https://qcmb.psychopen.eu/index.php/qcmb <p>The journal <em>Quantitative and Computational Methods in Behavioral Sciences</em> (QCMB) strives to foster the development of methods in psychology and related fields. To achieve this aim, QCMB publishes scientific articles that are suited to extend the understanding of foundational mathematics used in psychological methods, development of new methods and software or hardware for those, comparison of existing or new methods, and dissemination of this knowledge to a broader audience of scientists in psychology or related fields. QCMB is dedicated to Open Science: All published articles are openly available for free. There is no publication fee for the review process or the publication. QCMB makes all articles available as pre-prints as soon as they go into the review process and keeps them available regardless of the formal decision.</p> <p>QCMB publishes articles in two sections:</p> <p>The <strong>fundamental research section</strong> targets an audience of quantitative psychologists, mathematicians, and statisticians with an interest in psychological applications of computational, statistical, and mathematical models. This section publishes articles that advance fundamental research in the field of quantitative psychology. Topics include but are not limited to mathematical foundations of statistical methods, introduction and investigation of heuristics that guide data-analytic decisions, introduction of algorithms, performance analyses of existing software, simulation studies that provide insights into mathematical properties of models, the adoption of statistical learning and machine learning paradigms to psychological inquiries, and mathematical theories related to statistical analyses in quantitative psychology.</p> <p>The<strong>&nbsp;method dissemination section</strong> concentrates on methodological articles that target an audience of social scientists that want to apply top-notch analysis methods. Topics include but are not limited to introducing new statistical methods or software, tutorials explaining the application of statistical methods or software, simulation studies to compare the benefit of methods across different domains, introduction of new data tools used for methodology, and experimental design methods or tools, as for example for optimal study design planning.</p> <p>QCMB is <strong>indexed</strong> (amongst others) in:</p> <ul> <li><a href="https://app.dimensions.ai/discover/publication" target="_blank" rel="noopener">Dimensions</a></li> <li><a href="https://app.scilit.net/sources/110430" target="_blank" rel="noopener">Scilit</a></li> <li><a href="https://essentials.ebsco.com/" target="_blank" rel="noopener">EBSCO</a></li> </ul> PsychOpen, a project by Leibniz Institute for Psychology (ZPID) en-US Quantitative and Computational Methods in Behavioral Sciences 2699-8432 <p>Authors who publish with <em>Quantitative and Computational Methods in Behavioral Sciences</em> (QCMB) agree to the following terms:</p> <p><a href="https://creativecommons.org/licenses/by/4.0/" target="_blank" rel="noopener"><img style="border-width: 0; float: left; margin-right: 2em; margin-bottom: 1em;" src="https://i.creativecommons.org/l/by/4.0/88x31.png" alt="Creative Commons License"></a></p> <p>Articles are published under the <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank" rel="noopener">Creative Commons Attribution 4.0 International License</a> (CC BY 4.0).</p> <p>Under the CC BY license, authors retain ownership of the copyright for their article, but authors grant others permission to use the content of publications in QCMB in whole or in part provided that the original work is properly cited. Users (redistributors) of QCMB are required to cite the original source, including the author's names, QCMB as the initial source of publication, year of publication, volume number and DOI (if available).</p> <p>Authors may publish the manuscript in any other journal or medium but any such subsequent publication must include a notice that the manuscript was initially published by <em>Quantitative and Computational Methods in Behavioral Sciences</em>.</p> <p>Authors grant QCMB the right of first publication. Although authors remain the copyright owner, they grant the journal the irrevocable, nonexclusive rights to publish, reproduce, publicly distribute and display, and transmit their article or portions thereof in any manner.</p> https://qcmb.psychopen.eu/index.php/qcmb/article/view/10087 <p>Various likelihood-based methods are available for the parameter estimation of item response theory models (IRT), leading to comparable parameter estimates. Considering multistage testing (MST) designs, Glas (1988; https://doi.org/10.2307/1164950) stated that the conditional maximum likelihood (CML) method in its original formulation leads to severely biased parameter estimates. A modified CML estimation method for MST designs proposed by Zwitser and Maris (2015; https://doi.org/10.1007/s11336-013-9369-6) finally provides asymptotically unbiased item parameter estimates. Steinfeld and Robitzsch (2021b; https://doi.org/10.31234/osf.io/ew27f) complemented this method to MST designs with probabilistic routing strategies. For both proposed modifications additional software solutions are required since design-specific information must be incorporated into the estimation process. An R package that has implemented both modifications is "tmt". In this article, first, the proposed solutions of the CML estimation in MST designs are illustrated, followed by the main part, the demonstration of the CML item parameter estimation with the R package "tmt". The demonstration includes the process of model specification, data simulation, and item parameter estimation, considering two different routing types of deterministic and probabilistic MST designs.</p> Jan Steinfeld Alexander Robitzsch Copyright (c) 2023 Jan Steinfeld, Alexander Robitzsch https://creativecommons.org/licenses/by/4.0 2023-11-06 2023-11-06 1 32 10.5964/qcmb.10087 https://qcmb.psychopen.eu/index.php/qcmb/article/view/12069 <p>The use of classifiers provides an alternative to conventional statistical methods. This involves using the accuracy with which data is correctly assigned to a given group by the classifier to apply tests to compare the performance of classifiers. The conventional validation methods for determining the accuracy of classifiers have the disadvantage that the distribution of correct classifications does not follow any known distribution, and therefore, the application of statistical tests is problematic. Independent validation circumvents this problem and allows the use of binomial tests to assess the performance of classifiers. However, independent validation accuracy is subject to bias for small training datasets. The present study shows that a hyperbolic function can be used to estimate the loss in classifier accuracy for independent validation. This function is used to develop three new methods to estimate the classifier accuracy for small training sets more precisely. These methods are compared to two existing methods in a simulation study. The results indicate overall small errors in the estimation of classifier accuracy and indicate that independent validation can be used with small samples. A least square estimation approach seems best suited to estimate the classifier accuracy.</p> Tina Braun Hannes Eckert Timo von Oertzen Copyright (c) 2023 Tina Braun, Hannes Eckert, Timo von Oertzen https://creativecommons.org/licenses/by/4.0 2023-12-22 2023-12-22 1 30 10.5964/qcmb.12069