 12 publications

Jackknifing Techniques for Evaluation of Equating Accuracy

Jackknifing Techniques for Evaluation of Equating Accuracy

The effects of scanning position on evaluation of cerebral atrophy level: assessed by item response theory

Development of an item bank for computerized adaptive test (CAT) measurement of pain

Contact Info

Product

Resources

About