Optimal Scores: An Alternative to Parametric Item Response Theory and Sum Scores

Wiberg, Marie; Ramsay, J. O.; Li, Juan

doi:10.1007/s11336-018-9639-4

Cited by 15 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The lower dimensional curves defined by models such as the two-parameter graded response and partial credit models have the interpretational advantage of having the parameters a and b that directly reflect what we call sensitivity and location, but the disadvantage of being unable to capture more complex variation such as we see in Figures 8 and 9. We believe, however, that these and other simple IRT models would also deliver significant improvement in intensity estimation accuracy over that provided by the sum score, and have already reported on this in in the multiple choice testing context in [15].…”

Section: Discussionmentioning

confidence: 81%

“…Further information on splines and curve estimating with spline can be found in [13,14]. A comparison of results using two-parameter logistic curves and using splines curves for correct responses for a large-scale multiple choice test is [15]. The bottom panel of Figure 8 displays the surprisal curves in 5-bit units for a heterogeneous graded response model corresponding to the probability curves in the top panels.…”

Section: The Multinomial Surprisal Vectormentioning

confidence: 99%

See 1 more Smart Citation

Better Rating Scale Scores with Information–Based Psychometrics

Ramsay

Wiberg

2020

Psych

Self Cite

View full text Add to dashboard Cite

Diagnostic scales are essential to the health and social sciences, and to the individuals that provide the data. Although statistical models for scale data have been researched for decades, it remains nearly universal that scale scores are sums of weights assigned a priori to question choice options (sum scores), respectively. We propose several modifications of psychometric testing theory that together demonstrate remarkable improvements in the quality of rating scale scores. Our model represents performance as a space with a metric structure by transforming probability into surprisal or information. The estimation algorithm permits the analysis of data from tens and hundreds of thousands of test takers in a few minutes on consumer level computing equipment. Standard errors of performance estimates are shown to be as small as a quarter of those of sum scores. Open access software resources are presented.

show abstract

Section: Discussionmentioning

confidence: 81%

Section: The Multinomial Surprisal Vectormentioning

confidence: 99%

Better Rating Scale Scores with Information–Based Psychometrics

Ramsay

Wiberg

2020

Psych

Self Cite

View full text Add to dashboard Cite

show abstract

“…proposed by Arenson and Karabatsos (2018), which uses no specific parametric function as base distribution, was able to perform better than the parametric 2PL model, especially when symmetric priors, as the ones used in the present study, are used. Finally, these models can all be compared to sum scores, which can be considered as lower bound benchmarks for the performance of the models (Wiberg, Ramsay, & Li, 2018).…”

Section: Procedures' Performance and Hypothesismentioning

confidence: 99%

Nonparametric Item Response Models: A Comparison on Recovering True Scores

Vr¹,

Wiberg²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Nonparametric procedures are used to add flexibility to models. Three nonparametric item response models have been proposed, but not directly compared: the Kernel smoothing (KS-IRT); the Davidian-Curve (DC-IRT); and the Bayesian semiparametric Rasch model (SP-Rasch). The main aim of the present study is to compare the performance of these procedures in recovering simulated true scores, using sum scores as benchmarks. The secondary aim is to compare their performances in terms of practical equivalence with real data. Overall, the results show that, apart from the DC-IRT, which is the model that performs the worse, all the other models give results quite similar to those when sum scores are used. These results are followed by a discussion with practical implications and recommendations for future studies.

show abstract

“…In recent years, flexible item response models have increasingly been used in confirmatory contexts. For example, flexible item response models have recently been applied to computerized adaptive testing [7,8], the creation of item banks for measuring health outcomes [9], and the development of optimal scoring procedures [10].…”

Section: Introductionmentioning

confidence: 99%

Flexible Item Response Modeling in R with the flexmet Package

Feuerstahler

2021

Psych

View full text Add to dashboard Cite

The filtered monotonic polynomial (FMP) model is a semi-parametric item response model that allows flexible response function shapes but also includes traditional item response models as special cases. The flexmet package for R facilitates the routine use of the FMP model in real data analysis and simulation studies. This tutorial provides several code examples illustrating how the flexmet package may be used to simulate FMP model parameters and data (both for dichotomous and polytomously scored items), estimate FMP model parameters, transform traditional item response models to different metrics, and more. This tutorial serves as both an introduction to the unique features of the FMP model and as a practical guide to its implementation in R via the flexmet package.

show abstract

Optimal Scores: An Alternative to Parametric Item Response Theory and Sum Scores

Cited by 15 publications

References 13 publications

Better Rating Scale Scores with Information–Based Psychometrics

Better Rating Scale Scores with Information–Based Psychometrics

Nonparametric Item Response Models: A Comparison on Recovering True Scores

Flexible Item Response Modeling in R with the flexmet Package

Contact Info

Product

Resources

About