Article
Psychology, Applied
Philseok Lee, Seang-Hwane Joo, Stephen Stark
Summary: The study investigated DIF detection with three-alternative MFC items based on the Thurstonian IRT model, showing that the free baseline strategy was highly effective in detecting DIF in large sample sizes and large DIF conditions, with similar effectiveness in impact and no-impact conditions.
ORGANIZATIONAL RESEARCH METHODS
(2021)
Article
Psychology, Educational
Zhichen Guo, Daxun Wang, Yan Cai, Dongbo Tu
Summary: Forced-choice measures are widely used in personality or attitude tests to reduce response biases, and incorporating response time into these measures can better reveal respondents' behaviors or preferences. This study proposes a new item response theory model that integrates response time to improve personality assessment. Simulation studies and real data application show that this model can effectively enhance the estimation accuracy of personality traits and explain the differences compared to conventional models.
EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT
(2023)
Article
Mathematics, Interdisciplinary Applications
Susanne Frick
Summary: This study introduces a mixture item response theory model to estimate faking risk in the multidimensional forced-choice (MFC) format. Simulation and empirical validation results indicate that matching is necessary to reduce faking risk, but not sufficient to completely eliminate it.
Article
Mathematics, Interdisciplinary Applications
Xuelan Qiu, Jimmy de la Torre
Summary: This study developed a new item response theory (IRT) model for polytomous multidimensional forced-choice (MFC) items, which can assess individual psychological differentiation in interests, values, and emotions and compare the differentiation levels between individuals. Simulation studies showed that the model can recover the parameters well with sufficient sample size. Although some aspects of the model require further validation, the approach shows promise.
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY
(2023)
Article
Psychology, Mathematical
Qin Wang, Yi Zheng, Kai Liu, Yan Cai, Siwei Peng, Dongbo Tu
Summary: Multidimensional computerized adaptive testing for forced-choice items (MFC-CAT) combines the benefits of MFC items and CAT, providing unbiased responses and reduced testing time. Previous studies focused on item selection methods based on Fisher information (FI), which may be unstable at early stages. This study proposes new item selection methods based on KL information for MFC-CAT, and compares their performance with FI-based methods through simulation studies. The results demonstrate that the proposed KL-based methods generate accurate trait estimation and even utilization of the item pool, outperforming the FI-based methods.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Mathematics, Interdisciplinary Applications
Seang-Hwane Joo, Philseok Lee, Stephen Stark
Summary: This research developed a new ideal point-based item response theory (IRT) model, called ZG-MUPP, for multidimensional forced choice (MFC) measures. Through simulation study, it was found that the proposed model accurately estimated model parameters even with low sample sizes, and the scores obtained from the ZG-MUPP model were comparable to those from other models.
MULTIVARIATE BEHAVIORAL RESEARCH
(2023)
Article
Psychology, Applied
Yin Lin
Summary: Forced-choice assessments are popular in high-stakes organizational testing scenarios due to their resistance against response distortions, and the measurement precision of FC assessment scores is crucial. Various reliability estimates are reported for FC assessment scores, but there is a lack of consensus on best practices.
ORGANIZATIONAL RESEARCH METHODS
(2022)
Article
Mathematics, Interdisciplinary Applications
Luning Sun, Zijie Qin, Shan Wang, Xuetao Tian, Fang Luo
Summary: Forced-choice questionnaires involve presenting items in blocks and asking respondents to rank the items within each block. To prevent response distortions, blocks are formed with items of similar desirability. Constructing item pairs that are matched by desirability requires careful consideration of the relationships between the items.
MULTIVARIATE BEHAVIORAL RESEARCH
(2023)
Article
Psychology, Educational
Niklas Schulte, Heinz Holling, Paul-Christian Buerkner
Summary: Forced-choice questionnaires can prevent faking and response biases, but derived trait scores are often unreliable and ipsative, making interindividual comparisons difficult in high-stakes situations. Research suggests that increasing the number of measured traits can alleviate these issues.
EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT
(2021)
Article
Mathematics, Interdisciplinary Applications
Kyosuke Bunji, Kensuke Okada
Summary: This study introduces a joint model combining response time and MAFC item responses, using the linear ballistic accumulator model and MCMC algorithm for parameter estimation. Simulation and empirical applications demonstrate the effectiveness of the proposed model in jointly modeling MAFC item responses and response times.
MULTIVARIATE BEHAVIORAL RESEARCH
(2022)
Article
Multidisciplinary Sciences
Anna Frey, Tobias Leutritz, Joy Backhaus, Alexander Hoernlein, Sarah Koenig
Summary: Testing based on multiple choice questions (MCQ) is a widely used assessment method. This study evaluated the suitability and efficiency of extended matching questions (EMQ) as an assessment method, comparing it with three other MCQ types. The study found that EMQ had higher item discrimination and discrimination index compared to other item types, although they were slightly more difficult to score. EMQ required more processing time but were perceived as challenging and clinically relevant by students. Using a small number of EMQ items still maintained statistical quality of the overall examination.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Interdisciplinary Applications
Paul-Christian Buerkner
Summary: Item response theory (IRT) is often implemented in the human sciences using R packages like brms and the probabilistic programming language Stan. These tools allow for the specification and fitting of Bayesian IRT models with various response distributions and parameter configurations, making it easy to extract posterior distributions and evaluate model fit.
JOURNAL OF STATISTICAL SOFTWARE
(2021)
Article
Engineering, Industrial
Kevin Lai, Jason T. Jameson, Adam T. Biggs, Peter G. Roma, Dale W. Russell
Summary: Effective workplace operations rely on norms and practices that create an overall safety climate. Developing practical measurement tools for safety climate has been a challenge. This study used statistical techniques to create a Safety Climate Index (SCI) that proved to be reliable and operationally useful for assessing safety climate in high-risk organizations.
Article
Mathematics, Interdisciplinary Applications
Hui-Fang Chen, Kuan-Yu Jin
Summary: This study examined the impact of negatively worded items in mixed-format designs and found that they had an effect on item responses, with a minimal affirmative preference regardless of the proportion of negatively worded items. Additionally, participants showed a tendency to prefer extreme responses in mixed-format designs.
MULTIVARIATE BEHAVIORAL RESEARCH
(2022)
Article
Mathematics, Interdisciplinary Applications
Susanne Frick, Anna Brown, Eunike Wetzel
Summary: The study evaluated the normativity of Thurstonian IRT trait estimates in multidimensional forced-choice (MFC) data through simulation and empirical analysis. The results showed that Thurstonian IRT trait estimates were normative in contrast to classical partially ipsative estimates. Factors such as unbalanced numbers of items per trait, few opposite keyed items, traits correlated positively or assessing fewer traits did not significantly decrease measurement precision. Measurement precision was lower than that of rating scale data. The study also found that relative MFC responses did not provide better differentiation of behaviors within persons compared to absolute TF responses.
MULTIVARIATE BEHAVIORAL RESEARCH
(2023)
Article
Psychology, Biological
Markus T. Jansen, Sascha Tamm, Thomas Merten, Anett Tamm, Markus J. Hofmann
INTERNATIONAL JOURNAL OF PSYCHOPHYSIOLOGY
(2020)
Article
Psychology, Social
Sascha Schwarz, Lisa Kluemper, Markus Thomas Jansen, Maria Agthe
Summary: The researchers found that individual differences, particularly in pathogen and moral disgust sensitivity, can predict perceptions of violations against COVID-19 mitigation actions. These findings suggest that individual differences play an important role in evaluating violations against mitigation actions.
SOCIAL AND PERSONALITY PSYCHOLOGY COMPASS
(2023)