MultiLevel simultaneous component analysis: A computational shortcut and software package

Examination of differential ratings of perceived exertion (dRPE) during bio-banded small-sided games

The aims of the current study were to investigate the use of dRPE with academy soccer players to: 1) examine the effect of bio-banded and non-bio-banded maturity groups within SSG on players dRPE; 2) describe the multivariate relationships between dRPE measures investigating the sources of intra and inter-individual variation, and the effects of maturation and bio-banding. Using 32 highly trained under (U) 12 to U14 soccer players (mean (SD) age 12.9 (0.9) years, body mass 46.4 (8.5) kg and stature 158.2 (14.9) cm) academy soccer players from two English professional male soccer academies. Players were categorised according to somatic maturity status using estimated percentage of adult stature attainment, with players randomly assigned into teams to play 4v4 SSG. The study used a repeated measures design, whereby the selected players participated within 6 bio-banded (maturity matched [pre-PHV Vs pre-PHV and post-PHV vs post PHV] and miss-matched [pre-PHV vs post-PHV] and 6 mixed maturity SSG at their respective clubs. Using mixed and fixed effect regression models, it was established hat pre-PHV players exhibited higher dRPE compared with their post-PHV counterparts. Mixed bio-banded games reported higher dRPE outputs overall. Variation in dRPE measures across a series of bio-banded games are caused by both between and within sources of variation in relatively equal amounts. Across a series of bio-banded games, the four dRPE measures do not provide unique information, and between variation is best expressed by one or two highly correlated components, with within variation best explained by a single equally loaded component. Using a bio-banding SSG design study, we have shown that pre-PHV players report higher subjective measures of exertion than post-PHV players during. Additionally, when evenly mixing players based on measures of maturation, higher measures of perceived exertion were generally reported.

A Systematic Study into the Factors that Affect the Predictive Accuracy of Multilevel VAR(1) Models

The use of multilevel VAR(1) models to unravel within-individual process dynamics is gaining momentum in psychological research. These models accommodate the structure of intensive longitudinal datasets in which repeated measurements are nested within individuals. They estimate within-individual auto- and cross-regressive relationships while incorporating and using information about the distributions of these effects across individuals. An important quality feature of the obtained estimates pertains to how well they generalize to unseen data. Bulteel and colleagues (Psychol Methods 23(4):740-756, 2018a) showed that this feature can be assessed through a cross-validation approach, yielding a predictive accuracy measure. In this article, we follow up on their results, by performing three simulation studies that allow to systematically study five factors that likely affect the predictive accuracy of multilevel VAR(1) models: (i) the number of measurement occasions per person, (ii) the number of persons, (iii) the number of variables, (iv) the contemporaneous collinearity between the variables, and (v) the distributional shape of the individual differences in the VAR(1) parameters (i.e., normal versus multimodal distributions). Simulation results show that pooling information across individuals and using multilevel techniques prevent overfitting. Also, we show that when variables are expected to show strong contemporaneous correlations, performing multilevel VAR(1) in a reduced variable space can be useful. Furthermore, results reveal that multilevel VAR(1) models with random effects have a better predictive performance than person-specific VAR(1) models when the sample includes groups of individuals that share similar dynamics.

The role of social connection on the experience of COVID-19 related post-traumatic growth and stress

BACKGROUND

Historically social connection has been an important way through which humans have coped with large-scale threatening events. In the context of the COVID-19 pandemic, lockdowns have deprived people of major sources of social support and coping, with others representing threats. Hence, a major stressor during the pandemic has been a sense of social disconnection and loneliness. This study explores how people's experience of compassion and feeling socially safe and connected, in contrast to feeling socially disconnected, lonely and fearful of compassion, effects the impact of perceived threat of COVID-19 on post-traumatic growth and post-traumatic stress.

METHODS

Adult participants from the general population (N = 4057) across 21 countries worldwide, completed self-report measures of social connection (compassion for self, from others, for others; social safeness), social disconnection (fears of compassion for self, from others, for others; loneliness), perceived threat of COVID-19, post-traumatic growth and traumatic stress.

RESULTS

Perceived threat of COVID-19 predicted increased post-traumatic growth and traumatic stress. Social connection (compassion and social safeness) predicted higher post-traumatic growth and traumatic stress, whereas social disconnection (fears of compassion and loneliness) predicted increased traumatic symptoms only. Social connection heightened the impact of perceived threat of COVID-19 on post-traumatic growth, while social disconnection weakened this impact. Social disconnection magnified the impact of the perceived threat of COVID-19 on traumatic stress. These effects were consistent across all countries.

CONCLUSIONS

Social connection is key to how people adapt and cope with the worldwide COVID-19 crisis and may facilitate post-traumatic growth in the context of the threat experienced during the pandemic. In contrast, social disconnection increases vulnerability to develop post-traumatic stress in this threatening context. Public health and Government organizations could implement interventions to foster compassion and feelings of social safeness and reduce experiences of social disconnection, thus promoting growth, resilience and mental wellbeing during and following the pandemic.

State Attachment Variability: Between- and within-Person Level Associations with Trait Attachment and Psychological Problems

Research suggests that inter-individual differences in the degree of state attachment variability are related to differences in trait attachment and psychological problems between children. In this study, we tested whether such associations are also relevant at a within-person level, and if so, whether intra-individual fluctuations in the degree of variability were predictive of or predicted by intra-individual fluctuations in trait attachment and psychological problems. Children (N = 152; M_age = 10.41 years, SD_age = 0.60 at time 1) were tested three times over a period of one year. At each timepoint, children reported on their expectations of maternal support in different distressing situations. Additionally, we administered measures of trait attachment to children and psychological problems to children and their mothers. We used Random-Intercept Cross-Lagged Panel Models to distinguish between-person from within-person associations between these constructs over time. The results revealed that the degree of state attachment variability was mainly relevant to understand differences between children in trait attachment and psychological problems: children who overall showed more state attachment variability were overall less securely attached at a trait-level and reported more psychological problems. Although evidence for within-person associations was less robust, there was some indication that the degree of state attachment variability might be related to the development of trust and psychological problems at a within-person level.

Similarity-driven multi-view embeddings from high-dimensional biomedical data

Diverse, high-dimensional modalities collected in large cohorts present new opportunities for the formulation and testing of integrative scientific hypotheses. Similarity-driven multi-view linear reconstruction (SiMLR) is an algorithm that exploits inter-modality relationships to transform large scientific datasets into smaller, more well-powered and interpretable low-dimensional spaces. SiMLR contributes an objective function for identifying joint signal, regularization based on sparse matrices representing prior within-modality relationships and an implementation that permits application to joint reduction of large data matrices. We demonstrate that SiMLR outperforms closely related methods on supervised learning problems in simulation data, a multi-omics cancer survival prediction dataset and multiple modality neuroimaging datasets. Taken together, this collection of results shows that SiMLR may be applied to joint signal estimation from disparate modalities and may yield practically useful results in a variety of application domains.

Multilevel LASSO-based NIR temperature-correction modeling for viscosity measurement of bisphenol-A

Temperature variation will affect the prediction accuracy when using the near-infrared (NIR) analytical model to measure the viscosity of bisphenol-A. In order to correct the effect of temperature on the prediction performance of NIR model, a multilevel least-absolute shrinkage and selection operator (LASSO) is proposed in this paper. The multilevel LASSO algorithm combines LASSO with the multilevel simultaneous component analysis (MLSCA) to enhance the robustness of the model under temperature changes and external disturbances. MLSCA is applied to decompose the molecular spectral data into two parts. One part denotes the property caused by temperature, the other means the changes of concentration. LASSO, a sparse regression model, is used to select the variables and perform the regularization to further enhance the robustness and interpretability of the model. Experimental results demonstrate the effectiveness of the proposed model in measuring bisphenol-A viscosity, which provides a more stable prediction result compared with the existing ones without temperature corrections.

Detecting which variables alter component interpretation across multiple groups: A resampling-based method

In psychology, many studies measure the same variables in different groups. In the case of a large number of variables when a strong a priori idea about the underlying latent construct is lacking, researchers often start by reducing the variables to a few principal components in an exploratory way. Herewith, one often wants to evaluate whether the components represent the same construct in the different groups. To this end, it makes sense to remove outlying variables that have significantly different loadings on the extracted components across the groups, hampering equivalent interpretations of the components. Moreover, identifying such outlying variables is important when testing theories about which variables behave similarly or differently across groups. In this article, we first scrutinize the lower bound congruence method (LBCM; De Roover, Timmerman, & Ceulemans in Behavior Research Methods, 49, 216-229, 2017), which was recently proposed for solving the outlying-variable detection problem. LBCM investigates how Tucker's congruence between the loadings of the obtained cluster-loading matrices improves when specific variables are discarded. We show that LBCM has the tendency to output outlying variables that either are false positives or concern very small, and thus practically insignificant, loading differences. To address this issue, we present a new heuristic: the lower and resampled upper bound congruence method (LRUBCM). This method uses a resampling technique to obtain a sampling distribution for the congruence coefficient, under the hypothesis that no outlying variable is present. In a simulation study, we show that LRUBCM outperforms LBCM. Finally, we illustrate the use of the method by means of empirical data.

Accounting for Differential Item Functioning Using Bayesian Approximate Measurement Invariance

The main thesis of the present study is to use the Bayesian structural equation modeling (BSEM) methodology of establishing approximate measurement invariance (A-MI) using data from a national examination in Saudi Arabia as an alternative to not meeting strong invariance criteria. Instead, we illustrate how to account for the absence of measurement invariance using relative compared to exact criteria. A secondary goal was to compare latent means across groups using invariant parameters only and through utilizing exact and relative evaluative-MI protocol suggested equivalence of the thresholds using prior variances equal to 0.10. Subsequent differences between groups were evaluated using effect size criteria and the prior-posterior predictive p-value (PPPP), which proved to be invaluable in attesting for differences that are beyond zero, some meaningless nonzero estimate, and the three commonly used indices of effect sizes described by Cohen in 1988 (i.e., .20, .50, and .80). Results substantiated the use of the PPPP for evaluating mean differences across groups when utilizing nonexact evaluative criteria.

Improved Insight into and Prediction of Network Dynamics by Combining VAR and Dimension Reduction

To understand within-person psychological processes, one may fit VAR(1) models (or continuous-time variants thereof) to multivariate time series and display the VAR(1) coefficients as a network. This approach has two major problems. First, the contemporaneous correlations between the variables will frequently be substantial, yielding multicollinearity issues. In addition, the shared effects of the variables are not included in the network. Consequently, VAR(1) networks can be hard to interpret. Second, crossvalidation results show that the highly parametrized VAR(1) model is prone to overfitting. In this article, we compare the pros and cons of two potential solutions to both problems. The first is to impose a lasso penalty on the VAR(1) coefficients, setting some of them to zero. The second, which has not yet been pursued in psychological network analysis, uses principal component VAR(1) (termed PC-VAR(1)). In this approach, the variables are first reduced to a few principal components, which are rotated toward simple structure; then VAR(1) analysis (or a continuous-time analog) is applied to the rotated components. Reanalyzing the data of a single participant of the COGITO study, we show that PC-VAR(1) has the better predictive performance and that networks based on PC-VAR(1) clearly represent both the lagged and the contemporaneous variable relations.