When analyzing data, researchers are often confronted with a model selection problem (e.g., determining the number of components/factors in principal components analysis PCA/factor analysis or ...identifying the most important predictors in a regression analysis). To tackle such a problem, researchers may apply some objective procedure, like parallel analysis in PCA/factor analysis or stepwise selection methods in regression analysis. A drawback of these procedures is that they can only be applied to the model selection problem at hand. An interesting alternative is the CHull model selection procedure, which was originally developed for multiway analysis (e.g., multimode partitioning). However, the key idea behind the CHull procedure—identifying a model that optimally balances model goodness of fit/misfit and model complexity—is quite generic. Therefore, the procedure may also be used when applying many other analysis techniques. The aim of this article is twofold. First, we demonstrate the wide applicability of the CHull method by showing how it can be used to solve various model selection problems in the context of PCA, reduced
K
-means, best-subset regression, and partial least squares regression. Moreover, a comparison of CHull with standard model selection methods for these problems is performed. Second, we present the CHULL software, which may be downloaded from
http://ppw.kuleuven.be/okp/software/CHULL/
, to assist the user in applying the CHull procedure.
In consumer studies, segmentation has been widely applied to identify consumer subsets on the basis of their preference for a set of products. From the last decade onwards, a more comprehensive ...evaluation of product performance has led to take into account various information such as consumer emotion assessment or hedonic measures on several aspects, like taste, visual and flavor. This multi-attribute evaluation of products naturally yields a three-way (products by consumers by attributes) data structure. In order to identify segments of consumers on the basis of such three-way data, the Three-Way Cluster analysis around Latent Variables (CLV3W) approach (Wilderjans & Cariou, 2016) is considered. This method groups the consumers into clusters and estimates for each cluster an associated latent product variable and attribute weights, along with a set of consumer loadings, which may be used for the purpose of cluster-specific product characterization. As consumers who rate the products along the attributes in an opposite way (i.e., raters’ disagreement) should not be in the same cluster, in this paper, we propose to add a non-negativity constraint on the consumer loadings and to integrate this constraint within the versatile CLV3W approach. This non-negatively constrained criterion implies that the latent variable for each cluster is determined such that consumers within each cluster are as much related – in terms of a positive covariance – as possible with this latent product component. This approach is applied to a consumer emotion ratings dataset related to coffee aromas.
People read for many different reasons. These goals affect the cognitive processes and strategies they use during reading. Understanding how reading goals exert their effects requires investigation ...of whether and how they affect specific component processes, such as validation. We investigated the effects of reading goal on text-based and knowledge-based validation processes during reading and on the resulting offline mental representation. We employed a self-paced sentence-by-sentence contradiction paradigm with versions of texts containing target sentences that varied systematically in congruency with prior text and accuracy with background knowledge. Participants were instructed to read for general comprehension or for study. Memory for text information was assessed the next day. We also measured the degree to which each text topic was novel to a reader, as well as his or her working memory capacity. Results show that reading goals affect readers' general processing as indicated by overall reading times, but provide no evidence that they influence validation processes. Reading goals did affect readers' memory for target information but this effect depended on congruency between that information and the preceding text: Reading for study generally resulted in better memory for target information than reading for comprehension did, but not for target information that was incongruent with prior text. These results suggest that reading goals may not influence validation processes directly but affect subsequent representation-building processes after the detection of an (in)consistency-particularly in the case of incongruencies with prior text.
Educational Impact and Implications StatementThis study advances our understanding of how readers' purpose for reading impacts how they process texts containing information that contradicts what they know (i.e., their background knowledge) and/or what they just read (i.e., the preceding text) and what they remember from those texts.
Dual-pathway models suggest that poor self-regulation (immature regulatory combined with strong reactive processes) is an important factor underlying addictive behaviors among adolescents. This study ...examined whether there are different self-regulation profiles among community adolescents, and how these profiles are related to the presence, severity and comorbidity of different addictive behaviors. A community sample of 341 adolescents (54.5% female; 13–17 years) was recruited. Participants self-reported on regulatory (inhibitory control) and reactive (reward and punishment sensitivity) processes, as well as on different addictive behaviors (binge eating, tobacco-, cannabis- and alcohol use, gaming, gambling and pathological buying). A model-based clustering analysis found evidence for three meaningful profiles: ‘impulsive/under-controlled’, ‘anxious’ and ‘protective’. The ‘impulsive/under-controlled’ profile was characterized by the highest prevalence and severity of cannabis use and the most severe alcohol use. The ‘impulsive/under-controlled’ and ‘protective’ profiles demonstrated the highest prevalence and severity of tobacco use, whereas the ‘impulsive/under-controlled’ and ‘anxious’ profiles showed the highest binge eating scores. Adolescents who reported more than three types of addictive behaviors generally belonged to the ‘impulsive/under-controlled’ profile. The profiles did not differ for gaming, gambling and pathological buying. The ‘impulsive/under-controlled’ profile emerged as the most vulnerable profile in the context of addictive behaviors (especially for binge eating and substance use).
Mixture analysis is commonly used for clustering objects on the basis of multivariate data. When the data contain a large number of variables, regular mixture analysis may become problematic, because ...a large number of parameters need to be estimated for each cluster. To tackle this problem, the mixtures-of-factor-analyzers (MFA) model was proposed, which combines clustering with exploratory factor analysis. MFA model selection is rather intricate, as both the number of clusters and the number of underlying factors have to be determined. To this end, the Akaike (AIC) and Bayesian (BIC) information criteria are often used. AIC and BIC try to identify a model that optimally balances model fit and model complexity. In this article, the CHull (Ceulemans & Kiers,
2006
) method, which also balances model fit and complexity, is presented as an interesting alternative model selection strategy for MFA. In an extensive simulation study, the performances of AIC, BIC, and CHull were compared. AIC performs poorly and systematically selects overly complex models, whereas BIC performs slightly better than CHull when considering the best model only. However, when taking model selection uncertainty into account by looking at the first three models retained, CHull outperforms BIC. This especially holds in more complex, and thus more realistic, situations (e.g., more clusters, factors, noise in the data, and overlap among clusters).
Abstract This study investigated the role of arousal and effort costs in the cognitive benefits of alternating between sitting and standing postures using a sit‐stand desk, while measuring executive ...functions, self‐reports, physiology, and neural activity in a 2‐h laboratory session aimed to induce mental fatigue. Two sessions were conducted with a one‐week gap, during which participants alternated between sitting and standing postures each 20‐min block in one session and remained seated in the other. In each block, inhibition, switching, and updating were assessed. We examined effects of time‐on‐task, acute (local) effects of standing versus sitting posture, and cumulative (global) effects of a standing posture that generalize to the subsequent block in which participants sit. Results ( N = 43) confirmed that time‐on‐task increased mental fatigue and decreased arousal. Standing (versus sitting) led to acute increases in arousal levels, including self‐reports, alpha oscillations, and cardiac responses. Standing also decreased physiological and perceived effort costs. Standing enhanced processing speed in the flanker task, attributable to shortened nondecision time and speeded evidence accumulation processes. No significant effects were observed on higher‐level executive functions. Alternating postures also increased heart rate variability cumulatively over time. Exploratory mediation analyses indicated that the positive impact of acute posture on enhanced drift rate was mediated by self‐reported arousal, whereas decreased nondecision time was mediated by reductions in alpha power. In conclusion, alternating between sitting and standing postures can enhance arousal, decrease effort costs, and improve specific cognitive and physiological outcomes.
We studied the impact of standing on executive functions and physiology using a unique multi‐method approach that investigated the effect of alternating between sitting and standing on cardiac physiology, neural oscillations, and executive functions. Standing improved neural and physiological arousal, enhanced overall task performance in a flanker task, while also decreasing physiological and perceived effort costs, and increasing heart rate variability over time. These findings emphasize the benefits of alternating between seated and standing postures for the body and the brain.
•In the scope of sensory profiling, the CLV approach is extended to three-way data.•CLV3W exhibits clusters of sensory descriptors along with their latent dimensions.•A weight is determined for each ...combination of an assessor and a latent dimension.•CLV3W detects poor performing assessors along with the problematic latent dimension.
To detect panel disagreement, we propose the clustering around latent variables for three-way data (CLV3W) approach which extends the clustering of variables around latent components (CLV) approach to three-way data typically obtained from a conventional sensory profiling procedure (i.e., assessors rating products on various descriptors). The CLV3W method groups the descriptors into Q clusters and estimates for each cluster an associated latent sensory component such that the attributes within each cluster are as much related (i.e., highest squared covariance) as possible with the latent component. Simultaneously, for each latent sensory component separately, a system of weights is estimated that yields information regarding the extent to which an assessor (dis)agrees with the rest of the panel according to the latent sensory component under study. Our new approach is illustrated with a dataset pertaining to Quantitative Descriptive Analysis applied to cider varieties. It is shown that CLV3W, as opposed to related approaches, is able to detect differential panel disagreement on various latent sensory components.
The set‐up of comprehensive studies in life sciences involving a longitudinal dimension—as appears in time‐scale metabolomics—calls for the use of dimension reduction techniques for three‐way data ...structures (e.g., samples by variables by time points). For this purpose, a clustering around latent variables for three‐way data approach, CLV3W, has been proposed. CLV3W aims at both partitioning the variables into nonoverlapping clusters and estimating within each cluster a rank‐one Parafac model consisting of a latent component (resp. a weighting system) associated with the first mode (resp. third mode) and a vector of loadings reflecting the degree of closeness of each variable of the second mode to its cluster. In this paper, two constrained CLV3W models are discussed. First, a nonnegativity constraint is defined implying that clusters are composed of positively correlated variables. Second, it is proposed to constrain the weighting system to be the same for all clusters. These two constraints aim at providing more parsimonious models with configurations that are easier to interpret. The appropriateness of both constraints is evaluated in a simulation study and illustrated on two case studies pertaining to sensory evaluation and metabolomics data. Regarding the first case study, CLV3W yields the identification of two consumer segments together with one common emotional pleasantness dimension associated with coffee aromas. CLV3W analysis of human preterm breast milk metabolomics data provided three clusters of lipid species that are responsible for specific functions (i.e., milk fat globules membrane‐constituents, fatty acid oxidation‐products, lipid mediators as eicosanoids and endocannabinoids).
A clustering around latent variables for three‐way data (CLV3W) approach is presented. Constraints on the configuration aim at facilitating the interpretation of the CLV3W solutions. Nonnegativity constraint on loadings requires clusters with positively correlated variables only. Application of CLV3W to time‐scale metabolomics data provides a partitioning into consistent groups of bio‐markers.
Purpose
Few studies have investigated possible predictors of positive outcomes for youths in foster care. The aim of this prospective follow-up study was to examine quality of life (QoL) among youths ...in foster care and to assess whether contextual and child factors predicted QoL.
Methods
Online questionnaires were completed by carers in Norway in 2012 (T1,
n
= 236, child age 6–12 years) and by youths and carers in 2017 (T2,
n
= 405, youth age 11–18 years). We received responses on 116 of the youths at both T1 and T2, and our final sample consisted of 525 youths with responses from T1 and/or T2. Child welfare caseworkers reported preplacement maltreatment and service use at T1. We assessed mental health and prosocial behavior at T1 by having carers complete the Strength and Difficulties Questionnaire and QoL at T2 with youth-reported KIDSCREEN-27. We analyzed the data using descriptive statistics,
t
-tests and multiple linear regressions, and we used multiple imputation to handle missing data.
Results
Youths in foster care had lower QoL across all dimensions compared to a Swedish general youth sample. QoL scores among our sample were similar to Norwegian youths with ill or substance abusing parents and to European norm data. Youths reported the highest QoL scores on the parent relations and autonomy dimension. Male gender, younger age, kinship care and prosocial behavior five years earlier predicted higher QoL.
Conclusion
Similar to other at-risk youths, youths in foster care seem to have lower QoL than the general Scandinavian population. Despite early adversities, they had good relations with their current carers. Adolescent girls seem especially vulnerable to low QoL and might need extra support to have good lives in foster care.
The main purpose of the study was the development of the Sensory Processing Sensitivity Questionnaire (SPSQ), designed to measure Sensory Processing Sensitivity, defined as a person's sensitivity to ...subtle stimuli, the depth with which these stimuli are processed, and its impact on emotional reactivity. The item pool generated for the development of the SPSQ consisted of 60 items. After exploratory factor analysis, 43 items remained, divided into six specific factors: (1) Sensory Sensitivity to Subtle Internal and External Stimuli, (2) Emotional and Physiological Reactivity, (3) Sensory Discomfort, (4) Sensory Comfort, (5) Social-Affective Sensitivity, and (6) Esthetic Sensitivity. Confirmatory factor analysis indicated that a higher-order bi-factor model consisting of two higher-order factors (a positive and negative dimension), a general sensitivity factor and six specific factors had the best fit. Strong positive associations were found between Emotional and Physiological Reactivity, the negative higher-order dimension, and Neuroticism; the same holds for the association between Esthetic Sensitivity, the positive higher-order dimension, and Openness. Emotional and Physiological Reactivity and the negative higher-order dimension showed clear associations with clinical outcomes. The relationships between the SPSQ and similar scales - the Highly Sensitive Person Scale and part of the Adult Temperament Questionnaire - were in the expected direction.