Multivariate Behavioral Research最新文献

Ambulatory assessment has gained widespread popularity among researchers who study the dynamics of everyday experiences and behaviors, such as sleep patterns or emotional states. In this paper, we focus on the challenge that arises when we want to investigate the dynamic relations between variables measured at different frequencies. As a running example, we use a sleep quality variable measured once every morning and a momentary experience variable measured multiple times during the day for multiple days. We propose two N = 1 models that imply different processes; the first focuses on dynamic relations from day to day between sleep quality and a daily factor of the momentary experience variable, and the second focuses on dynamic relations from one measurement occasion to the next, which prioritizes when the variables affect each other. Additionally, we introduce a combination of these two models and demonstrate their accuracy with a simulation study. An empirical N = 1 example of daily sleep quality and momentary self-doubt demonstrates that dynamic relations exist between sleep quality and self-doubt at certain moments in the day and the daily factor of self-doubt. Researchers may adjust the proposed dynamic models to align with their own theories and to accommodate different data or research interests.

Growth mixture models (GMMs) are popular approaches for modeling unobserved population heterogeneity over time. GMMs can be extended with covariates, predicting latent class (LC) membership, the within-class growth trajectories, or both. However, current estimators are sensitive to misspecifications in complex models. We propose extending the two-step estimator for LC models to GMMs, which provides robust estimation against model misspecifications (namely, ignored and overfitted the direct effects) for simpler LC models. We conducted several simulation studies, comparing the performance of the proposed two-step estimator to the commonly-used one- and three-step estimators. Three different population models were considered, including covariates that predicted only the LC membership (I), adding direct effects to the latent intercept (II), or to both growth factors (III). Results show that when predicting LC membership alone, all three estimators are unbiased when the measurement model is strong, with weak measurement model results being more nuanced. Alternatively, when including covariate effects on the growth factors, the two-step, and three-step estimators show consistent robustness against misspecifications with unbiased estimates across simulation conditions while tending to underestimate the standard error estimates while the one-step estimator is most sensitive to misspecifications.

Single-case experimental designs (SCEDs) involve repeated measurements of a small number of cases under different experimental conditions, offering valuable insights into treatment effects. However, challenges arise in the analysis of SCEDs when autocorrelation is present in the data. Recently, generalized linear mixed models (GLMMs) have emerged as a promising statistical approach for SCEDs with count outcomes. While prior research has demonstrated the effectiveness of GLMMs, these studies have typically assumed error independence, an assumption that may be violated in SCEDs due to serial dependency. This study aims to evaluate two possible solutions for autocorrelated SCED count data: 1) to assess the robustness of previously introduced GLMMs such as Poisson, negative binomial, and observation-level random effects models under various levels of autocorrelation, and 2) to evaluate the performance of a new GLMM and a linear mixed model (LMM), both of which incorporate an autoregressive error structure. Through a Monte Carlo simulation study, we have examined bias, coverage rates, and Type I error rates of treatment effect estimators, providing recommendations for handling autocorrelation in the analysis of SCED count data. A demonstration with real SCED count data is provided. The implications, limitations, and future research directions are also discussed.

In this study, we extend the dynamic fit index (DFI) developed by McNeish and Wolf to the context of time series analysis. DFI is a simulation-based method for deriving fit index cutoff values tailored to the specific model and data characteristics. Through simulations, we show that DFI cutoffs for detecting an omitted path in time series network models tend to be closer to exact fit than the popular benchmark values developed by Hu and Bentler. Moreover, cutoff values vary by number of variables, network density, number of time points, and form of misspecification. Notably, using 10% as the upper limit of Type I and Type II error rates, the original DFI approach fails to identify cutoffs for detecting an omitted path when effect size and/or sample size is small. To address this problem, we propose two alternatives that allow for the derivation of cutoffs using more lenient criteria. DFI_A extends the original DFI approach by removing the upper limit of Type I and Type II error rates, whereas DFI_B aims at maximizing classification quality measured by the Matthews correlation coefficient. We demonstrate the utility of these approaches using simulation and empirical data and discuss their implications in practice.

It is a well-known fact that for the bivariate normal distribution the ratio between the point-polyserial correlation (the linear correlation after one of the two variables is discretized into k categories with probabilities $p_i\text{,}$ $i=1,\dots ,k$) and the polyserial correlation $\rho$ (the linear correlation between the two normal components) remains constant with $\rho \text{,}$ keeping the $p_i$'s fixed. If we move away from the bivariate normal distribution, by considering non-normal margins and/or non-normal dependence structures, then the constancy of this ratio may get lost. In this work, the magnitude of the departure from the constancy condition is assessed for several combinations of margins (normal, uniform, exponential, Weibull) and copulas (Gauss, Frank, Gumbel, Clayton), also varying the distribution of the discretized variable. The results indicate that for many settings we are far from the condition of constancy, especially when highly asymmetrical marginal distributions are combined with copulas that allow for tail-dependence. In such cases, the linear correlation may even increase instead of decreasing, contrary to the usual expectation. This implies that most existing simulation techniques or statistical models for mixed-type data, which assume a linear relationship between point-polyserial and polyserial correlations, should be used very prudently and possibly reappraised.

Numerous studies have shown that motor inhibition can be triggered automatically when the cognitive system encounters interfering stimuli, even a suspicious stimulus in the absence of perceptual awareness (e.g., the negative compatibility effect). This study investigated the effect of temporal expectation, a top-down active preparation for future events, on unconscious inhibitory processing both in the local expectation context on a trial-by-trial basis (Experiment 1) and in the global expectation context on a block-wise basis (Experiment 2). Modeling of the behavioral data using a drift-diffusion model showed that temporal expectation can accelerate the evidence accumulation and improve response caution, regardless of context. Importantly, the acceleration is lower when the target is consistent with the suspicious response tendency induced by the subliminal prime than when the target is inconsistent with that, which is significantly correlated with the behavioral RTs (i.e., the compatibility effect). The results provide evidence for a framework in which temporal expectation enhances inhibitory control of unconscious processes. The mechanism is likely to be that temporal expectation enhances the activations afforded by subliminal stimuli and the strength of cognitive monitoring, so that the cognitive system suppresses these suspicious activations more strongly, preventing them from escaping and interfering with subsequent processing.

Multilevel latent class profile analysis (MLCPA) is a recently developed technique for understanding latent class dynamics in longitudinal studies; however, conventional maximum likelihood (ML) estimation may face challenges, particularly with small sample sizes or boundary solutions. As an alternative method, we propose a Bayesian estimation for MLCPA by employing non-informative prior distributions. In addition, we shed light on the underflow problem, which denotes a phenomenon such that the logarithm of the likelihood is negative infinity due to the multilevel structure. We perform extensive numerical studies to compare the behaviors of the MLE and the Bayesian estimates and investigate the accuracies of approximated model selection criteria. The simulation study revealed that Bayesian estimates are preferred to ML estimates when the underlying latent classes are well-separated, while the ML estimates are preferred when the underlying latent classes overlap. Utilizing the Progress Monitoring and Reporting Network data, which includes longitudinal academic performance metrics, our analysis uncovers distinct pathways of latent classes for students, further differentiated by latent groups of schools. These findings shed light on the considerable variations in academic proficiency trajectories and thus may offer new perspectives on academic proficiency patterns, with important implications for policy development and targeted educational interventions.

The interrater reliability (IRR) of observational data is often estimated by means of intraclass correlation coefficients (ICCs), which are flexible IRR estimators that are based on the variance decomposition of scores obtained by observations. ICCs are typically estimated using mean squares from an ANOVA model, the computation of which is not straightforward for incomplete data. However, many studies in behavioral research use planned missing observational designs, in which the raters partially vary across subjects. Planned missing designs result in incomplete data. Therefore, we simulated planned incomplete data and compared the computational accuracy (bias of point estimates, bias of variability estimates, root mean squared error, and coverage rates) and computational feasibility (convergence rates and estimation time) of three recently proposed estimation methods for ICCs: Markov chain Monte Carlo estimation of Bayesian hierarchical linear models, maximum likelihood estimation of random-effects models, and maximum likelihood estimation of common-factor models. Maximum likelihood estimation of random-effects models with Monte-Carlo confidence intervals is preferred based on all criteria. This article is accompanied by R code, which enables researchers to apply these estimation methods. A demonstration of the R code to a real-data set from an educational context is provided.

Intensive longitudinal data with a large number of timepoints per individual are becoming increasingly common. Such data allow going beyond the classical growth model situation and studying population effects and individual variability not only in trends over time but also in autoregressive effects, cross-lagged effects, and the noise term. Dynamic structural equation models (DSEMs) have become very popular for analyzing intensive longitudinal data. However, when the data contain trends, cycles, or time-varying predictors which have nonlinear effects on the outcome, DSEMs require the practitioner to specify the correct parametric form of the effects, which may be challenging in practice. In this paper, we show how to alleviate this issue by introducing regression splines which are able to flexibly learn the underlying function shapes. Our main contribution is thus a building block to the DSEM modeler's toolkit, and we discuss smoothing priors and hierarchical smooth terms using the special cases of two-level lag-1 autoregressive and vector autoregressive models as examples. We illustrate in simulation studies how ignoring nonlinear trends may lead to biased parameter estimates, and then show how to use the proposed framework to model weekly cycles and long-term trends in diary data on alcohol consumption and perceived stress.

Variable selection in structural equation modeling has merged as a new concern in social and psychological studies. Researchers often aim to strike a balance between achieving predictive accuracy and fostering parsimonious explanations by identifying the most informative variables. While recent developments in Bayesian regularization methods offer promising solutions to promote model sparsity with much fewer "active" variables, their computational burden due to reliance on the Markov chain Monte Carlo technique limits practical utility. In response, this study proposes a variational Bayesian expectation-maximum algorithm (VBEM) for variable selection to extend the multiple-indicators multiple-causes (MIMIC) model. On the basis of traditional MIMIC models, a partially confirmatory framework that operates within the exploratory-confirmatory continuum is introduced, allowing for the flexible incorporation of substantive knowledge and regularization into both measurement and structural parts while accounting for factor correlation. The proposed method demonstrated its flexibility, reliability, and efficiency on both simulated and real data.