scholarly journals Escaping the snare of chronological growth and launching a free curve alternative: General deviance as latent growth model

2013 ◽  
Vol 25 (3) ◽  
pp. 739-754 ◽  
Author(s):  
Phillip Karl Wood ◽  
Kristina M. Jackson
Keyword(s):  
Problem Behavior ◽  
Problem Behaviors ◽  
Growth Model ◽  
Repeated Measures ◽  
Linear Models ◽  
Model Fit ◽  
Special Cases ◽  
Over Time ◽  
Better Than ◽  
General Deviance

AbstractResearchers studying longitudinal relationships among multiple problem behaviors sometimes characterize autoregressive relationships across constructs as indicating “protective” or “launch” factors or as “developmental snares.” These terms are used to indicate that initial or intermediary states of one problem behavior subsequently inhibit or promote some other problem behavior. Such models are contrasted with models of “general deviance” over time in which all problem behaviors are viewed as indicators of a common linear trajectory. When fit of the “general deviance” model is poor and fit of one or more autoregressive models is good, this is taken as support for the inhibitory or enhancing effect of one construct on another. In this paper, we argue that researchers consider competing models of growth before comparing deviance and time-bound models. Specifically, we propose use of the free curve slope intercept (FCSI) growth model (Meredith & Tisak, 1990) as a general model to typify change in a construct over time. The FCSI model includes, as nested special cases, several statistical models often used for prospective data, such as linear slope intercept models, repeated measures multivariate analysis of variance, various one-factor models, and hierarchical linear models. When considering models involving multiple constructs, we argue the construct of “general deviance” can be expressed as a single-trait multimethod model, permitting a characterization of the deviance construct over time without requiring restrictive assumptions about the form of growth over time. As an example, prospective assessments of problem behaviors from the Dunedin Multidisciplinary Health and Development Study (Silva & Stanton, 1996) are considered and contrasted with earlier analyses of Hussong, Curran, Moffitt, and Caspi (2008), which supported launch and snare hypotheses. For antisocial behavior, the FCSI model fit better than other models, including the linear chronometric growth curve model used by Hussong et al. For models including multiple constructs, a general deviance model involving a single trait and multimethod factors (or a corresponding hierarchical factor model) fit the data better than either the “snares” alternatives or the general deviance model previously considered by Hussong et al. Taken together, the analyses support the view that linkages and turning points cannot be contrasted with general deviance models absent additional experimental intervention or control.

scholarly journals Validity Evidences for the Portuguese Version of the Problem Behavior Scale: Preschool and Kindergarten Behavior Scales-2

2016 ◽  
Vol 26 (65) ◽  
pp. 273-281
Author(s):  
Sofia Major ◽  
Maria João Seabra-Santos
Keyword(s):  
Factor Analysis ◽  
Problem Behavior ◽  
Problem Behaviors ◽  
Early Identification ◽  
Total Sample ◽  
Model Fit ◽  
Portuguese Version ◽  
Confirmatory Factor ◽  
High Internal Consistency ◽  
Behavior Scale

ABSTRACT The early identification of problem behaviors is essential in preschool. This paper presents evidence of validity (confirmatory factor analysis) for the Problem Behavior scale of the Portuguese version of the Preschool and Kindergarten Behavior Scales - Second Edition (PKBS-2). Analyses were performed for the scale's 46 items, which were grouped into 16 item-parcels. Once it was verified that the model fit the total sample (N = 2000; CFI = .98; RMSEA = .06), analyses were replicated for the samples collected at home and at school (n = 1000 per setting). Results indicate a factor structure equivalent to the original version, with five supplemental subscales, distributed into two subscales (Externalizing and Internalizing), stable for the two subsamples, with high internal consistency levels (α = .78-.97). The discussion highlights the utility/validity of the Portuguese version of the Problem Behavior scale among preschoolers.

Linear mixed model better than repeated measures analysis

2019 ◽  
Vol 30 (6) ◽  
pp. NP1-NP2 ◽  
Author(s):  
Işıl Kutluturk Karagoz ◽  
Berhan Keskin ◽  
Flora Özkalaycı ◽  
Ali Karagöz
Keyword(s):  
Mixed Models ◽  
Repeated Measures ◽  
Mixed Model ◽  
Linear Mixed Model ◽  
Linear Time ◽  
Case Series ◽  
Model Fit ◽  
Repeated Measures Analysis ◽  
Technical Issues ◽  
Better Than

We have some criticism regarding some technical issues. Mixed models have begun to play a pivotal role in statistical analyses and offer many advantages over more conventional analyses regarding repeated variance analyses. First, they allow to avoid conducting multiple t-tests; second, they can accommodate for within-patient correlation; third, they allow to incorporate not only a random coefficient, but also a random slope, typically ‘linear’ time in longitudinal case series when there are enough data and patients’ trajectories vary a lot and improving model fit.

Optimizing dosing in navigation interventions across the cancer continuum: A planned secondary data analysis from the Patient Care Connect implementation.

2017 ◽  
Vol 35 (31_suppl) ◽  
pp. 174-174
Author(s):  
Elizabeth Ann Kvale ◽  
Maria J Pisu ◽  
Courtney Williams ◽  
Kelly Kenzik ◽  
Andres Azuero ◽  
...  
Keyword(s):  
Repeated Measures ◽  
Linear Models ◽  
Secondary Data ◽  
Administrative Claims ◽  
Cost Models ◽  
Total Cost ◽  
Baseline Cost ◽  
Optimal Approach ◽  
The Impact ◽  
Over Time

174 Background: Patient navigation programs in cancer care have historically focused on assisting persons to overcome barriers to accessing care. Evidence is emerging to support the impact of navigation interventions across the cancer continuum. However, navigation programs have varied designs, resulting in a lack of clarity about the optimal approach to delivering services to patients, and a lack of evidence linking program design to outcomes. Methods: A planned retrospective analysis of Medicare administrative claims for a population of older beneficiaries diagnosed with cancer: The main exposure was the number of contacts in person or over the phone with PCCP navigators in the 6 month period starting from the quarter in which patients enrolled in the PCCP. Repeated measures generalized linear models with normal distribution were used to evaluate trends in total cost over time based on: number of contacts, quarters post-enrollment (TIME), and the interaction between number of contacts and TIME. Intra-correlation was controlled for repeated measures. Results: 4,337 patients were included in this analysis. 17.9% had one contact, 17.7% had two, 22.2% had 3-4, 24.2% had 5-10, and 18.0% had more than 10 contacts. African Americans had a greater number of participants with more than 10 navigator contacts, as stage 4 cancers, and initial or end-of-life phase of care. Patients who received more than 3 contacts had significantly higher levels of baseline cost. Models to evaluate total cost over time demonstrate an effect of navigator contact on cost that is associated with number of contacts. This trend is statistically significant at 3-4 contacts or more, and remains significant at 10 or more contacts. Conclusions: Increased navigator contact is associated with increased slope of decline in utilization and cost indicates that navigation programs should be adequately resourced to deliver care that enables navigators to have contact with patients a minimum of 3-4 contacts over a six month period.

Revisiting the role of dissimilarity of host communities in driving dissimilarity of ectoparasite assemblages: non-linear vs linear approach

Parasitology ◽  
2017 ◽  
Vol 144 (10) ◽  
pp. 1365-1374 ◽  
Author(s):  
LUTHER VAN DER MESCHT ◽  
IRINA S. KHOKHLOVA ◽  
ELIZABETH M. WARBURTON ◽  
BORIS R. KRASNOV
Keyword(s):  
Linear Models ◽  
Species Turnover ◽  
Geographic Distance ◽  
Parasite Community ◽  
Model Fit ◽  
Linear Approach ◽  
Non Linear ◽  
Using Data ◽  
Better Than

SUMMARYWe revisited the role of dissimilarity of host assemblages in shaping dissimilarity of flea assemblages using a non-linear approach. Generalized dissimilarity models (GDMs) were applied using data from regional surveys of fleas parasitic on small mammals in four biogeographical realms. We compared (1) model fit, (2) the relative effects of host compositional and phylogenetic turnover and geographic distance on flea compositional and phylogenetic turnover, and (3) the rate of flea turnover along gradients of host turnover and geographic distance with those from earlier application of a linear approach. GDMs outperformed linear models in explaining variation in flea species turnover and host dissimilarity was the best predictor of flea dissimilarity, irrespective of scale. The shape of the relationships between flea compositional turnovers along host compositional turnover was similar in all realms, whereas turnover along geographic distance differed among realms. In contrast, the rate of flea phylogenetic turnover along gradients of host phylogenetic turnover differed among realms, whereas flea phylogenetic turnover did not depend on geographic distance in any realm. We demonstrated that a non-linear approach (a) explained spatial variation in parasite community composition better than and (b) revealed patterns that were obscured by earlier linear analyses.

scholarly journals Repeated measures ASCA+ for analysis of longitudinal intervention studies with multivariate outcome data

2020 ◽  
Author(s):  
Torfinn S. Madssen ◽  
Guro F. Giskeødegård ◽  
Age K. Smilde ◽  
Johan A. Westerhuis
Keyword(s):  
Clinical Trials ◽  
Repeated Measures ◽  
Intervention Studies ◽  
Linear Models ◽  
Principal Component ◽  
Component Analysis ◽  
Outcome Data ◽  
Biological Data ◽  
Changes Over Time ◽  
Over Time

AbstractLongitudinal intervention studies with repeated measurements over time are an important type of experimental design in biomedical research. Due to the advent of “omics”-sciences (genomics, transcriptomics, proteomics, metabolomics), longitudinal studies generate increasingly multivariate outcome data. Analysis of such data must take both the longitudinal intervention structure and multivariate nature of the data into account. The ASCA+-framework combines general linear models with principal component analysis, and can be used to separate and visualize the multivariate effect of different experimental factors. However, this methodology has not yet been developed for the more complex designs often found in longitudinal intervention studies, which may be unbalanced, involve randomized interventions, and have substantial missing data. Here we describe a new methodology, repeated measures ASCA+ (RM-ASCA+), and show how it can be used to model metabolic changes over time, and compare metabolic changes between groups, in both randomized and non-randomized intervention studies. Tools for both visualization and model validation are discussed. This approach can facilitate easier interpretation of data from longitudinal clinical trials with multivariate outcomes.Author summaryClinical trials are increasingly generating large amounts of complex biological data. Examples can include measuring metabolism or gene expression in tissue or blood sampled repeatedly over the course of a treatment. In such cases, one might wish to compare changes in not one, but hundreds, or thousands of variables simultaneously. In order to effectively analyze such data, both the study design and the multivariate nature of the data should be considered during data analysis. ANOVA simultaneous component analysis+ (ASCA+) is a statistical method which combines general linear models with principal component analysis, and provides a way to separate and visualize the effects of different factors on complex biological data. In this work, we describe how repeated measures linear mixed models, a class of models commonly used when analyzing changes over time and treatment effects in longitudinal studies, can be used together with ASCA+ for analyzing clinical trials in a novel method called repeated measures-ASCA+ (RM-ASCA+).

scholarly journals Spatial and Temporal Variation of Mortality and Deprivation 2: Statistical Modelling

1998 ◽  
Vol 30 (10) ◽  
pp. 1815-1834 ◽  
Author(s):  
M L Senior ◽  
H C W L Williams ◽  
G Higgs
Keyword(s):  
Cross Sections ◽  
Repeated Measures ◽  
Fixed Effects ◽  
Linear Models ◽  
Nonlinear Effects ◽  
Statistical Modelling ◽  
Spatial And Temporal Variation ◽  
Fixed Effects Model ◽  
Poisson Models ◽  
Over Time

Building on the tabular analyses exemplified in our first paper and widely used in the medical literature, we use generalised linear models to provide a formal, statistical approach to the analysis of mortality and deprivation relationships, and their change over time. Three types of fixed effects model are specified and estimated with the same ward-level data sets for Wales examined in our first paper. They are: Poisson models for analysing mortality and deprivation at a single cross section in time; repeated-measures Poisson models for analysing mortality–deprivation relations, not only at cross sections in time, but also their changes over time; and logit models focusing on temporal changes in mortality–deprivation relationships. Nonlinear effects of deprivation on mortality have been explored by using dummy variables representing deprivation categories to establish the connection between formal statistical models and the tabular approach.

Threshold models improve estimates of molt parameters in datasets with small sample sizes

The Auk ◽  
2021 ◽  
Author(s):  
Ryan S Terrill ◽  
Youyi Fong ◽  
Jared D Wolfe ◽  
Amanda J Zellmer
Keyword(s):  
Linear Models ◽  
Global Climate ◽  
Threshold Model ◽  
Life History Evolution ◽  
Small Sample ◽  
Model Fit ◽  
Change Points ◽  
Least Squares Regression ◽  
Sample Sizes ◽  
Over Time

Abstract The timing of events in birds’ annual cycles is important to understanding life history evolution and response to global climate change. Molt timing is often measured as an index of the sum of grown feather proportion or mass within the primary flight feathers. The distribution of these molt data over time has proven difficult to model with standard linear models. The parameters of interest are at change points in model fit over time, and so least-squares regression models that assume molt is linear violate the assumption of even variance. This has led to the introduction of other nonparametric models to estimate molt parameters. Hinge models directly estimate changes in model fit and have been used in many systems to find change points in data distributions. Here, we apply a hinge model to molt timing, through the introduction of a double-hinge (DH) threshold model. We then examine its performance in comparison to current models using simulated and empirical data. Our results suggest that the Underhill–Zucchini (UZ) and Pimm models perform well under many circumstances and appear to outperform the DH model in datasets with high variance. The DH model outperforms the UZ model at low sample sizes of birds in active molt and shorter molt durations and provides more realistic confidence intervals at smaller sample sizes. The DH model provides a novel addition to the toolkit for estimating molt phenology, expanding the conditions under which molt can accurately be estimated.

scholarly journals Repeated measures ASCA+ for analysis of longitudinal intervention studies with multivariate outcome data

2021 ◽  
Vol 17 (11) ◽  
pp. e1009585
Author(s):  
Torfinn S. Madssen ◽  
Guro F. Giskeødegård ◽  
Age K. Smilde ◽  
Johan A. Westerhuis
Keyword(s):  
Repeated Measures ◽  
Intervention Studies ◽  
Linear Models ◽  
Principal Component ◽  
Outcome Data ◽  
Repeated Measurements ◽  
Metabolic Changes ◽  
Randomized Intervention ◽  
Over Time ◽  
Multivariate Outcome

Longitudinal intervention studies with repeated measurements over time are an important type of experimental design in biomedical research. Due to the advent of “omics”-sciences (genomics, transcriptomics, proteomics, metabolomics), longitudinal studies generate increasingly multivariate outcome data. Analysis of such data must take both the longitudinal intervention structure and multivariate nature of the data into account. The ASCA+-framework combines general linear models with principal component analysis and can be used to separate and visualize the multivariate effect of different experimental factors. However, this methodology has not yet been developed for the more complex designs often found in longitudinal intervention studies, which may be unbalanced, involve randomized interventions, and have substantial missing data. Here we describe a new methodology, repeated measures ASCA+ (RM-ASCA+), and show how it can be used to model metabolic changes over time, and compare metabolic changes between groups, in both randomized and non-randomized intervention studies. Tools for both visualization and model validation are discussed. This approach can facilitate easier interpretation of data from longitudinal clinical trials with multivariate outcomes.

The development of problem behavior syndromes in at-risk adolescents

1994 ◽  
Vol 6 (2) ◽  
pp. 323-342 ◽  
Author(s):  
Joseph P. Allen ◽  
Bonnie J. Leadbeater ◽  
J. Lawrence Aber
Keyword(s):  
At Risk ◽  
Problem Behavior ◽  
Problem Behaviors ◽  
Unprotected Sexual Intercourse ◽  
Service Programs ◽  
Multiple Paths ◽  
Hard Drugs ◽  
Behavior Syndrome ◽  
At Risk Adolescents ◽  
Over Time

AbstractThis study examined multiple paths that can explain the co-occurrence of behaviors comprising a problem behavior syndrome in adolescence. Two hundred sixteen 15–18-year-olds in service programs for at-risk adolescents were assessed twice over a 6–12-month period to examine predictors of changes in levels of their delinquency, unprotected sexual intercourse, and use of soft and of hard drugs. This study considered (a) potential common predictors of multiple behaviors, (b) predictive links among behaviors over time, and (c) whether or not important unique aspects of individual behaviors remain in spite of their co-occurrence. Results were consistent with the hypothesis that the co-occurrence of problem behaviors results from multiple pathways of influence. The future occurrence of several problem behaviors was predicted by adolescents' initial negative expectations in social interactions. In addition, alcohol and marijuana use predicted increases in several other problem behaviors over time. Finally, individual problem behaviors retained important unique characteristics, suggesting the need for further research examining both their syndromal and unique aspects.

Exploring the Multidimensional Facets of Authoritarianism: Authoritarian Aggression and Social Dominance Orientation

2008 ◽  
Vol 67 (1) ◽  
pp. 51-60 ◽  
Author(s):  
Stefano Passini
Keyword(s):  
Social Dominance ◽  
Factor Model ◽  
Three Dimensional ◽  
Social Dominance Orientation ◽  
Model Fit ◽  
Regression Analyses ◽  
One Dimensional ◽  
Confirmatory Factor ◽  
The One ◽  
Better Than

The relation between authoritarianism and social dominance orientation was analyzed, with authoritarianism measured using a three-dimensional scale. The implicit multidimensional structure (authoritarian submission, conventionalism, authoritarian aggression) of Altemeyer’s (1981, 1988) conceptualization of authoritarianism is inconsistent with its one-dimensional methodological operationalization. The dimensionality of authoritarianism was investigated using confirmatory factor analysis in a sample of 713 university students. As hypothesized, the three-factor model fit the data significantly better than the one-factor model. Regression analyses revealed that only authoritarian aggression was related to social dominance orientation. That is, only intolerance of deviance was related to high social dominance, whereas submissiveness was not.

Sign in / Sign up

Export Citation Format

Share Document