Article
Education & Educational Research
Francesco Innocenti, Math J. J. M. Candel, Frans E. S. Tan, Gerard J. P. van Breukelen
Summary: Normative studies are necessary for comparing individuals on relevant measures, and norms can be obtained efficiently through regression-based approaches. Multivariate regression-based approaches are important in considering correlations between measures and reducing the number of significance tests. A new approach using Mahalanobis distance is proposed for combining multiple measures and providing an overall performance indicator.
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICS
(2023)
Article
Psychology, Mathematical
Susan Teubner-Rhodes, Andrew Luu, Rebecca Dunterman, Kenneth I. Vaden Jr
Summary: Conflicting tasks can improve performance in subsequent trials. This study controlled associative learning and found that performance adjustments still occurred. After conflict, participants had higher word recognition rates in incongruent trials and made more phonologically related errors. Postconflict improvements seem to be due to better resolution of phonological conflict rather than increased attention to the picture or speech signal.
PSYCHONOMIC BULLETIN & REVIEW
(2023)
Article
Psychology, Mathematical
Rani Gera, Segev Barak, Tom Schonberg
Summary: Habits are prominent features of human behavior, but there are currently no well-established experimental procedures for inducing habits in humans. To address this issue, researchers have developed a unique smartphone application that allows participants to freely enter and use the app, providing a novel method for studying the neurobehavioral and psychological mechanisms underlying habits in humans.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Mathematical
Chen-Wei Liu
Summary: This paper proposes a novel statistical approach, namely the beta copula model and the competing logit-normal model, for analyzing testlet-based visual analogue scale (VAS) data. The empirical data analysis demonstrates that the beta copula model has a superior fit, and the simulation studies show good parameter recovery.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Mathematical
Michael John Ilagan, Carl F. Falk
Summary: This article introduces a model-agnostic and unsupervised bot detection algorithm that uses permutation test with leave-one-out calculations of outlier statistics. The algorithm provides a p value for each respondent to determine if they are a bot. The simulation study shows that the proposed algorithm outperforms naive alternatives in terms of sensitivity calibration and classification accuracy.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Mathematics, Interdisciplinary Applications
Chenchen Ma, Jing Ouyang, Chun Wang, Gongjun Xu
Summary: Survey instruments and assessments are widely used in social science research. Multidimensional item response theory (MIRT) provides a framework and statistical tool for analyzing, calibrating, and scoring these instruments. However, traditional estimation methods face computational challenges when dealing with large dimensions, sample sizes, and test lengths. Variational estimation methods, such as Gaussian variational expectation-maximization (GVEM), offer a faster solution but may introduce bias in discrimination parameters. This study proposes an importance-weighted version of GVEM (IW-GVEM) to correct for such bias and shows promising results through simulations.
Review
Psychology, Mathematical
Adam F. Osth, Lyulei Zhang
Summary: In recognition memory, retrieval is believed to be based on the global similarity between the probe and studied items. This study integrates perceptual representations of letter strings with global similarity models and finds that relative position models are favored. When semantic representations are incorporated into the models, it is found that orthographic representations are almost equally important as semantic representations in determining inter-item similarity and false recognition errors. The model is able to modestly capture individual word variability in false alarm rates, but has limitations in capturing variability in hit rates.
PSYCHONOMIC BULLETIN & REVIEW
(2023)
Article
Psychology, Educational
Sandip Sinharay, Matthew S. Johnson
Summary: This paper introduces a method of reusing the multigroup multidimensional Rasch model to model and analyze data from culturally responsive assessments, and provides conditions for computing comparable scores. The performance of the model on simulated data from hypothetical culturally responsive assessments is evaluated, and recommendations are made for measurement practitioners interested in this area.
JOURNAL OF EDUCATIONAL MEASUREMENT
(2023)
Article
Psychology, Mathematical
Alice Mason, Elliot A. Ludvig, Marcia L. Spetch, Christopher R. Madan
Summary: This study examines the impact of the combination of rarity and extremity biases on decision-making. It finds that when events are both rare and extreme, people's risk preferences shift.
PSYCHONOMIC BULLETIN & REVIEW
(2023)
Article
Mathematics, Interdisciplinary Applications
Yang Liu, Weimeng Wang
Summary: It is believed that a joint factor analysis of item responses and response time (RT) can improve the precision of ability scores. This study analyzed the 2015 Programme for International Student Assessment mathematics data using a semiparametric simple-structure model and found that a simple factor structure can fit the data well under certain assumptions. Furthermore, the study discovered a strong but nonlinear association between latent ability and speed/slowness.
Correction
Psychology, Mathematical
Ana Isabel Correia, Margherita Vincenzi, Patricia Vanzella, Ana P. Pinheiro, Cesar F. Lima, E. Glenn Schellenberg
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Mathematical
Tianqi Wang, Xu Xu
Summary: Word affective ratings play a crucial role in psycholinguistic research and natural language processing. However, the existing norms for affective ratings are often limited in scale. In this study, a computational neural network was implemented to extrapolate the affective values of Chinese words based on their vector-based semantic representations. The resulting estimates of affective values correlated well with human ratings and also captured the variability in human ratings. The extrapolated affective values for over 38,000 Chinese words were made available in the Open Science Framework.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Mathematical
Alaa Alzahrani
Summary: This study created and validated a quick test for measuring L2 Arabic proficiency, addressing the lack of research and tools in this area. The test, called LexArabic, demonstrated good reliability, validity, and accuracy, making it a valuable tool for experimental research and standardized proficiency assessment of Arabic speakers.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Mathematics, Interdisciplinary Applications
Jason Hou-Liu, Ryan P. Browne
Summary: In this paper, a model formulation and estimation procedure for clustering with nested Gaussian clusters in orthogonal intrinsic variable subspaces are described. The model considers observed variables to be rotations of intrinsic primary and secondary clustering subspaces with additional noise subspaces. An estimation procedure using the expectation-maximization algorithm is provided with model selection via Bayesian information criterion.
JOURNAL OF CLASSIFICATION
(2023)
Article
Psychology, Educational
Philippe Goldammer, Peter Lucas Stockli, Yannik Andrea Escher, Hubert Annen, Klaus Jonas
Summary: Indirect indices for faking detection in questionnaires use a respondent's deviant or unlikely response pattern to identify them as a faker, offering advantages over direct faking indices. This study compared the performance of different indirect faking detection indices and found that the Likert-type item response process tree model, proportion of desirable scale endpoint responses, and covariance index performed the best. Using indirect indices in combination resulted in comparable or better detection rates than direct faking measures. Some effective indirect indices had minimal correlation with substantive scales, making them useful for detecting faking without losing substance. Researchers are encouraged to use indirect indices for detecting faking in their data.
EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT
(2023)
Article
Psychology, Mathematical
Joaquina Couto, Mael Lebreton, Leendert van Maanen
Summary: This paper presents an empirical diagnostic method to detect a mixture of processes and uses resampling of real experimental data to simulate the effects of variations in experimental design. The study aims to validate the fixed-point property of binary mixture data and provide performance metrics for researchers testing the fixed-point property on their experimental data.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Mathematics, Interdisciplinary Applications
Xiaying Zheng
Summary: This study proposes two new single-level methods to support random-slope estimation and compares them with existing methods. The findings suggest that two existing single-level methods can support random-intercept models, while one proposed single-level method presents an efficient alternative to multilevel latent regression and recovers acceptable estimates.
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY
(2023)
Article
Psychology, Mathematical
Robin Hake, Michel Buergel, Ninh K. Nguyen, Alinka Greasley, Daniel Muellensiefen, Kai Siedenburg
Summary: Auditory scene analysis (ASA) is a process in the auditory system where it organizes sound mixtures into meaningful events and streams. This study introduces a new tool, Musical Scene Analysis (MSA) test, to assess ASA abilities in the context of music for both normal-hearing and hearing-impaired individuals. The results reveal important factors affecting item difficulty, such as the level ratio between the target instrument and mixture and the number of instruments in the mixture, while stereo width has minimal impact.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Mathematical
Elena Benini, Malte Moeller, Iring Koch, Andrea M. Philipp, Ruyi Qiu, Susanne Mayr
Summary: In task switching, repeating a response usually improves performance, but only when the task also repeats. This study provides evidence that repeating response errors are more likely in task repetitions than in task switches, supporting the importance of task-response binding in the repetition effect.
PSYCHONOMIC BULLETIN & REVIEW
(2023)
Article
Psychology, Educational
Okan Bulut, Guher Gorgun, Hacer Karamese
Summary: The use of multistage adaptive testing (MST) has been increasing in large-scale testing programs as it offers a balance between linear test design and item-level adaptive testing. However, research has shown that a lack of test-taking engagement can impact the measurement accuracy of MST. To address this issue, test-taking engagement can be incorporated into the on-the-fly module assembly procedure to minimize the impact of noneffortful responses.
JOURNAL OF EDUCATIONAL MEASUREMENT
(2023)
