Psychological methods最新文献

Many psychological theories assume that observable responses are determined by multiple latent processes. Multinomial processing tree (MPT) models are a class of cognitive models for discrete responses that allow researchers to disentangle and measure such processes. Before applying MPT models to specific psychological theories, it is necessary to tailor a model to specific experimental designs. In this tutorial, we explain how to develop, fit, and test MPT models using the classical pair-clustering model as a running example. The first part covers the required data structures, model equations, identifiability, model validation, maximum-likelihood estimation, hypothesis tests, and power analyses using the software multiTree. The second part introduces hierarchical MPT modeling which allows researchers to account for individual differences and to estimate the correlations of latent processes among each other and with additional covariates using the TreeBUGS package in R. All examples including data and annotated analysis scripts are provided at the Open Science Framework (https://osf.io/24pbm/). (PsycInfo Database Record (c) 2025 APA, all rights reserved).

The outcomes in single-case experimental designs (SCEDs) are often counts or proportions. In our study, we provided a colloquial illustration for a new class of generalized linear mixed models (GLMMs) to fit count and proportion data from SCEDs. We also addressed important aspects in the GLMM framework including overdispersion, estimation methods, statistical inferences, model selection methods by detecting overdispersion, and interpretations of regression coefficients. We then demonstrated the GLMMs with two empirical examples with count and proportion outcomes in SCEDs. In addition, we conducted simulation studies to examine the performance of GLMMs in terms of biases and coverage rates for the immediate treatment effect and treatment effect on the trend. We also examined the empirical Type I error rates of statistical tests. Finally, we provided recommendations about how to make sound statistical decisions to use GLMMs based on the findings from simulation studies. Our hope is that this article will provide SCED researchers with the basic information necessary to conduct appropriate statistical analysis of count and proportion data in their own research and outline the future agenda for methodologists to explore the full potential of GLMMs to analyze or meta-analyze SCED data. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Continuous-time (CT) models are a flexible approach for modeling longitudinal data of psychological constructs. When using CT models, a researcher can assume one underlying continuous function for the phenomenon of interest. In principle, these models overcome some limitations of discrete-time (DT) models and allow researchers to compare findings across measures collected using different time intervals, such as daily, weekly, or monthly intervals. Theoretically, the parameters for equivalent models can be rescaled into a common time interval that allows for comparisons across individuals and studies, irrespective of the time interval used for sampling. In this study, we carry out a Monte Carlo simulation to examine the capability of CT autoregressive (CT-AR) models to recover the true dynamics of a process when the sampling interval is different from the time scale of the true generating process. We use two generating time intervals (daily or weekly) with varying strengths of the AR parameter and assess its recovery when sampled at different intervals (daily, weekly, or monthly). Our findings indicate that sampling at a faster time interval than the generating dynamics can mostly recover the generating AR effects. Sampling at a slower time interval requires stronger generating AR effects for satisfactory recovery, otherwise the estimation results show high bias and poor coverage. Based on our findings, we recommend researchers use sampling intervals guided by theory about the variable under study, and whenever possible, sample as frequently as possible. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

In psychology, researchers often predict a dependent variable (DV) consisting of multiple measurements (e.g., scale items measuring a concept). To analyze the data, researchers typically aggregate (sum/average) scores across items and use this as a DV. Alternatively, they may define the DV as a common factor using structural equation modeling. However, both approaches neglect the possibility that an independent variable (IV) may have different relationships to individual items. This variance in individual item slopes arises because items are randomly sampled from an infinite pool of items reflecting the construct that the scale purports to measure. Here, we offer a mixed-effects model called random item slope regression, which accounts for both similarities and differences of individual item associations. Critically, we argue that random item slope regression poses an alternative measurement model to common factor models prevalent in psychology. Unlike these models, the proposed model supposes no latent constructs and instead assumes that individual items have direct causal relationships with the IV. Such operationalization is especially useful when researchers want to assess a broad construct with heterogeneous items. Using mathematical proof and simulation, we demonstrate that random item slopes cause inflation of Type I error when not accounted for, particularly when the sample size (number of participants) is large. In real-world data (n = 564 participants) using commonly used surveys and two reaction time tasks, we demonstrate that random item slopes are present at problematic levels. We further demonstrate that common statistical indices are not sufficient to diagnose the presence of random item slopes. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

The accuracy of factor retention methods for structures with one or more general factors, like the ones typically encountered in fields like intelligence, personality, and psychopathology, has often been overlooked in dimensionality research. To address this issue, we compared the performance of several factor retention methods in this context, including a network psychometrics approach developed in this study. For estimating the number of group factors, these methods were the Kaiser criterion, empirical Kaiser criterion, parallel analysis with principal components (PA_PCA) or principal axis, and exploratory graph analysis with Louvain clustering (EGA_LV). We then estimated the number of general factors using the factor scores of the first-order solution suggested by the best two methods, yielding a "second-order" version of PA_PCA (PAP_CA-FS) and EGA_LV (EGA_LV-FS). Additionally, we examined the direct multilevel solution provided by EGA_LV. All the methods were evaluated in an extensive simulation manipulating nine variables of interest, including population error. The results indicated that EGA_LV and PA_PCA displayed the best overall performance in retrieving the true number of group factors, the former being more sensitive to high cross-loadings, and the latter to weak group factors and small samples. Regarding the estimation of the number of general factors, both PAP_CA-FS and EGA_LV-FS showed a close to perfect accuracy across all the conditions, while EGA_LV was inaccurate. The methods based on EGA were robust to the conditions most likely to be encountered in practice. Therefore, we highlight the particular usefulness of EGA_LV (group factors) and EGA_LV-FS (general factors) for assessing bifactor structures with multiple general factors. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Decades of published methodological research have shown the chi-square test of model fit performs inconsistently and unreliably as a determinant of structural equation model (SEM) fit. Likewise, SEM indices of model fit, such as comparative fit index (CFI) and root-mean-square error of approximation (RMSEA) also perform inconsistently and unreliably. Despite rather unreliable ways to statistically assess model fit, researchers commonly rely on these methods for lack of a suitable inferential alternative. Marcoulides and Yuan (2017) have proposed the first inferential test of SEM fit in many years: an equivalence test adaptation of the RMSEA and CFI indices (i.e., RMSEA_t and CFI_t). However, the ability of this equivalence testing approach to accurately judge acceptable and unacceptable model fit has not been empirically tested. This fully crossed Monte Carlo simulation evaluated the accuracy of equivalence testing combining many of the same independent variable (IV) conditions used in previous fit index simulation studies, including sample size (N = 100-1,000), model specification (correctly specified or misspecified), model type (confirmatory factor analysis [CFA], path analysis, or SEM), number of variables analyzed (low or high), data distribution (normal or skewed), and missing data (none, 10%, or 25%). Results show equivalence testing performs rather inconsistently and unreliably across IV conditions, with acceptable or unacceptable RMSEA_t and CFIt model fit index values often being contingent on complex interactions among conditions. Proportional z-tests and logistic regression analyses indicated that equivalence tests of model fit are problematic under multiple conditions, especially those where models are mildly misspecified. Recommendations for researchers are offered, but with the provision that they be used with caution until more research and development is available. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Omega squared (ω^2) is a measure of effect size for analysis of variance (ANOVA) designs. It is less biased than eta squared, but reported less often. This is in part due to lack of clear guidance on how to calculate it. In this paper, we discuss the logic behind effect size measures, the problem with eta squared, the history of omega squared, and why it has been underused. We then provide a user-friendly guide to omega squared and partial omega squared for ANOVA designs with fixed factors, including one-way, two-way, and three-way designs, using within-subjects factors and/or between-subjects factors. We show how to calculate omega squared using output from SPSS. We provide information on the calculation of confidence intervals. We examine the problems of nonadditivity, and intrinsic versus extrinsic factors. We argue that statistical package developers could play an important role in making the calculation of omega squared easier. Finally, we recommend that researchers report the formulas used in calculating effect sizes, include confidence intervals if possible, and include ANOVA tables in the online supplemental materials of their work. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Meta-analyses in the psychological sciences typically examine moderators that may explain heterogeneity in effect sizes. One of the most commonly examined moderators is gender. Overall, tests of gender as a moderator are rarely significant, which may be because effects rarely differ substantially between men and women. While this may be true in some cases, we also suggest that the lack of significant findings may be attributable to the way in which gender is examined as a meta-analytic moderator, such that detecting moderating effects is very unlikely even when such effects are substantial in magnitude. More specifically, we suggest that lack of between-primary study variance in gender composition makes it exceedingly difficult to detect moderation. That is, because primary studies tend to have similar male-to-female ratios, there is very little variance in gender composition between primaries, making it nearly impossible to detect between-study differences in the relationship of interest as a function of gender. In the present article, we report results from two studies: (a) a meta-meta-analysis in which we demonstrate the magnitude of this problem by computing the between-study variance in gender composition across 286 meta-analytic moderation tests from 50 meta-analyses, and (b) a Monte Carlo simulation study in which we show that this lack of variance results in near-zero moderator effects even when male-female differences in correlations are quite large. Our simulations are also used to show the value of single-gender studies for detecting moderating effects. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

Given recent evidence challenging the replicability of results in the social and behavioral sciences, critical questions have been raised about appropriate measures for determining replication success in comparing effect estimates across studies. At issue is the fact that conclusions about replication success often depend on the measure used for evaluating correspondence in results. Despite the importance of choosing an appropriate measure, there is still no widespread agreement about which measures should be used. This article addresses these questions by describing formally the most commonly used measures for assessing replication success, and by comparing their performance in different contexts according to their replication probabilities-that is, the probability of obtaining replication success given study-specific settings. The measures may be characterized broadly as conclusion-based approaches, which assess the congruence of two independent studies' conclusions about the presence of an effect, and distance-based approaches, which test for a significant difference or equivalence of two effect estimates. We also introduce a new measure for assessing replication success called the correspondence test, which combines a difference and equivalence test in the same framework. To help researchers plan prospective replication efforts, we provide closed formulas for power calculations that can be used to determine the minimum detectable effect size (and thus, sample sizes) for each study so that a predetermined minimum replication probability can be achieved. Finally, we use a replication data set from the Open Science Collaboration (2015) to demonstrate the extent to which conclusions about replication success depend on the correspondence measure selected. (PsycInfo Database Record (c) 2025 APA, all rights reserved).