Two-part mixed-effects location scale models

Alternative covariance structures in mixed-effects models: Addressing intra- and inter-individual heterogeneity

Mixed-effects models for repeated measures and longitudinal data include random coefficients that are unique to the individual, and thus permit subject-specific growth trajectories, as well as direct study of how the coefficients of a growth function vary as a function of covariates. Although applications of these models often assume homogeneity of the within-subject residual variance that characterizes within-person variation after accounting for systematic change and the variances of the random coefficients of a growth model that quantify individual differences in aspects of change, alternative covariance structures can be considered. These include allowing for serial correlations between the within-subject residuals to account for dependencies in data that remain after fitting a particular growth model or specifying the within-subject residual variance to be a function of covariates or a random subject effect to address between-subject heterogeneity due to unmeasured influences. Further, the variances of the random coefficients can be functions of covariates to relax the assumption that these variances are constant across subjects and to allow for the study of determinants of these sources of variation. In this paper, we consider combinations of these structures that permit flexibility in how mixed-effects models are specified to understand within- and between-subject variation in repeated measures and longitudinal data. Data from three learning studies are analyzed using these different specifications of mixed-effects models.

Detecting influential subjects in intensive longitudinal data using mixed-effects location scale models

BACKGROUND

Collection of intensive longitudinal health outcomes allows joint modeling of their mean (location) and variability (scale). Focusing on the location of the outcome, measures to detect influential subjects in longitudinal data using standard mixed-effects regression models (MRMs) have been widely discussed. However, no existing approach enables the detection of subjects that heavily influence the scale of the outcome.

METHODS

We propose applying mixed-effects location scale (MELS) modeling combined with commonly used influence measures such as Cook's distance and DFBETAS to fill this gap. In this paper, we provide a framework for researchers to follow when trying to detect influential subjects for both the scale and location of the outcome. The framework allows detailed examination of each subject's influence on model fit as well as point estimates and precision of coefficients in different components of a MELS model.

RESULTS

We simulated two common scenarios in longitudinal healthcare studies and found that influence measures in our framework successfully capture influential subjects over 99% of the time. We also re-analyzed data from a health behavior study and found 4 particularly influential subjects, among which two cannot be detected by influence analyses via regular MRMs.

CONCLUSION

The proposed framework can help researchers detect influential subject(s) that will be otherwise overlooked by influential analysis using regular MRMs and analyze all data in one model despite influential subjects.

Lifetime and Current Self-Harm Thoughts and Behaviors and Their Relationship to Parent and Peer Attachment

Background: Previous research suggests attachment is a vulnerability factor for self-harm thoughts and behaviors in adults. Yet, few studies have investigated this relationship during adolescence, although adolescence is a critical period for changes in attachment relationships and self-harm onset. Whether and how attachment relates to self-harm thoughts and behaviors as measured in daily life is also unknown. Aims: To investigate whether and how paternal, maternal, and peer attachment are associated with lifetime and current adolescent self-harm thoughts and behaviors. Additionally, to examine how different attachment bonds interact in relation to lifetime and current adolescent self-harm thoughts and behaviors. Method: Pre-existing data from N = 1,913 adolescents of the SIGMA study were used. Attachment and lifetime history of self-harm thoughts and behaviors were measured via retrospective questionnaires. Current self-harm thoughts and behaviors were assessed 10 times per day for 6 days using the experience sampling method (ESM). Results: Paternal and maternal attachments were associated with lifetime self-harm thoughts and behaviors and current self-harm thoughts. No significant associations were found between peer attachment and self-harm outcomes. Limitations: Some analyses were underpowered. Conclusion: Our results highlight the importance of parent-child attachment relationships, which may be intervention targets for prevention and treatment of adolescent self-harm.

Predictors of individual differences in minimally verbal peer communication exchanges following peer-oriented social intervention

School-age children on the autism spectrum who are minimally verbal (MVAS) use a limited repertoire of spontaneous communicative spoken words and reveal large heterogeneity in cognitive functioning. Despite the challenges to form peer social engagement posed by their unique social-communicative profile, few interventions have targeted peer interactions in the MVAS population. This study explored predictors of individual differences in treatment response among 54 school-age minimally verbal autistic children (8-16 years) following an RCT "school-based peer social intervention" (S-PSI) that compared two peer-oriented intervention modalities (conversation versus collaboration) versus a waitlisted control group. We examined autistic-symptom severity, age, verbal and nonverbal IQ, executive functions, and sensory-processing profile for their contribution to children's ability to form relevant spontaneous communication exchanges with a peer partner. Main findings revealed that larger deficits in sensory-processing (sensory-avoidance and sensory low-registration) and in executive functions contributed to greater growth in "relevant" (i.e., adequately attuned, participatory, reciprocal) communication following both interventions, but not for the waitlisted controls. Additionally, older participants with lower verbal and nonverbal IQ improved communication's relevancy more after the conversation intervention, whereas the collaboration intervention was more beneficial for younger participants. Lower autistic-symptom severity contributed to larger growth in relevancy for all groups. By identifying individual-level predictors of spontaneous, attuned, participatory, and reciprocal (i.e., "relevant") communication exchanges with peers, we optimized S-PSI personalization for this uniquely nonverbal, heterogeneous MVAS population. These new channels for tailoring peer interventions to better meet individuals' needs may result in reduced social isolation and loneliness and enhanced well-being.

Variability in negative affect is an important feature of neuroticism above mean negative affect once measurement issues are accounted for

Neuroticism is an important predictor of well-being that is conceptualized by high levels of mean negative affect and negative affect variability. However, research has shown that negative affect variability only explained limited additional variance in neuroticism when the confound with mean negative affect was accounted for using a modified version of the standard deviation ( SD), the relative standard deviation ( RSD). Here, we (a) examined the suitability of the RSD as a variability measure, (b) introduced the number of negative affect episodes as an alternative measure of negative affect variability, and (c) investigated the relationship between neuroticism and negative affect variability, accounting for measurement error. Re-analyzing three experience sampling datasets ( N = 430 participants), we found several issues with the RSD, which limits its use as a negative affect variability measure, and which were not found for the number of negative affect episodes measure. Moreover, only the SD and the number of negative affect episodes explained substantial variance in neuroticism above mean negative affect. Thus, neuroticism was associated with experiencing negative affect more strongly and more often in daily life, when measurement error was accounted for, which demonstrates the importance to model reliability and to correct accordingly.

Catching Up on Multilevel Modeling

This review focuses on the use of multilevel models in psychology and other social sciences. We target readers who are catching up on current best practices and sources of controversy in the specification of multilevel models. We first describe common use cases for clustered, longitudinal, and cross-classified designs, as well as their combinations. Using examples from both clustered and longitudinal designs, we then address issues of centering for observed predictor variables: its use in creating interpretable fixed and random effects of predictors, its relationship to endogeneity problems (correlations between predictors and model error terms), and its translation into multivariate multilevel models (using latent-centering within multilevel structural equation models). Finally, we describe novel extensions—mixed-effects location–scale models—designed for predicting differential amounts of variability.