Article

Growth Mixture Modeling

July 2007
Organizational Research Methods 10(4):635-656

July 2007
10(4):635-656

Authors:

Portland State University

An important limitation of conventional latent-growth modeling (LGM) is that it assumes that all individuals are drawn from one or more observed populations. However, in many applied-research situations, unobserved subpopulations may exist, and their different latent trajectories may be the focus of research to test theory or to resolve inconsistent prior research findings. Conventional LGM does not help to identify and predict these unobserved subpopulations. This article introduces the growth-mixture modeling (GMM) method for these purposes. Given that GMM handles longitudinal data (i.e., nesting of time observations within individuals) and identifies unobserved subpopulations (i.e., the nesting of individuals within latent classes), GMM can be construed as a multilevel modeling technique. The modeling procedure of GMM is illustrated on a simulated data set. Steps in the modeling process are highlighted and limitations, cautions, recommendations, and extensions of using GMM are discussed. Technical references for additional information are noted throughout.

One slope does not fit all: longitudinal trajectories of quality of life in older adulthood

Article

Full-text available

Aug 2021
QUAL LIFE RES

Purpose: Maintaining or improving quality of life (QoL) in later life has become a major policy objective. Yet we currently know little about how QoL develops at older ages. The few studies that have modelled QoL change across time for older adults have used 'averaged' trajectories. However, this ignores the variations in the way QoL develops between groups of older adults. Methods: We took a theoretically informed 'capabilities approach' to measuring QoL. We used four waves of data, covering 6 years, from the New Zealand Health, Work and Retirement Study (NZHWR) (N = 3223) to explore whether distinct QoL trajectories existed. NZHWR is a nationally representative longitudinal study of community-dwelling adults aged 50 + in New Zealand. Growth mixture modelling was applied to identify trajectories over time and multinomial regressions were calculated to test baseline differences in demographic variables (including age, gender, ethnicity, education and economic living standards). Results: We found five QoL trajectories: (1) high and stable (51.94%); (2) average and declining (22.74%); (3) low and increasing (9.62%); (4) low and declining (10.61%); (5) low and stable (5.09%). Several differences across profiles in baseline demographic factors were identified, with economic living standards differentiating between all profiles. Conclusions: The trajectory profiles demonstrate that both maintaining and even improving QoL in later life is possible. This has implications for our capacity to develop nuanced policies for diverse groups of older adults.

A Person-Centered Approach to Behaving Badly at Work: An Examination of Workplace Deviance Patterns

Article

Full-text available

May 2024

To investigate research questions surrounding workplace deviance, scholars have primarily applied variable-centered approaches, such as overall deviance measures or those that separate interpersonal deviance and organizational deviance. These approaches, however, ignore that individuals might employ more complex combinations of deviance behaviors that do not fit neatly within the existing variable frameworks. The present study explores whether person-centered deviance classes emerge in a comprehensive database of the prior studies. We then investigated whether these classes showed differences in antecedents and correlates in an independent sample of working adults from multiple industries. In Study 1, a multilevel latent class analysis of 20 independent samples and 6,218 individuals revealed five classes of workplace deviance, thus providing preliminary support for a person-centered approach. In Study 2, a time-lagged sample of 553 individuals showed the emergence of five classes that largely reflected the patterns found in Study 1. Study 2 points to meaningful differences between classes of deviance behaviors and antecedents, including abusive supervision, Openness, Conscientiousness, Agreeableness, Emotional Stability, and psychological entitlement; classes are also uniquely associated with correlates such as organizational citizenship behaviors, turnover intentions, job performance, and job satisfaction. Altogether, this work is an important first step toward understanding workplace deviance with a person-centered lens.

Latent trajectories of subjective well- being: An application of Latent Growth Curve and Latent Class Growth Modeling

Article

Jun 2023

This study proposed a three-stage measurement model utilizing the Latent Growth Curve Modeling and Latent Class Growth Analysis. The measurement model was illustrated using repeated data collected through a four-week prospective study tracking the subjective well-being of volunteer college students (n=154). Firstly, several unconditional growth models were estimated to define the model, providing a better representation of individual growth trajectories. Secondly, several conditional growth models were formulated to test the usefulness of covariate variables hypothesized to explain observed variance in growth factors. Finally, latent class growth models were estimated to further explore different latent trajectory classes. Results showed that students' subjective well-being changed over time, and the rate of this change and its covariates were not constant for the entire sample. This study clearly illustrates how a longitudinal measurement approach can enhance the scope of findings and the depth of inferences when repeated measurements are available.

Classification and prediction of cognitive trajectories of cognitively unimpaired individuals

Article

Full-text available

Mar 2023

Objectives Efforts to prevent Alzheimer’s disease (AD) would benefit from identifying cognitively unimpaired (CU) individuals who are liable to progress to cognitive impairment. Therefore, we aimed to develop a model to predict cognitive decline among CU individuals in two independent cohorts.MethodsA total of 407 CU individuals from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and 285 CU individuals from the Samsung Medical Center (SMC) were recruited in this study. We assessed cognitive outcomes by using neuropsychological composite scores in the ADNI and SMC cohorts. We performed latent growth mixture modeling and developed the predictive model.ResultsGrowth mixture modeling identified 13.8 and 13.0% of CU individuals in the ADNI and SMC cohorts, respectively, as the “declining group.” In the ADNI cohort, multivariable logistic regression modeling showed that increased amyloid-β (Aβ) uptake (β [SE]: 4.852 [0.862], p < 0.001), low baseline cognitive composite scores (β [SE]: −0.274 [0.070], p < 0.001), and reduced hippocampal volume (β [SE]: −0.952 [0.302], p = 0.002) were predictive of cognitive decline. In the SMC cohort, increased Aβ uptake (β [SE]: 2.007 [0.549], p < 0.001) and low baseline cognitive composite scores (β [SE]: −4.464 [0.758], p < 0.001) predicted cognitive decline. Finally, predictive models of cognitive decline showed good to excellent discrimination and calibration capabilities (C-statistic = 0.85 for the ADNI model and 0.94 for the SMC model).Conclusion Our study provides novel insights into the cognitive trajectories of CU individuals. Furthermore, the predictive model can facilitate the classification of CU individuals in future primary prevention trials.

BMI Trends for Veterans Up to 10 Years After VA Enrollment Following Military Discharge

Article

Oct 2022
J GEN INTERN MED

Background: Obesity (body mass index [BMI]≥30kg/m2) among US adults has tripled over the past 45 years, but it is unclear how this population-level weight change has occurred. Objective: We sought to identify distinct long-term BMI trajectories and examined associations with demographic and clinical characteristics. Design: The design was latent trajectory modeling over 10 years of a retrospective cohort. Subgroups were identified via latent class growth mixture models, separately by sex. Weighted multinomial logistic regressions identified factors associated with subgroup membership. Participants: Participants were a retrospective cohort of 292,331 males and 62,898 females enrolled in VA. Main measures: The main outcome measure was 6-month average VA-measured BMI over the course of 10 years. Additional electronic health record measures on demographic, clinical, and services utilization characteristics were also used to characterize latent trajectories. Key results: Four trajectories were identified for men and for women, corresponding to standard BMI categories "normal weight" (BMI <25), "overweight" (BMI 25-29.99), and "with obesity" (BMI ≥30): "normal weight" and increasing (males: 28.4%; females: 22.8%), "overweight" and increasing (36.4%; 35.6%), "with obesity" and increasing (33.6%; 40.0%), and "with obesity" and stable (males: 1.6%) or decreasing (females: 1.6%). Race, ethnicity, comorbidities, mental health diagnoses, and mental health service utilization discriminated among classes. Conclusions: BMI in the 10 years following VA enrollment increased modestly. VA should continue prioritizing weight management interventions to the large number of veterans with obesity upon VA enrollment, because the majority remain with obesity.

Growth in basic science knowledge in first-year medical school and USMLE Step 1 results: a longitudinal investigation at one school

Article

Full-text available

Aug 2022
ADV HEALTH SCI EDUC

Purpose Our US medical school uses National Board of Medical Examiners (NBME) tests as progress tests during the pre-clerkship curriculum to assess students. In this study, we examined students’ growth patterns using progress tests in the first year of medical school to identify students at risk for failing United States Medical Licensing Examination (USMLE) Step 1. Method Growth Mixture Modeling (GMM) was used to examine the growth trajectories based on NBME progress test scores in the first year of medical school. Achieving a passing score on the USMLE Step 1 at the end of the second year of medical school was used as the distal outcome, controlling for Medical College Admissions Test (MCAT) scores and underrepresented in medicine (URiM) status. Results A total of 518 students from a US medical school were included in the analysis. Five different growth patterns were identified based on students’ NBME test results. Seventy-eight students identified in Group 1 had the lowest starting NBME test score (mean = 33.6, 95% CI 32.0–35.2) and lowest growth rate (mean = 2.30, 95% CI 2.06–2.53). All 26 students who failed Step 1 at the end of the second year were in Group 1 (failing rate = 33%). Meanwhile Group 4 (n = 65 students) had moderate starting NBME test scores (mean = 37.9, 95% CI 36.3–39.0) but the highest growth rate with mean slope at 6.07 (95% CI 5.40–6.73). This group of students achieved significant higher USMLE Step1 scores comparing with the 3 other groups of students (P < 0.05). Conclusions Our study found students had heterogeneous growth patterns in progress test results in their first year of medical school. Growth patterns were highly predictive of USMLE step 1 results. This study can provide performance benchmarks for our future students to assess their progress and for medical educators to identify students who need support and guidance.

Detecting Patterns of Engagement in a Digital Cognitive Skills Training Game

Article

Jan 2021
COMPUT EDUC

Indicators of behavioral engagement derived from log data may provide insight about variation in participants’ interactions with and the efficacy of digital cognitive skills training games. We first sought to determine whether distinct groups of adolescents (N = 163; Mean age = 14.1 years, SD =1.3) could be identified based on variables derived from digital log data collected while participants played a game designed to enhance inhibitory control. We then examined whether these data-driven participant groupings were associated with improvement in inhibitory control. Latent class mixture modeling was conducted both with reaction time and a measure of response accuracy (d’) of log data. Results indicated two distinct classes based on reaction time, and four classes based on response accuracy over the course of training. Class membership based on reaction time was associated with differential improvements in performance on a subsequent standardized measure of inhibitory control. The findings point towards the need for formative metrics of progress, as well as the need for more adaptive and user-centered cognitive skills interventions. Our findings suggest that there may be some utility in analyzing log data as an indicator of student engagement, particularly when used in complement with more traditional measures of performance.

Patterns of Early Change in Interpersonal Problems and Their Relationship to Nonverbal Synchrony and Multidimensional Outcome

Article

Full-text available

Jul 2020

Early change is an increasing area of investigation in psychotherapy research. In this study, we analyzed patterns of early change in interpersonal problems and their relationship to nonverbal synchrony and multiple outcome measures for the first time. We used growth mixture modeling to identify different latent classes of early change in interpersonal problems with 212 patients who underwent cognitive-behavioral treatment including interpersonal and emotion-focused elements. Furthermore, videotaped sessions were analyzed using motion energy analysis, providing values for the calculation of nonverbal synchrony to predict early change in interpersonal problems. The relationship between early change patterns and symptoms as well as overall change in interpersonal problems was also investigated. Three latent subgroups were identified: 1 class with slow improvement (n = 145), 1 class with fast improvement (n = 12), and 1 early deterioration class (n = 55). Lower levels of early nonverbal synchrony were significantly related to fast improvement in interpersonal change patterns. Furthermore, such patterns predicted treatment outcome in symptoms and interpersonal problems. The results suggest that nonverbal synchrony is associated with early change patterns in interpersonal problems, which are also predictive of treatment outcome. Limitations of the applied methods as well as possible applications in routine care are discussed. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

A Latent Class Analysis of Adverse Life Events for Kenyan Adolescents

Article

Full-text available

Dec 2023
J Child Adolesc Trauma

Extant evidence indicates that exposure to adverse childhood experiences (ACE) tend to cluster among children and adolescents. Considering that adolescents from African countries present higher risk of being exposed to multiple ACE compared to other countries, the identification of victimization profiles in this population is clearly warranted. The aim of this study was to determine meaningful clusters of individuals with similar experiences of ACE in a sample of Kenyan adolescents. Latent class analysis (LCA) was conducted to identify latent classes of exposure to ACE. In addition, the relationships between the latent classes and gender, parental education, living arrangements and diagnosis of post-traumatic stress disorder (PTSD) were estimated. A three-class solution was found to be the best description of ACE, and the classes were labelled ‘‘Low Risk’’, ‘‘Intermediate Risk’’, and ‘‘High Risk’’. Compared with the Low-Risk class, the High-Risk class was found to be significantly more likely to have a diagnosis of PTSD and being a female may be an antecedent risk factor for high exposure to ACE. The Intermediate Risk class was significantly less likely to have parents with high school or college education. This paper indicates that Kenyan adolescents present higher risk of being exposed to multiple ACE and that trauma research may turn its focus on the individual as the unit of analysis rather than traumatic events.

Employability in the post-job security era: testing competing effects of perceived job insecurity on perceived employability change

Article

Full-text available

Oct 2023

Perceived job insecurity and perceived employability are often mentioned in one breath with employability typically referred to as the “modern response to job insecurity”. Yet our understanding of how individuals’ perceptions of employability may change over time in response to job insecurity is limited. Both positive and negative changes seem plausible: Job insecurity may trigger employees to invest in employability, making them feel more employable. However, job insecurity may also elicit a defensive response in employees that undermines their perceived employability. We tested these two competing hypotheses against the background of conservation of resources theory in a sample of 358 employees surveyed on three occasions across 3.5 years. Using latent change score modelling, our findings suggest that job insecurity increases perceived employability. That is, the state-level of perceived job insecurity predicts a positive subsequent change in perceived employability. These findings highlight the importance of considering the dynamic within-person perspective to understand the relationship between job insecurity and perceived employability, and illustrate that results observed in prior static research may lead to different conclusions in within-person longitudinal studies. Implications of our findings for theory and practice are discussed.

Profiles of Early Childhood Adversity in an Urban Pediatric Clinic: Implications for Pediatric Primary Care

Article

Full-text available

Jun 2023

Pediatricians are well-positioned to screen for early childhood adversities, but effective responses to positive screens require an understanding of which adversities typically co-occur, and to what extent they are associated with other risk or protective factors. Among children seen at an urban academic pediatric practice, this study aimed to (1) examine the prevalence of different types of early adversity and protective experiences reported by primary caregivers, and (2) define latent classes of co-occurring adversities. Of 1434 children whose parents completed the Safe Environment for Every Kid (SEEK) at well-child visits during November 2019–January 2021, three classes of adverse experiences emerged, including those reporting low adversity (L; 73%), caregiver stress (CS; 17%), and both caregiver stress and depression (CSD; 10%). Among those who also completed the Adverse Childhood Experiences Questionnaire (ACE-Q, n = 1373) and the Protective and Compensatory Experiences Scale (PACES, n = 1377), belonging to the L class was associated with lower ACE-Q and higher PACES scores. For parent-respondents only, ACE-Q scores were significantly greater for the CSD class compared to the CS and L classes. Pediatricians should attend to the needs of caregivers reporting both stress and depression, as these families may face especially high levels of adversity and low levels of protective factors.

When time is running out: A growth curve analysis of older workers' retirement intentions

Article

Feb 2023

Using dynamic theory and methods, we investigate the phenomenon of older workers who withdraw from paid work while still healthy. We focus on intention to retire as the penultimate stage in the retirement process. We extend socio‐emotional selectivity theory to explain the growth of intention to retire. Older workers have a rising perception of time running out but good health allows for an ongoing choice between remaining in work or active retirement. While, in general, older people in poor health have a greater intention to retire than those in good health, we hypothesize that the passage of time motivates the healthy to increase their intention to retire, especially when manager support is low. We examine longitudinal data consisting of three waves of survey responses (2011, 2012, 2013) from 495 workers in their fiftieth year and older. We employ growth curve analysis (random coefficient modeling). The findings show that over a two‐year period, in contrast to other older workers whose retirement intention remains stable, individuals in consistently good health but with low manager support demonstrate a growth in intention to retire. That is, we identify the “queue jumpers”: those workers who speeded up their retirement process relative to other older workers.

Timing and Frequency of Screening in Schools: A Latent Variable Analysis of Behavioral Stability Over Time

Article

Full-text available

Aug 2022

The growth of school-based initiatives incorporating multitiered systems of support (MTSS) for social, emotional, and behavioral domains has fueled interest in behavioral assessment. These assessments are foundational to determining risk for behavioral difficulties, yet research to date has been limited with regard to when and how often to administer them. The present study evaluated these questions within the framework of behavioral stability and examined the extent to which behavior is stable when measured by two school-based behavioral assessments: the Direct Behavior Rating-Single-Item Scales (DBR-SIS), and the Behavioral and Emotional Screening System (BESS). Participants included 451 students rated three times per year across 4 years, with the primary teacher from each year providing the within-year ratings. Latent variable models were employed to measure the constructs underlying the observed assessment scores. Models demonstrated that the DBR-SIS best captured changes within the year, whereas the BESS scores remained stable across time points within a year. Across years, scores from both assessments captured changes. The unique contributions of each assessment in the data-based decision-making process are discussed, and recommendations are given for their combined use within and across school years. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Differences in body mass index trajectories of adolescent psychiatric inpatients by sex, age, diagnosis and medication: an exploratory longitudinal, mixed effects analysis

Article

Full-text available

Jul 2022
Child Adolesc Ment Health

Background Adolescents in secure psychiatric care typically report high obesity rates. However, longitudinal research exploring the rate and extent of change is sparse. This study aimed to analyse sex differences in longitudinal body mass index (BMI) change for adolescents receiving treatment in a secure psychiatric hospital. Methods The sample comprised 670 adolescents in secure psychiatric care. BMI trajectories from admission to 50 months of hospitalisation were produced using sex‐stratified multilevel models. Systematic difference in mean BMI trajectories according to age at admission (14, 15, 16, or 17 years), medication (Olanzapine or Sodium Valproate), and primary diagnosis (Psychotic, non‐Psychotic or Functional/behavioural disorders) were investigated. Results Together, males and females experienced a mean BMI increase of 2.22 m/kg² over the 50‐month period. For females, BMI increased from 25.69 m/kg² to 30.31 m/kg², and for males, reduced from 25.01 m/kg² to 23.95 m/kg². From 30 to 50 months, a plateau was observed for females and a reduction in BMI observed for males. Psychotic disorders in males (β 3.87; CI 1.1–6.7) were associated with the greatest rate of BMI change. For medication, Olanzapine in females was associated with the greatest rate of change (β1.78; CI −.89–4.47). Conclusions This is the first longitudinal study exploring longitudinal BMI change for adolescent inpatients. Results highlight that individual differences in adolescent inpatients result in differing levels of risk to weight gain in secure care. Specifically, males with psychotic disorders and females taking Olanzapine present the greatest risk of weight gain. This has implications for the prioritisation of interventions for those most at risk of weight gain.

Dynamics of Language Learning Motivation and Emotions: A Parallel-Process Growth Mixture Modeling Approach

Article

Full-text available

Jun 2022

The present study adopted a novel parallel-process growth mixture modeling (GMM) technique to research the adaptive interaction between foreign language learners’ learning motivation and emotions, with a view to advancing our understanding of how language learning motivation and emotions (enjoyment and anxiety) adaptively interact with each other over time. The present study, situated in the Chinese English as a foreign language (EFL) learning context, collected learning motivation and emotion data from 176 Chinese EFL learners over a period of two semesters (12 months). The GMM technique adopted in the study identified three developmental profiles of motivation and two of emotions, respectively. The study further distilled salient patterns of motivation–emotion interaction over time, patterns significant for designing and implementing pedagogical interventions for motivation enhancement. The parallel-process GMM technique was also proven to be a useful approach to parsing learner variety and learning heterogeneity, efficiently summarizing the complex, dynamic processes of motivation and emotion development.

Article

Full-text available

May 2022

Satu Laitinen

The aim of this cross-sectional study was to examine the well-being of Finnish health education teachers (n = 108) by examining the latent profiles of work burnout and work engagement by using a person-centered approach. Additionally, this study explored to what extent different job and personal resources (social support, pedagogical self-efficacy, and social belonging) and job demands (work overload) are associated with teachers’ belonging to the work-related well-being profiles. The Job Demands-Resources model was used as the theoretical framework for this study. The study found that three different work-related well-being profiles could be identified among health education teachers: those who were engaged (45%), those who were already experiencing burnout (43%), and those at risk of burnout (12%). The more demands the teachers experienced, the likelier they were to belong to the burnout profile. Experiences of pedagogical self-efficacy, social belonging, and social support increased the probability of belonging to the engaged profile group. Determining job and personal resources and job demands might be beneficial for health education teacher well-being.

Behavior change versus stability during the college‐to‐work transition: Life course and the “stickiness” of alcohol misuse at career entry

Article

Full-text available

May 2022
PERS PSYCHOL

To what extent and under what conditions do college graduates disengage from employment‐incompatible behaviors during the college‐to‐work transition? Drawing from the life course perspective, we proposed a model highlighting considerable stability of employment‐incompatible behaviors during initial months of organizational socialization. Our model predicted that maturing out of such behaviors, which is expected by employers and beneficial to career adjustment, would be more likely to occur given a conducive transition context. Using a large dataset tracking graduates from their last semester in college to up to approximately 1.5 years after graduation and with alcohol use as our empirical referent, we demonstrated that a pattern of high‐risk drinking behavior may remain even after the transition into full‐time employment. We further showed that lower levels of perceived cohort drinking norms and higher levels of mentoring were associated with a higher probability of maturing out, manifesting in a transition from a high‐risk drinking profile before graduation to a moderate drinking profile after starting full‐time employment. Finally, we found that maturing out was associated with lagged outcomes including lower levels of sleep problems and depression and fewer work days lost to absenteeism, thus underscoring the consequential nature of behavior profile shifts during the college‐to‐work transition. This article is protected by copyright. All rights reserved

Examining Subtypes of Emerging Adult Situational Drinking: A Replication and Extension Study

Article

Full-text available

Nov 2021

Background: Emerging adulthood is associated with heavy drinking. Despite overall heavy use, studies show considerable heterogeneity in emerging adult drinking habits. Lau-Barraco and colleagues (2016 b) identified three subtypes (high, moderate, low) of emerging adult heavy drinkers based on patterns of use across common drinking situations. Heavy situational drinkers had more alcohol problems, mental health symptoms, and coping/conformity motives for alcohol use. Objective: Our goal was to replicate and extend the aforementioned study, expecting to find the same subgroups, then examining whether certain risk factors predicted subgroup membership. Methods/Results: Undergraduates (N = 497) completed online self-report measures and a latent profile analysis (LPA) found support for three similar subtypes; low, “moderate” (higher endorsement of pleasant emotion/social pressure situations, relative to the low group), and high. Univariate ANOVAs, followed by pairwise comparisons, found that heavy situational drinkers scored highest on measures of alcohol problems, problem gambling, drug use, depression, and anxiety compared to the other two groups, and consistent with previous findings. Conclusions: This study showed that emerging adults who drink heavily across various situations are likely to engage in other addictive behaviors and struggle with mental health symptoms. Identifying one’s personal risk factors and their riskiest drinking situations is critical for developing targeted intervention programs and increasing the understanding of the heterogeneous nature of drinking behaviors in emerging adults in Canada.

Cognitive Development in the Context of Trauma: Longitudinal Implications of Intimate Partner Violence for Women and Children

Thesis

Jan 2019

Hannah Clark

Nearly one in three women in the United States have experienced intimate partner violence (IPV), and children are often direct eyewitnesses to these events. IPV places women and children at risk for a range of health problems, and mounting literature indicates that IPV threatens cognitive development as well, particularly with regard to executive functioning (EF). EF refers to the ability to plan, think flexibly, inhibit responses, and redirect attention. Bioecological and developmental cascade models suggest that EF may influence the relation between IPV and a host of mental health concerns in women and children. One such concern involves disruptions in children’s abilities to direct and sustain attention, as executive dysfunction is a core component of Attention-Deficit/Hyperactivity Disorder. Accordingly, this dissertation assessed EF and attention problems in women and children with histories of IPV. The first study of this dissertation examined factors associated with mothers’ EF following IPV. Using longitudinal data, this study revealed unique relationships between IPV and women’s performance on tasks measuring distinct EF domains. Specifically, recent—but not remote—experiences of IPV were associated with impaired interference control, whereas women’s cognitive flexibility was not significantly impacted by recent or remote IPV. Women’s post-intervention depressive symptoms were predictive of impairments in cognitive flexibility eight years later, suggesting that treatment-resistant depression may increase risk for poor EF in women with histories of IPV. By assessing the differential effects of recent and remote IPV on distinct EF domains, this study addressed a gap in the literature on the links between trauma, mental health, and EF. Examining these relationships in women with children is critically important, as deficits in mothers’ EF following IPV may have lasting effects on their children’s development. The second dissertation study evaluated speeded control, an aspect of EF influenced by processing speed, in IPV-exposed children. Results indicated that children’s IPV exposure during the preschool years had a significant, negative impact on their speeded control in late childhood, eight years later. This relation was mediated by the remote effects of IPV on their mothers; specifically, IPV was positively associated with maternal depression, which in turn contributed to greater use of negative parenting strategies when children were of preschool age. Children’s IPV exposure during late childhood was not predictive of their concurrent speeded control performance, suggesting that the preschool years may be a sensitive period for EF development. This study was the first to assess how the detrimental effects of IPV on women affect their children’s cognition in the long term, and provides compelling evidence for developmental cascade models that emphasize the role of parent-child relationships during early childhood. The final dissertation study assessed the effectiveness of a ten-session intervention in reducing children’s attention problems. Results indicated that IPV exposure interacted with experimental group assignment such that among children exposed to high levels of IPV, those in the Treatment group exhibited fewer attention problems one year post-intervention relative to Controls. There was no treatment effect for children exposed to low levels of IPV. These results inform the implementation of evidence-based interventions for IPV-exposed children and their mothers that address the effects of violence on cognitive development. In doing so, this dissertation has implications for research on both cognitive development and family violence, contributing to two fields of psychology that, until recently, were rarely integrated.

Proactive Personality and Social Support With Pre-retirement Anxiety: Mediating Role of Subjective Career Success

Article

Full-text available

Jul 2021

The main objective of this paper is to investigate the mediating role of subjective career success (SCS) in the relationship between proactive personality, social support (SS), and pre-retirement anxiety. Using a two-wave longitudinal design, 624 pre-retirees were sampled (M = 56.49 years; SD = 4.56); of these, 237 (37.98%) were males and 387 (62.02%) were females. Measurement model and mediation test were performed using the SmartPLS and IBM SPSS Amos software. The result indicated that proactive personality, SS, and SCS showed negative relationships with the dimensions of pre-retirement anxiety (financial preparedness, social obligation, and social alienation). Subjective career success mediated the relationship between proactive personality and pre-retirement anxiety.

Trajectory patterns of blood pressure change up to six years and the risk of dementia: a nationwide cohort study

Article

Full-text available

Jul 2021

The present study aimed to investigate the associations between the trajectory of blood pressure (BP) change and the risk of subsequent dementia and to explore the differences in age, gender, and hypertension subgroups. We included 10,660 participants aged ≥ 60 years from 1998 to 2018 waves of the Chinese Longitudinal Healthy Longevity Survey. Latent growth mixture models were used to estimate BP trajectories. Cox-proportional hazard models were used to analyze the effects of BP trajectories on the risk of dementia. According to the results, stabilized systolic BP (SBP) was found to be associated with a higher risk of dementia compared with normal SBP [adjusted hazard ratio (aHR): 1.62; 95% confidence interval (CI): 1.27-2.07] and elevated SBP (aHR: 2.22; 95% CI: 1.51-3.28) in and only in the subgroups of the oldest-old, women, and subjects without hypertension at baseline. Similarly, stabilized pulse pressure (PP) was associated with a higher risk of dementia compared with normal PP (aHR: 1.52; 95% CI: 1.24-1.88) and elevated PP (aHR: 2.12; 95% CI: 1.48-3.04) in and only in the subgroups of the oldest-old, women, and subjects with hypertension at baseline. These findings suggest that stabilized SBP and PP have predictive significance for the occurrence of dementia in late life, and the factors of age, gender, and late-life hypertension should be considered when estimating the risk of BP decline on dementia.

Developmental Trajectories of Children’s Spatial Skills: Influencing Variables and Associations with Later Mathematical Thinking

Article

Full-text available

Jun 2021
LEARN INSTR

Few studies have examined the long-term relations between children’s early spatial skills and their later mathematical abilities. In the current study, we investigated children’s developmental trajectories of spatial skills across four waves from age 3-7 years and their association with children’s later mathematical understanding. We assessed children’s development in a large, heterogeneous sample of children (N=586) from diverse cultural backgrounds and mostly low-income homes. Spatial and mathematical skills were measured using standardized assessments. Children’s starting points and rate of growth in spatial skills were investigated using latent growth curve models. We explored the influence of various covariates on spatial skill development and found that socioeconomic status, language skills, and sex, but not migration background predicted children’s spatial development. Furthermore, our findings showed that children’s initial spatial skills––but not their rate of growth––predicted later mathematical understanding, indicating that early spatial reasoning may play a crucial role for learning mathematics.

Sustainable Growth of Social Tourism: A Growth Mixture Modeling Approach Using Heterogeneous Travel Frequency Trajectories

Article

Full-text available

May 2021
Int J Environ Res Publ Health

As travel activity has gained attention as one of the essential ways of understanding the sustainable growth of social tourism, a growing number of research projects have been conducted to elucidate the relationship between residents’ travel quantity (frequency) and quality (experience) in both macro and micro perspectives. Yet, very little research has highlighted that travel opportunities are not equally available to residents, especially a longitudinal perspective. The current study classified domestic travelers into four distinct classes using four years of longitudinal data from 5054 Korean residents. Latent growth curve modeling (LGCM) and growth mixture modeling (GMM) were employed to find out (1) the optimal number of classes, (2) the longitudinal travel frequency trajectory of each class, and (3) the distinctive demographic and travel characteristics of the four classes. This study provides some practical implications for policymakers when optimizing available resources for sustainable travel opportunities to relevant target sub-populations. Furthermore, detailed step-by-step analytic tutorials are also introduced for the extended application of longitudinal latent variable analysis in the tourism and hospitality fields, providing additional insights for relevant stakeholders.

Residual-Based Algorithm for Growth Mixture Modeling: A Monte Carlo Simulation Study

Article

Full-text available

Feb 2021

Growth mixture models are regularly applied in the behavioral and social sciences to identify unknown heterogeneous subpopulations that follow distinct developmental trajectories. Marcoulides and Trinchera (2019) recently proposed a mixture modeling approach that examines the presence of multiple latent classes by algorithmically grouping or clustering individuals who follow the same estimated growth trajectory based on an evaluation of individual case residuals. The purpose of this article was to conduct a simulation study that examines the performance of this new approach for determining the number of classes in growth mixture models. The performance of the approach to correctly identify the number of classes is examined under a variety of longitudinal data design conditions. The findings demonstrated that the new approach was a very dependable indicator of classes across all the design conditions considered.

Trajectories of Nicotine and Cannabis Vaping and Polyuse From Adolescence to Young Adulthood

Article

Full-text available

Oct 2020

Importance: Little is known about cannabis vaping trajectories across adolescence and young adulthood or the co-occurrence with nicotine vaping. Objective: To evaluate nicotine vaping and cannabis vaping trajectories from late adolescence to young adulthood (≥18 years of age) and the extent of polysubstance vaping. Design, setting, and participants: In this prospective cohort study, 5 surveys (including information on substance vaped) were completed at 10 high schools in the Los Angeles, California, metro area. Students were surveyed at 6-month intervals from fall of 11th grade (October to December 2015; wave 5) through spring of 12th grade (March to June 2017; wave 8) and again approximately 1 to 2 years after high school (October 2018 to October 2019; wave 9). Exposures: Past 30-day nicotine and cannabis vaping frequency across 5 waves. Main outcomes and measures: Self-reported frequency of nicotine vaping and cannabis vaping within the past 30 days across 5 time points from late adolescence to young adulthood. Trajectories were measured with these past 30-day use frequencies at each wave. Parallel growth mixture modeling estimated conditional probabilities of polysubstance vaping. Results: The analytic sample included 3322 participants with at least 1 time point of data (mean [SD] age, 16.50 [0.42] years at baseline; 1777 [53.5%] female; 1573 [47.4%] Hispanic or Latino). Growth mixture modeling identified the 5-trajectory model as optimal for both nicotine vaping and cannabis vaping. Trajectories for nicotine and cannabis vaping were similar (nonusers: 2246 [67.6%] nicotine, 2157 [64.9%] cannabis; infrequent users: 566 [17.0%] nicotine, 608 [18.3%] cannabis; moderate users: 167 [5.0%] nicotine, 233 [7.0%] cannabis; young adult-onset frequent users: 213 [6.4%] nicotine, 190 [5.7%] cannabis; adolescent-onset escalating frequent users: 131 [3.9%] nicotine, 134 [4.0%] cannabis). Males had greater odds of belonging to the adolescent-onset escalating frequent users nicotine (adjusted odds ratio, 2.88; 95% CI, 1.58-5.23; P < .01) and cannabis (adjusted odds ratio, 1.95; 95% CI,1.03-3.66; P < .05) vaping trajectories compared with nonusers. Polysubstance vaping was common, with those in trajectories reflecting more frequent nicotine vaping (adolescent-onset escalating frequent users and young adult-onset frequent users) having a high probability of membership (85% and 93%, respectively) in a cannabis-use trajectory. Conclusions and relevance: In this cohort study, the prevalence and type of nicotine vaping and cannabis vaping developmental trajectories from late adolescence to young adulthood were similar. Polysubstance vaping was common from late adolescence to young adulthood, particularly among those reporting more frequent vaping use. The findings suggest that public health policy and clinical interventions should address polysubstance vaping in both adolescence and young adulthood.

A Unifying Framework to Study Workplace Decision-Making Aptitude and Performance

Article

Full-text available

Jul 2020

Parental Psychosocial Well-Being as a Predictor of the Social Competence of a Child

Article

Full-text available

Nov 2020
J Child Fam Stud

Our study examined the interrelations between the psychosocial well-being of parents at the time of pregnancy and the social competence of their three-year-old child. Whereas most previous studies have linked the psychosocial well-being of one parent to the social development of their child, newer research has highlighted the importance of examining the psychosocial well-being of both parents and its’ effects to the development of the child. This study used data from the Steps to the Healthy Development and Well-being of Children follow-up study (The STEPS Study, n = 1075) to examine the interrelations between the psychosocial well-being of both the mother and the father during the period of pregnancy and the social competence of their three-year-old child. The interrelations between the psychosocial well-being of one parent and the social competence of their child were studied with regression analyses, and family-level interrelations were modeled with a latent profile analysis of family-level psychosocial well-being. At the dyadic level, the poorer psychosocial well-being of one parent during the pregnancy period mostly predicted poorer social competence in their child. However, at the family level, these links were not statistically significant. The higher level of psychosocial well-being experienced by one parent seemed to protect the development of the social competence of their child. This study emphasizes the need to consider the psychosocial well-being of both parents as a factor that influences the social development of their child.

Response Behavior in Work Stress Surveys: A Qualitative Study on Motivational and Cognitive Processes in Self-and Other-Reports

Article

Full-text available

Feb 2021

Work stressors have major consequences for employees’ health and performance, and it is thus in the interest of organizations to assess them. Although organizations often ask employees to fill out work stress surveys regarding stressors and resources, the literature on survey responding offers only limited advice on how to formulate work stress surveys. Furthermore, self-, supervisor-, and coworker-reports show only low convergence. To deepen our understanding of motivational and cognitive processes when individuals respond to work stress surveys, we used a qualitative, grounded theory approach. We interviewed employees after they responded to representative items, asking them about their thoughts, motivational processes, potential factors that might have biased their responses, and the contexts they considered when responding. Since organizations are often also interested in other-reports of stress at work, we also interviewed supervisors and coworkers. We reached theoretical saturation after 31 interviews. A multi-stage coding-process with three raters resulted in new theoretical findings regarding motivational processes, comparisons, and differences between self- and other-reports. For example, employees sometimes deliberately distort answers for fear of consequences. Furthermore, employees, supervisors, and coworkers undergo different comparison processes. The findings of this study suggest that more specific and context-rich wording of items may lead to a more reliable and comparable assessment of stressors and resources at work.

Describing Patterns of Early Literacy Skill Development in the First Year of School and Reading Instruction in a New Zealand Sample

Article

Full-text available

Jul 2020

This study describes trajectories of early literacy skill development of 99 children (n = 55 boys) in their first year of primary school in New Zealand (NZ). Children were assessed twice weekly for 8 weeks on Dynamic Indicators of Basic Early Literacy Skills (DIBELS; Good & Kaminski, 2011) First Sound Fluency (FSF) and AIMSweb Letter Sound Fluency (LSF; Shinn & Shinn, 2002), with other early literacy and beginning reading skills assessed before and after progress monitoring. FSF and LSF growth trajectories were modeled separately. Multilevel modeling indicated improved performance; however, growth mixture modeling indicated 3 growth trajectories (i.e., latent classes; FSF and LSF, respectively): typical (77.6% of children, 65.7%), developing (10.8%, 14.6%), and limited progress (11.6%, 19.7%). Beginning of year screening was sometimes associated with latent class membership, whereas latent class membership differentiated mid- and year-end literacy skills. Results support progress monitoring of early literacy skills within the NZ context to aid earlier identification of children at-risk for difficulties with reading acquisition. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Latent Class Evaluation in Educational Trials: What Percentage of Children Benefits from an Intervention?

Article

Full-text available

Jun 2020

Educational stakeholders are keen to know the magnitude and importance of different interventions. However, the way evidence is communicated to support understanding of the effectiveness of an intervention is controversial. Typically studies in education have used the standardised mean difference as a measure of the impact of interventions. This measure, commonly known as the effect size, is problematic, in terms of how it is interpreted and understood. In this study, we propose a “gain index” as an alternative metric for quantifying and communicating the effectiveness of an intervention. This is estimated as the difference in the percentage of children who make positive gains between the intervention and control groups. Analysis of four randomized controlled trials in education supports the expectation that most children make progress due to normal school activities, which is independent of the intervention. This study elaborates a method to illustrate how trials with a positive gain index and with a higher percentage of pupils with positive gain in the intervention group can be used to communicate which trials are effective in improving educational outcomes.

Latent profiles of dyadic parent–child interaction and associations with triadic family interaction in early childhood

Article

May 2024

Objective This longitudinal study examined latent profiles of parent–child interaction and their associations with triadic family interaction. Background A child's development is significantly influenced by early family relationships. Family systems theory emphasizes the interrelations between family subsystems, such as parent–child and parent–parent relationships, as well as the family as a whole. However, few studies have examined the relationship between each parent–child dyad and triadic family interaction. Method Fathers and mothers were separately videotaped interacting with their child ( n = 120) at 4 and 18 months and assessed using the Parent–Child Early Relational Assessment. Triadic family interaction was evaluated at 18 months using the Family Alliance Assessment Scale within a Lausanne Trilogue Play setting. Results Four latent profiles of parent–child interaction were identified. Dyadic interaction that was characterized by reciprocity, positive affect, and low negativity was associated with higher family coordination. However, interaction characterized by a lack of reciprocity, negativity, and dyadic tension was related to less coordinated triadic family interaction. Conclusions Well‐functioning parent–child interaction contributes to higher‐quality triadic family interaction. In contrast, challenges in early father–child interaction, including emotional distance, a lack of positive paternal involvement, and limited mutual engagement, are linked to lower‐quality triadic family interaction. Implications This study highlights the importance of supporting early parent–child relationships to promote well‐functioning and coordinated triadic family interaction, and the inclusion of fathers in interventions aiming to improve family dynamics.

Multiple Trajectories and Predictors of Self-Esteem Change in Later Life: A Latent Growth Mixture Modeling Approach

Article

Mar 2024
INT J AGING HUM DEV

Applying latent growth mixture modeling (GMM), this study delves into the examination of self-esteem trajectories in a sample of 5,597 older adults over a nine-year period. Four distinct patterns of self-esteem changes have emerged: low, decreasing, increasing, and high. Additionally, the study explores the relationships between each trajectory and various predictors encompassing demographic factors, socioeconomic status, health, and interpersonal relationships. The findings highlight the significance of these factors in predicting the likelihood of an individual following a specific self-esteem trajectory. Notably, maintaining employment, fostering satisfactory social relationships, and being free of frequent depressive feelings emerged as strong predictors for the stability and increase of high self-esteem. Intriguingly, an average or above-average income was unexpectedly associated with lower levels of self-esteem. The study emphasizes the contribution of GMM to advancing aging research.

Multivariate contaminated normal mixture regression modeling of longitudinal data based on joint mean‐covariance model

Article

Dec 2023

Outliers are common in longitudinal data analysis, and the multivariate contaminated normal (MCN) distribution in model‐based clustering is often used to detect outliers and provide robust parameter estimates in each subgroup. In this paper, we propose a method, the mixture of MCN (MCNM), based on the joint mean‐covariance model, specifically designed to analyze longitudinal data characterized by mild outliers. Our model can automatically detect outliers in longitudinal data and provide robust parameter estimates in each subgroup. We use iteratively expectation‐conditional maximization (ECM) algorithm and Aitken acceleration to estimate the model parameters, achieving both algorithm acceleration and stable convergence. Our proposed method simultaneously clusters the population, identifies progression patterns of the mean and covariance structures for different subgroups over time, and detects outliers. To demonstrate the effectiveness of our method, we conduct simulation studies under various cases involving different proportions and degrees of contamination. Additionally, we apply our method to real data on the number of people infected with AIDS in 49 countries or regions from 2001 to 2021. Results show that our proposed method effectively clusters the data based on various mean progression trajectories. In summary, our proposed MCNM based on the joint mean‐covariance model and MCD of covariance matrices provides a robust method for clustering longitudinal data with mild outliers. It effectively detects outliers and identifies progression patterns in different groups over time, making it valuable for various applications in longitudinal data analysis.

Clustering Countries based on the Trend of COVID-19 Mortality Rates: An Application of Growth Mixture Models

Article

Full-text available

Oct 2023
Open Publ Health J

Background The pattern of death due to COVID-19 is not the same worldwide and requires special approaches and strategies to identify. Objective This study aimed to investigate the pattern of COVID-19 mortality rates in different countries using the Growth Mixture Model (GMM). Methods This longitudinal study examined mortality trends due to COVID-19 for 214 countries during 2020-2022. Data were extracted from the World Health Organization reports. Countries were classified using Latent Growth Models (LGM) and GMM based on reported death trends. Results Countries worldwide were classified into four clusters with different mortality patterns due to COVID-19. The highest increase in the death rate was related to cluster 2, including three countries of Iran, Peru, and Spain. The lowest increase in the death rate in each period belonged to cluster 1, which included about 60% of the world's countries. In cluster 3, most European countries, the United States, and a few countries from South America and Southeast Asia were placed. Italy was the only country in the fourth cluster. Conclusion Our findings showed which countries performed better or worse in dealing with the COVID-19 pandemic.

Analysis of heterogeneous growth changes in longitudinal height of children

Article

Full-text available

Aug 2023
J HEALTH POPUL NUTR

Background There have been methodologies developed for a wide range of longitudinal data types; nevertheless, the conventional growth study is restricted if individuals in the sample have heterogeneous growth trajectories across time. Using growth mixture modeling approaches, we aimed to investigate group-level heterogeneities in the growth trajectories of children aged 1 to 15 years. Method This longitudinal study examined group-level growth heterogeneities in a sample of 3401 males and 3200 females. Data were analyzed using growth mixture modeling approaches. Results We examined different trajectories of growth change in children across four low- and middle-income countries using a data-driven growth mixture modeling technique. The study identified two-group trajectories: the most male samples group ( n = 4260, 69.7%) and the most female samples group ( n = 2341, 81.6%). The findings show that the two groups had different growth trajectories. Gender and country differences were shown to be related to growth factors; however, the association varied depending on the trajectory group. In both latent groups, females tended to have lower growth factors (initial height and rate of growth) than their male counterparts. Compared with children from Ethiopia, children from Peru and Vietnam tended to exhibit faster growth in height over time: In contrast, children from India showed a lower rate of change in both latent groups than that of children from Ethiopia. Conclusions The height of children in four low- and middle-income countries showed heterogeneous changes over time with two different groups of growth trajectories.

Assessing the Efficacy of Online Learning in Disparate Business Subjects: Lessons from Distributed Practice and Social Learning Theory

Article

Full-text available

Jun 2023
J Manag Educ

For many reasons, higher education is shifting toward more online instruction. As this shift occurs, educators and administrators should be aware that the efficacy of online courses may be influenced by course content. Specifically, student learning may suffer as courses utilizing significant quantitative content, such as accounting and finance, transition from in-person instruction to online. We investigate this possibility in three quasi-experiments, which compare online and face-to-face instruction in four disparate business minor courses. In our first study we obtain the predicted interaction, such that online students performed worse than traditional students, and more so in quantitatively heavy classes. In order to encourage better performance, we design an intervention based on distributed practice theory, encouraging students to engage with material more frequently. However, this intervention fails, replicating the interaction in Study 2. For Study 3, we design a more extensive intervention based on social learning theory, asking teachers to employ a variety of tactics to boost their feedback and interpersonal contact with students. This new intervention is successful, with online students performing equally well regardless of the volume of quantitative content in the course. Findings are discussed in terms of their teaching implications and the need for more theory-based research.

Does a change in moral neutralization from early to mid-adolescence predict a change in delinquency?

Article

Feb 2023

Glenn D. Walters

A growth mixture modeling (GMM) analysis of neutralization scores in 1,830 youth across six waves of data revealed evidence of a three-class model in which moral neutralization either increased (low accelerating), decreased (high decelerating), or remained the same (moderate stable) over time. Controlling for age, sex, race, group assignment, and Wave 1 delinquency, an analysis of covariance revealed a significantly greater increase in Wave 6 delinquency in the moderate stable group than in the low accelerating group. When the average neutralization score was added as a control variable, the low accelerating group was highest, the moderate stable group next highest, and the high decelerating group lowest on Wave 6 delinquency. These results indicate that change in the pattern of neutralization scores (accelerating, stable, decelerating) may be as important as the level of moral neutralization at any particular point in time in determining the effect of neutralization on future delinquency.

Profiling employee psychological responses during restructuring and downsizing in the public sector: “Flourishers”, “Recoverers” and “Ambivalents”

Article

Sep 2022
PERS REV

Purpose This study aims to examine configurations of person-centered psychological change during organizational restructuring and downsizing in a public sector setting. Drawing on a social cognitive framework of organizational change the authors explore and identify the existence of different groups of employees who demonstrate varied responses (on commitment, engagement and anxiety) to restructuring and downsizing. Design/methodology/approach Surveys were collected from employees in three longitudinal waves (Time 1 N = 253; Time 2 N = 107; Time 3 N = 93, twelve months apart) at a UK public sector organization shortly before, during and after restructuring and downsizing. Findings Three classes of response emerged based on levels of and change in anxiety, organizational commitment and work engagement: a positive “Flourishers” profile was identified along with two relatively negative response profiles, labeled as “Recoverers” and “Ambivalents”. Higher levels of job control accounted for membership of the more positive response profile; higher structural uncertainty predicted membership of the most negative response group. Practical implications Using a person-centered approach, the authors form an understanding of different types of employee responses to downsizing; along with potential factors that help explain why groups of employees may exhibit certain psychological response patterns and may need to be managed differently during change. Thus, this approach provides greater understanding to researchers and managers of the varied impact that restructuring/downsizing has on the workforce. Originality/value To date there has been little research exploring employee responses to organizational restructuring and downsizing that has attempted to take a person-centered approach, which assumes population heterogeneity. Unlike variable centered approaches, this unique approach helps identify different patterns of employee responses to restructuring and downsizing.

Understanding salient trajectories and emerging profiles in the development of Chinese learners’ motivation: a growth mixture modeling approach

Article

Full-text available

Jun 2022
Int Rev Appl Ling

Based on the theoretical framework of the L2 Motivational Self System (L2MSS), the present study aims to make a methodological contribution to L2 motivation research. With the application of a novel growth mixture modeling (GMM) technique, the study depicted developmental trajectories of three motivational variables (ideal L2 self, ought-to L2 self, and L2 learning experience) of 176 Chinese tertiary-level students over a period of two semesters. Results showed two to three salient classes with typical developmental patterns for the three motivational variables respectively, with which the study gained fresh insights into the developmental processes of motivation beyond the individual level. Our study further established three main multivariate profiles of motivation characterized by a distinct combination of different motivational variables. The findings extend our understanding of motivational dynamics, providing a nuanced picture of emergent motivational trajectories systemically. Additionally, GMM has shown to be an effective and applicable method for the identification of salient patterns in motivation development, which leads to practical implications.

Developmental trajectories of self-esteem, the related predictors, and depression: A growth mixture modeling approach

Article

May 2022
J AFFECT DISORDERS

Background This study aimed to identify sub-trajectory groups of self-esteem among adults aged 19–64 years and the factors impacting latent classes, as well as to assess differences in symptoms of depression. Methods Research data from the Korea Welfare Panel Study were analyzed, including those from 8866 adults who participated in the 6th, 9th, 12th, and 15th waves. The growth mixture modeling analysis was used to identify latent classes of self-esteem trajectories. Results Three classes of self-esteem trajectories were identified. The majority of adults (88.0%) reported stable high self-esteem over time. A second class (low-level increasing: 7.7%) reported low levels of self-esteem, which gradually increased to high levels by the end of the study. A third group, medium-level decreasing (4.3%), reported medium self-esteem levels, which decreased to the lowest level by the end of the study. Limitations The factors identified in previous studies as those closely associated with self-esteem, such as personality, quality of life, and life satisfaction, were not considered in this study. Additionally, although the absence or presence of chronic disease was included in the health factors, no further investigation was made to identify the effects of different chronic diseases on the dependent and outcome variables. Conclusions These results suggest that interventions designed to prevent depression among adults who are older, unemployed, at risk of alcoholism, or dissatisfied with their health and relationships may be beneficial. This study identified a relationship between unstable self-esteem and the risk of depression.

New insights on growth trajectory in infants with complex congenital heart disease

Article

Sep 2022
J Pediatr Nurs

Purpose We aimed to describe the weight-for-age Z-score growth trajectory (WAZ-GT) of infants with complex congenital heart disease (cCHD) after neonatal cardiac surgery in the first 4 months of life and assess potential risk factors. Methods We utilized data from a previously reported trial of the REACH telehealth home monitoring (NCT01941667) program which evaluated 178 infants with cCHD from 2012 to 2017. Over the first 4 months of life, weekly infant weights were converted to WAZ. WAZ-GT classes were identified using latent class growth modeling. Multinomial logistic regression models were used to examine the associations between potential risk factors and WAZ-GT classes. Results Four distinct classes of WAZ-GT were identified: maintaining WAZ > 0, 14%; stable around WAZ = 0, 35%; partially recovered, 28%; never recovered, 23%. Compared with reference group “stable around WAZ=0,” we identified clinical and sociodemographic determinants of class membership for the three remaining groups. “Maintaining WAZ > 0” had greater odds of having biventricular physiology, borderline appetite, and a parent with at least a college education. “Partially recovered” had greater odds of hospital length of stay>14 days and being a single child in the household. “Never recovered” had greater odds hospital length of stay >14 and > 30 days, tube feeding at discharge, and low appetite. Conclusions This study described distinct classes of WAZ-GT for infants with cCHD early in infancy and identified associated determinants. Practice implications Findings from this study can be used in the identification of infants at risk of poor WAZ-GT and in the design of interventions to target growth in this vulnerable patient population.

Age and Work: Advances in Theory, Methods, and Practice

Book

Full-text available

Jan 2022

The edited volume Age and Work: Advances in Theory, Methods, and Practice presents a systematic collection of key advances in theory, methods, and practice regarding age(ing) and work. This cutting-edge collection breaks new ground by developing novel and useful theory, explaining underutilized but important methodological approaches, and suggesting original practical applications of emerging research topics. The book begins with a prologue by the World Health Organization’s unit head for aging and health, an introduction on the topic by the editors, and an overview of past, current, and future workforce age trends. Subsequently, the frst main section outlines theoretical advances regarding alternative age constructs (e.g., subjective age), intersectionality of age with gender and social class, paradoxical age-related actions, generational identity, and integration of lifespan theories. The second section presents methodological advances regarding behavioral assessment, age at the team and organizational levels, longitudinal and diary methods, experiments and interventions, qualitative methods, and the use of archival data. The third section covers practical advances regarding age and job crafting, knowledge exchange, the work/nonwork interface, healthy aging, and absenteeism and presenteeism, and organizational meta-strategies for younger and older workers. The book concludes with an epilogue by an eminent scholar in age and work. Written in a scientifc yet accessible manner, the book ofers a valuable resource for undergraduate and graduate students, academics in the felds of psychology and business, as well as practitioners working in the areas of human resource management and organizational development.

Discrete Latent Variable Models

Article

Mar 2022

We review the discrete latent variable approach, which is very popular in statistics and related fields. It allows us to formulate interpretable and flexible models that can be used to analyze complex datasets in the presence of articulated dependence structures among variables. Specific models including discrete latent variables are illustrated, such as finite mixture, latent class, hidden Markov, and stochastic block models. Algorithms for maximum likelihood and Bayesian estimation of these models are reviewed, focusing, in particular, on the expectation–maximization algorithm and the Markov chain Monte Carlo method with data augmentation. Model selection, particularly concerning the number of support points of the latent distribution, is discussed. The approach is illustrated by summarizing applications available in the literature; a brief review of the main software packages to handle discrete latent variable models is also provided. Finally, some possible developments in this literature are suggested. Expected final online publication date for the Annual Review of Statistics and Its Application, Volume 9 is March 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Mothers’ Well-Being in Families and Family Structure: Examining Constellations of Stressors Across Life Domains

Article

Full-text available

Oct 2021

Identifying conditions under which parents thrive is a key concern of family research. Prior research often focused on mothers’ well-being in single life domains, yet it is more likely to be shaped by stressors that stem directly from the parenting role and related stressors emerging from spillover processes into other domains. We therefore examine how stressors concerning mothers’ subjective, relational, and financial well-being accumulate and combine within subgroups of mothers and whether the likelihood to belong to these multidimensional subgroups varies by family structure. Using representative German data (N = 11,242), latent class analysis revealed four distinct subgroups of maternal well-being with varying exposure to financial, psychological, and relational stressors. Regression models showed that particularly single mothers were at risk to belong to the most vulnerable group with exposure to multiple stressors. Findings are discussed in light of persisting disparities among post-separation families despite demographic trends toward growing family diversity.

Clustering Trend Changes of Lung Cancer Incidence in Europe via the Growth Mixture Model during 1990–2016

Article

Full-text available

Apr 2021

Background: Lung cancer accounts for half of all deaths from cancer in Europe and has the highest incidence in Southern Europe. The current study aimed to cluster trend changes of lung cancer incidence in Europe via the growth mixture model. Methods: The dataset included incidence rates of female and male lung cancer per 100,000 for 42 European countries during 1990-2016 compiled from the Gapminder database. The growth mixture model was implemented to recognize different longitudinal patterns and estimate the linear trend of each pattern in Mplus 7.4 software. Results: The observed overall trend of incidence for female and male lung cancer was raising and falling, respectively, and Iceland was the only country with higher incidence of female versus male lung cancer in 2016. The growth mixture model suggests 3 main patterns for the trend of lung cancer incidence both for males and females. In male lung cancer, a sharp decreasing pattern was detected for 6 countries including Belarus, Estonia, Russia, Slovenia, Ukraine, and the United Kingdom; also, a moderately decreasing pattern was observed among the other countries. In female lung cancer, a moderate increasing trend was observed for 8 countries including the United Kingdom, Denmark, Hungary, Iceland, Ireland, Montenegro, Netherlands, and Norway; the other patterns were categorized into two clusters with slow increasing trends. Conclusion: Given the raising patterns in the incidence of lung cancer among European females, especially in the United Kingdom, Denmark, Hungary, Iceland, Ireland, Montenegro, Netherlands, and Norway, urgent effective measures are recommended to be taken.

Sex specific trajectories of central adiposity, lipid indices, and glucose level with incident hypertension: 12 years Follow-up in Tehran lipid and glucose study

Article

Full-text available

Feb 2021
J TRANSL MED

Aims To identify sex specific trajectories of waist circumference (WC),triglyceride (TG), total cholesterol (TC), high density lipoprotein cholesterol (HDL-C) and fasting plasma glucose (FPG) during adulthood and examine their associations with incident hypertension. Methods The cohort consisted of 5030 participants (2051 males) with at least 2 repeated measurement during a median of 12 years follow up. We identified trajectory groups using latent class growth mixture model, their association with hypertension was examined using multivariate Cox-regression analysis. Results We found 997 cases of hypertension (483 male). For both exposures, three distinct trajectory groups were identified in both genders. For WC, in women: low-increasing, 82.4%; high-stable, 13.4%; high-increasing, 4.2% and in men: stable, 94.6%; low-increasing, 3.6% and for high- increasing, 1.7%. For TG, in women: stable, 91.3%; decreasing, 5.9%; inverse U-shape, 2.8%; in men: stable, 89.7%; inverse U- shape, 6.2% and for decreasing, 4.1%. Regarding WC, high stable and high-increasing trajectories were associated with hypertension in the multivariate model [(hazard ratio (HR) = 1.66 (95% CI 1.26–2.20) and 2.78(1.79–3.60), respectively]. Among men, this association was shown only for the low-increasing trajectory [2.76: 1.49–5.10]. For TG, among women decreasing and inverse U-shape trajectories were significantly associated with hypertension in the multivariate model [1.32:1.01–1.76] and [2.23:1.58–3.23, respectively]. We did not find any impact of increasing trajectories of FPG and HDL-C on incident hypertension. Considering TC, all individuals followed a stable trajectory. Conclusion WC dynamic changes in both gender and TG trajectory among women were significantly associated with incident hypertension.

Effortful Control and Community Violence Exposure as Predictors of Developmental Trajectories of Self-serving Cognitive Distortions in Adolescence: A Growth Mixture Modeling Approach

Article

Full-text available

Nov 2020
J YOUTH ADOLESCENCE

Self-serving cognitive distortions are biased or rationalizing beliefs and thoughts that originate from the individual persistence into immature moral judgment stages during adolescence and adulthood, increasing the individual’s engagement in antisocial or immoral conducts. To date, the literature examining trajectories of cognitive distortions over time and their precursors is limited. This study sought to fill this gap, by examining effortful control and community violence exposure as individual and environmental precursors to developmental trajectories of cognitive distortions in adolescence. The sample consisted of 803 Italian high school students (349 males; Mage = 14.19, SD = 0.57). Three trajectories of cognitive distortions were identified: (1) moderately high and stable cognitive distortions (N = 311), (2) moderate and decreasing cognitive distortions (N = 363), and (3) low and decreasing cognitive distortions (N = 129). Both low effortful control and high exposure to community violence were significant predictors for moderately high and stable trajectory of cognitive distortions. These results point to the importance of considering moral development as a process involving multiple levels of individual ecology, highlighting the need to further explore how dispositional and environmental factors might undermine developmental processes of morality.

Sustained Effects From a School-Based Intervention Pilot Study for Children With Asthma and Anxiety

Article

Jun 2020

Children with chronic conditions (i.e., asthma) are more likely to have anxiety or depressive symptoms. Comorbid asthma and anxiety in children leads to increased morbidity, causing children to miss instructional time and parent/caregiver (CG) work absences. Asthma educational programs and mental health interventions have been developed, though no scalable programs integrate asthma education and mental health behavioral interventions for school-aged children. This study evaluated the sustained preliminary effects of an integrated asthma education and cognitive behavioral skills-building program, Creating Opportunities for Personal Empowerment for Asthma. Thirty-two children ages 8–12 years with asthma and symptoms of anxiety received the intervention. At 6-weeks postintervention, anxiety and CG-reported behavioral symptoms were significantly reduced, there were fewer missed doses of asthma controller medications, and asthma-related self-efficacy, personal beliefs, and the children’s understanding of asthma significantly increased. Most children ( n = 29, 91%) reported continued use of coping skills.

Impact of adolescent obesity on middle‐age health of women given data MAR

Article

Jun 2020

We analyze adolescent BMI and middle‐age systolic blood pressure (SBP) repeatedly measured on women enrolled in the Fels Longitudinal Study (FLS) between 1929 and 2010 to address three questions: Do adolescent‐specific growth rates in body mass index (BMI) and menarche affect middle‐age SBP? Do they moderate the aging effect on middle‐age SBP? Have the effects changed over historical time? To address the questions, we propose analyzing a growth curve model (GCM) that controls for age, birth‐year cohort, and historical time. However, several complications in the data make the GCM analysis nonstandard. First, the person‐specific adolescent BMI and middle‐age SBP trajectories are unobservable. Second, missing data are substantial on BMI, SBP, and menarche. Finally, modeling the latent trajectories for BMI and SBP, repeatedly measured on two distinct sets of unbalanced time points, are computationally intensive. We adopt a bivariate GCM for BMI and SBP with correlated random coefficients. To efficiently handle missing values of BMI, SBP, and menarche assumed missing at random, we estimate their joint distribution by maximum likelihood via the EM algorithm where the correlated random coefficients and menarche are multivariate normal. The estimated distribution will be transformed to the desired GCM for SBP that includes the random coefficients of BMI and menarche as covariates. We demonstrate unbiased estimation by simulation. We find that adolescent growth rates in BMI and menarche are positively associated with, and moderate, the aging effect on SBP in middle age, controlling for age, cohort, and historical time, but the effect sizes are at most modest. The aging effect is significant on SBP, controlling for cohort and historical time, but not vice versa.

Retiring: Role identity processes in retirement transition

Article

Mar 2020

Retirement transition has become a prolonged process of adaptation, including changes in role identity. However, there is a dearth of research on the process by which retirees cope with the role transition, including how pre‐retirement role identities shape the transition, the forms of identity work undertaken by retirees, and the unfolding nature of retirement transition. In an in‐depth qualitative examination of the transition process, we identify pre‐retirement role identity profiles based on work and nonwork role identities. We then examine how pre‐retirement role identities influence the transition process, including the nature of identity work in transition and the transition pathways demonstrated by retirees. Our findings provide insights into strengths and limitations afforded by pre‐retirement identities: They facilitate agentic coping in which retirees shed old and adopt new identities but also impose inertia and prolong the transition until identity crises force the retirees to undergo identity exploration and adoption of new identities.

Goodness-of-Fit Indices in Confirmatory Factor Analysis: The Effect of Sample Size

Article

Full-text available

May 1992

This investigation examined the influence of sample size on different goodness-of-fit indices used in confirmatory factor analysis (CFA). The first two data sets were derived from large normative samples of responses to a multidimensional self-concept instrument and to a multidimensional instrument used to assess students' evaluations of teaching effectiveness. In the third set, data were simulated and generated according to the model to be tested. In the fourth, data were simulated and generated according to a three-factor model that did not have a simple structure. Twelve fit indicators were used to assess goodness-of-fit in all CFAs. All analyses were conducted with the LISREL V package. One-way ANOVAs and a visual inspection of graphs were used to assess the sample size effect on each index for the four data sets. Despite the inconsistency of the findings with previous claims, the results are consistent with the observation that the amount of random, unexplained variance varies inversely with sample size. Appendices include a set of computed statements, an explanation and listing of the 12 goodness-of-fit indicators, a bibliography, a table of results, and figures showing sample size effect. (Author/LMO)

Cross-Cultural Training Effectiveness: A Review and a Theoretical Framework for Future Research

Article

Full-text available

Jan 1990

Increased internationalization in the economic, political, and social arenas has led to greater interpersonal cross-cultural contact. Because much of this contact has not been successful, cross-cultural training has been proposed by many scholars as a means of facilitating more effective interaction. A review of the cross-cultural training literature is presented, and it is determined that cross-cultural training in general is effective. The article also offers a theoretical framework based on social learning theory for understanding past research and for guiding future research; this is important because in this context variables seem to operate differently in international versus domestic areas.

Input-Based and Time-Based Models of International Adjustment: Meta-Analytic Evidence and Theoretical Extensions

Article

Full-text available

Apr 2005

Integrating work on international assignments and domestic stress, we conducted meta-analyses of over 50 determinants and consequences of expatriate adjustment using data from 8,474 expatriates in 66 studies. We also examined the trajectory of adjustment over time, and time as a moderator of adjustment effects. Results emphasize the centrality, criticality, and complexity of adjustment, strongly supporting Black, Mendenhall, and Oddou's (1991) model. Structural modeling of proposed model extensions showed that adjustment uniquely affects job satisfaction, withdrawal cognitions, and performance.

Is the Relationship Between Cognitive Ability and Job Performance Stable Over Time?

Article

Full-text available

Sep 1989

Kevin Murphy

Schmidt, Hunter, and Outerbridge's (1986) causal model of job performance suggests that cognitive ability is the most important cause of job performance and that the relationship between ability and performance is stable over time. Research on both the stability of skilled performance and the ability requirements of tasks is inconsistent with this model. Our article describes an alternative model that ascribes a critical importance to ability during stages where workers are learning new tasks and performing unfamiliar functions (i.e., transition stages) but less so during stages where workers are performing well-learned, familiar tasks (i.e., maintenance stages). The alternative model is shown to account for the findings explained by the Schmidt et al. (1986) model, as well as for findings that cannot be accounted for by their model.

Using Covariance Structure Analysis to Detect Correlates and Predictors of Individual Change Over Time

Article

Full-text available

Sep 1994

Recently, methodologists have shown how 2 disparate conceptual arenas (individual growth modeling and covariance structure analysis) can be integrated. The integration brings the flexibility of covariance analysis to bear on the investigation of systematic interindividual differences in change and provides another powerful data-analytic tool for answering questions about the relationship between individual true change and potential predictors of that change. The individual growth modeling framework uses a pair of hierarchical statistical models to represent (1) within-person true status as a function of time and (2) between-person differences in true change as a function of predictors. This article explains how these models can be reformatted to correspond, respectively, to the measurement and structural components of the general LISREL model with mean structures and illustrates, by means of worked example, how the new method can be applied to a sample of longitudinal panel data. (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Overall Fit in Covariance Structure Models: Two Types of Sample Size Effects

Article

Full-text available

Mar 1990

Kenneth A Bollen

Acontroversial area in covariance structure models is the assessment of overall model fit. Researchers have expressed concern over the influence of sample size on measures of fit. Many contradictory claims have been made regarding which fit statistics are affected by N. Part of the confusion is due to there being two types of sample size effects that are confounded. The first is whether N directly enters the calculation of a fit measure. The second is whether the means of the sampling distributions of a fit index are associated with sample size. These types of sample size effects are explained and illustrated with the major structural equation fit indices. In addition, the current debate on sample size influences is examined in light of this distinction. (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Goodness-of-Fit Indexes in Confirmatory Factor Analysis: The Effect of Sample Size. [Miscellaneous Article]

Article

Full-text available

May 1988

In confirmatory factor analysis, hypothesized models reflect approximations to reality so that any model can be rejected if the sample size is large enough. The appropriate question is whether the fit is adequate to support the model, and a large number of fit indexes have been proposed for this purpose. In the present article, we examine the influence of sample size on different fit indexes for both real and simulated data. Contrary to claims by Bentler and Bonett (1980), their incremental fit index was substantially affected by sample size. Contrary to claims by Joreskog and Sorbom (1981), their goodness-of-fit indexes provided by LISREL were substantially affected by sample size. Contrary to claims by Bollen (1986), his new incremental fit index was substantially affected by sample size. Hoelter's (1983) critical N index was also substantially affected by sample size. Of the more than 30 indexes considered, the Tucker-Lewis (1973) index was the only widely used index that was relatively independent of sample size. However, four new indexes based on the same form as the Tucker-Lewis index were also relatively independent of sample size., (C) 1988 by the American Psychological Association <2>

Practical Issues in Structural Equation Modeling

Article

Full-text available

Aug 1987

Practical problems that are frequently encountered in applications of covariance structure analysis are discussed and solutions are suggested. Conceptual, statistical, and practical requirements for structural modeling are reviewed to indicate how basic assumptions might be violated. Problems associated with estimation, results, and model fit are also mentioned. Various issues in each area are raised, and possible solutions are provided to encourage more appropriate and successful applications of structural modeling

A maximum likelihood method for latent class regression involving a censored dependent variable

Article

Full-text available

Feb 1993
PSYCHOMETRIKA

The standard tobit or censored regression model is typically utilized for regression analysis when the dependent variable is censored. This model is generalized by developing a conditional mixture, maximum likelihood method for latent class censored regression. The proposed method simultaneously estimates separate regression functions and subject membership in K latent classes or groups given a censored dependent variable for a cross-section of subjects. Maximum likelihood estimates are obtained using an EM algorithm. The proposed method is illustrated via a consumer psychology application.

Gender and Life Satisfaction in Retirement: An Assessment of the Male Model

Article

Full-text available

Feb 1996

Toni Calasanti

This study seeks to extend the research on retirement in two ways. First, using a nationally representative sample, I attempt to clarify the relationship between gender and life satisfaction in retirement by explicitly considering how gender structures preretirement employment experiences. Second, I ask whether the “male model” of life satisfaction in retirement can be used to assess women's life satisfaction in retirement. It is hypothesized that employment structures, through their influence on sources of work satisfaction and world view, influence the sources of life satisfaction in retirement. While this hypothesis is generally supported, gender still appears to define a context for the variables of the male model beyond that encompassed by employment structures. The sources and implications of this gender influence for future social gerontological research, particularly in the area of retirement, are discussed.

The Retirement Adjustment Process: Changes in the Well-Being of Male Retirees across Time

Article

Full-text available

Jun 1997

The purpose of this prospective study was to (1) evaluate the impact of retirement, (2) monitor the change in adjustment across time, and (3) identify the resources predictive of short- and long-term adjustment in retirement. A sample of 117 male retirees was assessed on indices of physical and psychological health, perceived control, retirement satisfaction, and life satisfaction at 2–4 months preretirement, 1 year post-, and 6–7 years postretirement. The results provided support for a positive impact of retirement, as retirees evidenced increases in well-being during the first year. There was also evidence of a retirement adjustment process, in that aspects of well-being (i.e., psychological health) changed from short- to long-term retirement. Finally, physical health, income, and voluntary retirement status predicted short-term adjustment, while internal locus of control was an additional resource for long-term adjustment. Changes in resources over time also differentially predicted short- and long-term adjustment (e.g., an increase in internal locus of control predicted an increase in activity satisfaction at 1 year but not at 6-7 years postretirement).

The temporal factor of change in stressor-strain relationships: A growth curve model on a longitudinal study in East Germany

Article

Full-text available

Jun 2000

Several theoretical models describing how stressor-strain relationships unfold in time (e.g., M. Frese & D. Zapf, 1988) were tested with a longitudinal study, with 6 measurement waves, using multivariate latent growth curve models. The latent growth curve model made it possible to decompose trait and state components of strains and to show that both trait and state components are affected by work stressors. Because East Germany constitutes a high-change environment, it is an appropriate setting in which to study the relationship between work stressors and strains. The results showed that both the state and trait components of strains were affected by stressors. For example, individual trends in uncertainty (stressor) and worrying (strain) were related, whereas worrying also showed a short-term relationship with time pressure (another stressor). In particular, the decomposition into trait and state components was only possible with the growth curve method that was used.

Retirement Transitions, Gender, and Psychological Well-Being A Life-Course, Ecological Model

Article

Full-text available

Jun 2002

This longitudinal study investigated the relationship between retirement transitions and subsequent psychological well-being using data on 458 married men and women (aged 50–72 years) who were either still in their primary career jobs, retired, or had just made the transition to retirement over the preceding 2 years. The findings show that the relationship between retirement and psychological well-being must be viewed in a temporal, life course context. Specifically, making the transition to retirement within the last 2 years is associated with higher levels of morale for men, whereas being "continuously" retired is related to greater depressive symptoms among men. The results suggest the importance of examining various resources and contexts surrounding retirement transitions (gender, prior level of psychological well-being, spouses' circumstance, and changes in personal control, marital quality, subjective health, and income adequacy) to understand the dynamics of the retirement transition and its relationship with psychological well-being.

General growth mixture modeling for randomized preventive interventions

Article

Full-text available

Jan 2003

This paper proposes growth mixture modeling to assess intervention effects in longitudinal randomized trials. Growth mixture modeling represents unobserved heterogeneity among the subjects using a finite-mixture random effects model. The methodology allows one to examine the impact of an intervention on subgroups characterized by different types of growth trajectories. Such modeling is informative when examining effects on populations that contain individuals who have normative growth as well as non-normative growth. The analysis identifies subgroup membership and allows theory-based modeling of intervention effects in the different subgroups. An example is presented concerning a randomized intervention in Baltimore public schools aimed at reducing aggressive classroom behavior, where only students who were initially more aggressive showed benefits from the intervention.

The Role of Coding Time in Estimating and Interpreting Growth Curve Models

Article

Full-text available

Mar 2004

The coding of time in growth curve models has important implications for the interpretation of the resulting model that are sometimes not transparent. The authors develop a general framework that includes predictors of growth curve components to illustrate how parameter estimates and their standard errors are exactly determined as a function of receding time in growth curve models. Linear and quadratic growth model examples are provided, and the interpretation of estimates given a particular coding of time is illustrated. How and why the precision and statistical power of predictors of lower order growth curve components changes over time is illustrated and discussed. Recommendations include coding time to produce readily interpretable estimates and graphing lower order effects across time with appropriate confidence intervals to help illustrate and understand the growth process.

Variation in the Drinking Trajectories of Freshmen College Students

Article

Full-text available

Apr 2005

F. K. Del Boca, J. Darkes, P. E. Greenbaum, and M. S. Goldman (2004) examined temporal variations in drinking during the freshmen college year and the relationship of several risk factors to these variations. Here, using the same data, the authors investigate whether a single growth curve adequately characterizes the variability in individual drinking trajectories. Latent growth mixture modeling identified 5 drinking trajectory classes: light-stable, light-stable plus high holiday, medium-increasing, highdecreasing, and heavy-stable. In multivariate predictor analyses, gender (i.e., more women) and lower alcohol expectancies distinguished the light-stable class from other trajectories; only expectancies differentiated the high-decreasing from the heavy-stable and medium-increasing classes. These findings allow for improved identification of individuals at risk for developing problematic trajectories and for development of interventions tailored to specific drinker classes.

Profiling Retirees in the Retirement Transition and Adjustment Process: Examining the Longitudinal Change Patterns of Retirees' Psychological Well-Being

Article

Full-text available

Mar 2007

Mo Wang

The author used role theory, continuity theory, and the life course perspective to form hypotheses regarding the different retirement transition and adjustment patterns and how different individual and contextual variables related to those patterns. The longitudinal data of 2 samples (n(1) = 994; n(2) = 1,066) from the Health and Retirement Survey were used. Three latent growth curve patterns of retirees' psychological well-being were identified as coexisting in the retiree samples through growth mixture modeling (GMM) analysis. On the basis of the latent class membership derived from GMM, retiree subgroups directly linked to different growth curve patterns were profiled with individual (e.g., bridge job status) and contextual variables (e.g., spouse working status). By recognizing the existence of multiple retiree subgroups corresponding to different psychological well-being change patterns, this study suggests that retirees do not follow a uniform adjustment pattern during the retirement process, which reconciles inconsistent previous findings. A resource perspective is further introduced to provide a more integrated theory for the current findings. The practical implications of this study are also discussed at both individual level and policy level.

Statistical Analysis with Missing Data

Article

Nov 1988
TECHNOMETRICS

General growth mixture modeling for randomized preventive interventions

Article

Jan 2002
BIOSTATISTICS

Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives

Article

Jan 1998
STRUCT EQU MODELING

Comment: Principal stratification approach to broken randomized experiments: A case study of school choice vouchers in New York City

Article

Jun 2003

Estimating the dimension of a model

Article

Jan 1978
ANN STAT

G. Schwartz

New Look at Statistical-Model Identification

Article

Jan 1974

H Akaike

Principal Stratification Approach to Broken Randomized Experiments: A Case Study of School Choice Vouchers in New York City [with Comment]

Article

Jun 2003

Latent Growth Modeling

Article

Jan 2002

David Chan

Comparisons of two statistical approaches to study growth curves: The multilevel model and the latent curve analysis

Article

Jan 1998

This article compares two statistical approaches for modeling growth across time. The two statistical approaches are the multilevel model (MLM) and latent curve analysis (LCA), which have been proposed to depict change or growth adequately. These two approaches were compared in terms of the estimation of growth profiles represented by the parameters of initial status and the rate of growth. A longitudinal data set obtained from a school‐based substance‐use prevention trial for adolescents was used to illustrate the similarities and differences between the two approaches. The results indicated that the two approaches yielded very compatible results. The parameter estimates associated with regression weights are the same, whereas those associated with variances and covariances are similar. The MLM approach is easier for model specification and is more efficient computationally in yielding results. The LCA approach, however, has the advantage of providing model evaluation, that is, an overall test of goodness of fit, and is more flexible in modeling and hypothesis testing as demonstrated in this study.

The Health and Retirement Study: An Overview

Article

Nov 1994

Juster FT, Suzman R. An overview of the health and retirement study. J Hum Res 40: S7-S56

Article

Jan 1995

This paper examines the scientific, public policy, and organizational background out of which the Health and Retirement Study emerged. It describes the evolution of the major parameters of the survey and the unique planning structure designed to ensure that the substantive insights of the research community were fully reflected in the content of the database, highlights key survey innovations contained in the HRS, and provides a preliminary assessment of the quality of the data as reflected by sample size, sample composition, response rate, and survey content. The paper also describes the several types of administrative data that are expected to be added to the HRS data: earnings and benefits from Social Security files, and health insurance and pension data from the employers of survey respondents.

Testing the number of components in a normal mixture

Article

Sep 2001
BIOMETRIKA

We demonstrate that, under a theorem proposed by Q.H. Vuong [Econometrica 57, No. 2, 307-333 (1989; Zbl 0701.62106)], the likelihood ratio statistic based on the Kullback-Leibler information criterion or the null hypothesis that a random sample is drawn from a k 0 -component normal mixture distribution against the alternative hypothesis that the sample is drawn from a k 1 -component normal mixture distribution is asymptotically distributed as a weighted sum of independent chi-squared random variables with one degree of freedom, under general regularity conditions. We report simulation studies of two cases where we are testing a single normal versus a two-component normal mixture and a two-component normal mixture versus a three-component normal mixture. An empirical adjustment to the likelihood ratio statistic is proposed that appears to improve the rate of convergence to the limiting distribution.

Have Multilevel Models Been Structural Equation Models All Along?

Article

Oct 2003

Patrick Curran

A core assumption of the standard multiple regression model is independence of residuals, the violation of which results in biased standard errors and test statistics. The structural equation model (SEM) generalizes the regression model in several key ways, but the SEM also assumes independence of residuals. The multilevel model (MLM) was developed to extend the regression model to dependent data structures. Attempts have been made to extend the SEM in similar ways, but several complications currently limit the general application of these techniques in practice. Interestingly, it is well known that under a broad set of conditions SEM and MLM longitudinal "growth curve" models are analytically and empirically identical. This is intriguing given the clear violation of independence in growth modeling that does not detrimentally affect the standard SEM. Better understanding the source and potential implications of this isomorphism is my focus here. I begin by exploring why SEM and MLM are analytically equivalent methods in the presence of nesting due to repeated observations over time. I then capitalize on this equivalency to allow for the extension of SEMs to a general class of nested data structures. I conclude with a description of potential opportunities for multilevel SEMs and directions for future developments.

Interindividual Differences in Intraindividual Changes in Proactivty During Organizational Entry: A Latent Growth Modeling Approach to Understanding Newcomer Adaptation

Article

Apr 2000

Intraindividual change over time is the essence of the change phenomenon hypothesized to occur in the individual newcomer adaptation process. Many important adaptation questions cannot be answered without an adequate conceptualization and assessment of intraindividual change. Using a latent growth modeling approach to data collected from 146 doctoral program newcomers over 4 repeated measurements spaced at 1-month intervals, the authors explicitly modeled interindividual differences in intraindividual changes in newcomer proactivities (information seeking, relationship building) and proximal adaptation outcomes (task mastery, role clarity, social integration) during organizational entry. Results indicated that changes in proactivity may be related to newcomer characteristics and adaptation outcomes in interesting ways that have not been previously examined.

Power of Latent Growth Modeling for Detecting Group Differences in Linear Growth Trajectory Parameters

Article

Jul 2003

Xitao Fan

This simulation study focused on the power for detecting group differences in linear growth trajectory parameters within the framework of structural equation modeling (SEM) and compared the latent growth modeling (LGM) approach to the more traditional repeated-measures analysis of variance (ANOVA) approach. Several patterns of group differences in linear growth trajectories were considered. SEM growth modeling consistently showed higher statistical power for detecting group differences in the linear growth slope than repeated-measures ANOVA. For small group differences in the growth trajectories, large sample size (e.g., N > 500) would be required for adequate statistical power. For medium or large group differences, moderate or small sample size would be sufficient for adequate power. Some future research directions are discussed.

The Conceptualization and Analysis of Change Over Time: An Integrative Approach Incorporating Longitudinal Mean and Covariance Structures Analysis (LMACS) and Multiple Indicator Latent Growth Modeling (MLGM)

Article

Oct 1998

David Chan

The concept of change over time is fundamental to many phenomena investigated in organizational research. This didactically oriented article proposes an integrative approach incorporating longitudinal mean and covariance structures analysis and multiple indicator latent growth modeling to aid organizational researchers in directly addressing fundamental questions concerning the conceptualization and analysis of change over time. The approach is illustrated using a numerical example involving several organizationally relevant variables. Advantages, limitations, and extensions of the approach are discussed.

Goodness-of-fit indexes in confirmatory factor analysis: The effect of sample size

Article

May 1988
PSYCHOL BULL

In confirmatory factor analysis, hypothesized models reflect approximations to reality so that any model can be rejected if the sample size is large enough. The appropriate question is whether the fit is adequate to support the model, and a large number of fit indexes have been proposed for this purpose. In the present article, we examine the influence of sample size on different fit indexes for both real and simulated data. Contrary to claims by Bentler and Bonett (1980), their incremental fit index was substantially affected by sample size. Contrary to claims by Joreskog and Sorbom (1981), their goodness-of-ftt indexes provided by LISREL were substantially affected by sample size. Contrary to claims by Bollen (1986), his new incremental fit index was substantially affected by sample size. Hoelter's (1983) critical N index was also substantially affected by sample size. Of the more than 30 indexes considered, the Tucker-Lewis (1973) index was the only widely used index that was relatively independent of sample size. However, four new indexes based on the same form as the Tucker-Lewis index were also relatively independent of sample size. The purpose of the present investigation was to examine the influence of sample size on goodness-of-f it indicators used in confirmatory factor analysis (CFA). Although the present inves- tigation was limited to CFA, the problems, issues, and most of the results generalize to the analysis of covariance structures. The advantages of CFA are well-known, and numerous intro- ductions to the LISREL approach used in the present investiga- tion are available elsewhere (e.g., Bagozzi, 1980; Joreskog &

Second-generation structural equation modeling with a combination of categorical and continuous latent variables: New opportunities for latent class-latent growth modeling.

Chapter

Jan 2001

B. O. Muthen

Finite mixture model selection with psychometrics application

Article

Mar 1999

Chih-Chien Yang

Recent statistical advances, for example, Bandeen-Roche, Miglioretti, Zeger and Rathouz (1997); Jedidi, Jagpal and DeSarbo (1997); and Wang and Puterman (1998) have made it feasible to fit finite mixture models in a wide range of applications. With a collection of plausible models for a given data set, problems of model selection arise. Selection among finite mixture models often involves a choice among models with different number of latent classes. As pointed out by several researchers (Everitt, 1981; Aitkin & Rubin, 1985), this can be problematic for traditional likelihood ratio tests because of the unknown distribution of the likelihood ratio. Therefore, it is desirable to investigate information criteria for alternative model selection procedures. Lin and Dayton (1997) showed that the accuracy of widely used criteria, AIC/BIC/CAIC, can be dissatisfying for complex finite mixture models. To improve the accuracy, a relatively newer criterion (Draper, 1995) as well as an adjustment of standard criteria have been suggested. The aim of this study is to investigate the accuracy of information criteria and standard likelihood ratio testing methods for various finite mixture models. Monte Carlo studies in this dissertation show that improvements of accuracy by using the adjusted information criteria are considerable. Guidelines are also provided for practical use of these model fit indices in estimating the number of latent classes, especially when different sample sizes, parameter structures and model complexities are involved. Following these guidelines, the optimum accuracy rates of these model fit indices can be achieved; moreover, situations that can cause low accuracy can be avoided. Finite mixture models for an alcohol dependence and abuse study using datasets from the National Longitudinal Surveys of Youth (NLSY) are also illustrated. Both model selections and interpretations for the finite mixture models are emphasized using these examples. (PsycINFO Database Record (c) 2012 APA, all rights reserved)

The substantive nature of performance variability: Predicting interindividual differences in intraindividual performance

Article

Dec 2006
PERS PSYCHOL

The nature of intraindividual performance variability over time, along with individual difference predictors of such variability, was examined using latent growth curve methodology. Quarterly sales performance for a sample of securities analysts (n= 303) was measured at 8 times. Average intraindividual performance approximated a basic “learning” curve, although there were considerable individual differences in each of the latent performance growth parameters. Individual difference predictors from a biodata inventory were moderately related to these latent growth parameters. Theoretical and practical implications of performance variability for personnel selection are also discussed.

Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives

Article

Jan 1999

This article examines the adequacy of the “rules of thumb” conventional cutoff criteria and several new alternatives for various fit indexes used to evaluate model fit in practice. Using a 2‐index presentation strategy, which includes using the maximum likelihood (ML)‐based standardized root mean squared residual (SRMR) and supplementing it with either Tucker‐Lewis Index (TLI), Bollen's (1989) Fit Index (BL89), Relative Noncentrality Index (RNI), Comparative Fit Index (CFI), Gamma Hat, McDonald's Centrality Index (Mc), or root mean squared error of approximation (RMSEA), various combinations of cutoff values from selected ranges of cutoff criteria for the ML‐based SRMR and a given supplemental fit index were used to calculate rejection rates for various types of true‐population and misspecified models; that is, models with misspecified factor covariance(s) and models with misspecified factor loading(s). The results suggest that, for the ML method, a cutoff value close to .95 for TLI, BL89, CFI, RNI, and Gamma Hat; a cutoff value close to .90 for Mc; a cutoff value close to .08 for SRMR; and a cutoff value close to .06 for RMSEA are needed before we can conclude that there is a relatively good fit between the hypothesized model and the observed data. Furthermore, the 2‐index presentation strategy is required to reject reasonable proportions of various types of true‐population and misspecified models. Finally, using the proposed cutoff criteria, the ML‐based TLI, Mc, and RMSEA tend to overreject true‐population models at small sample size and thus are less preferable when sample size is small.

Fit Indices in Covariance Structure Modeling: Sensitivity to Underparameterized Model Misspecification

Article

Dec 1998
PSYCHOL METHODS

This study evaluated the sensitivity of maximum likelihood (ML)-, generalized least squares (GLS)-, and asymptotic distribution-free (ADF)-based fit indices to model misspecification, under conditions that varied sample size and distribution. The effect of violating assumptions of asymptotic robustness theory also was examined. Standardized root-mean-square residual (SRMR) was the most sensitive index to models with misspecified factor covariance(s), and Tucker-Lewis Index (1973; TLI), Bollen's fit index (1989; BL89), relative noncentrality index (RNI), comparative fit index (CFI), and the ML- and GLS-based gamma hat, McDonald's centrality index (1989; Mc), and root-mean-square error of approximation (RMSEA) were the most sensitive indices to models with misspecified factor loadings. With ML and GLS methods, we recommend the use of SRMR, supplemented by TLI, BL89, RNI, CFI, gamma hat, Mc, or RMSEA (TLI, Mc, and RMSEA are less preferable at small sample sizes). With the ADF method, we recommend the use of SRMR, supplemented by TLI, BL89, RNI, or CH. Finally, most of the ML-based fit indices outperformed those obtained from GLS and ADF and are preferable for evaluating model fit. (PsycINFO Database Record (c) 2012 APA, all rights reserved)

An introduction to latent growth curve modeling

Article

Mar 2004
BEHAV THER

Over the past 3 decades we have witnessed an increase in the complexity of theoretical models that attempt to explain development in a number of behavioral domains. The conceptual movement to examine behavior from both developmental and contextual perspectives parallels recent methodological advances in the analysis of change. These new analysis techniques have fundamentally altered how we conceptualize and study change, and have prompted researchers to identify larger frameworks to integrate knowledge. One such framework is latent growth modeling. This article presents a basic latent growth modeling approach for analyzing repeated measures data and delineates several of its extensions, including analyses for multiple populations, accelerated designs, multivariate associative models, and a framework for sample size selection and power estimation.

Finite Mixture Model

Article

Jan 2000
TECHNOMETRICS

Estimating the Dimension of a Model

Article

Mar 1978
ANN STAT

Gideon E. Schwarz

The problem of selecting one of a number of models of different dimensions is treated by finding its Bayes solution, and evaluating the leading terms of its asymptotic expansion. These terms are a valid large-sample criterion beyond the Bayesian context, since they do not depend on the a priori distribution.

Latent Curve Analysis

Article

Feb 1990
PSYCHOMETRIKA

As a method for representing development, latent trait theory is presented in terms of a statistical model containing individual parameters and a structure on both the first and second moments of the random variables reflecting growth. Maximum likelihood parameter estimates and associated asymptotic tests follow directly. These procedures may be viewed as an alternative to standard repeated measures ANOVA and to first-order auto-regressive methods. As formulated, the model encompasses cohort sequential designs and allow for period or practice effects. A numerical illustration using data initially collected by Nesselroade and Baltes is presented.

Application of Model-Selection Criteria to Some Problems in Multivariate Analysis, Psychometrika, Vol.52, No.3, Special Section, 333-343

Article

Feb 1987
PSYCHOMETRIKA

Stanley Sclove

A review of model-selection criteria is presented, with a view toward showing their similarities. It is suggested that some problems treated by sequences of hypothesis tests may be more expeditiously treated by the application of model-selection criteria. Consideration is given to application of model-selection criteria to some problems of multivariate analysis, especially the clustering of variables, factor analysis and, more generally, describing a complex of variables.

Retirement as a Critical Life Event: The Differential Experiences of Women and Men

Article

Jan 1988
RES AGING

This article examines the experience of adaptation to retirement among a sample of women (n = 124) and men (n = 176) retired an average of three years and living in an urban area of Ontario, Canada. The impact of retirement as a life event relative to other life experiences was examined, and found to be distinctly less critical than previous research would suggest.

Consequences of Retirement Activities for Distress and the Sense of Personal Control

Article

Jan 1999

We compare retirement with full-time employment on four forms of engaging activity and examine the consequences of retirement activities for the sense of control and psychological distress. We use a 1995 U.S. national telephone probability sample of 2,592 respondents with an oversample of persons aged sixty and older. In comparison to the activities of full-time employees, those of retirees are more alienating on some aspects but more engaging on others. Retiree activities are more routine, provide less of a chance to learn new things, provide less positive social interaction with others, and they are especially unlikely to involve problem-solving. However, retirees' activities are also equally enjoyable and more autonomous compared to those of full-time workers. Autonomous activities, fulfilling activities which are enjoyable and provide the opportunity to learn new things, and integrated activities are all positively associated with a sense of control and negatively associated with psychological distress. However, solving problems is associated with both high levels of control and high levels of distress. Retirees have a significantly lower sense of control than do full-time employees, in large part because of the characteristics of their daily activities. At the same time, retirees do not have significantly higher levels of psychological distress.

Finite Mixture Modeling with Mixture Outcomes Using the EM Algorithm

Article

Jul 1999

This paper discusses the analysis of an extended finite mixture model where the latent classes corresponding to the mixture components for one set of observed variables influence a second set of observed variables. The research is motivated by a repeated measurement study using a random coefficient model to assess the influence of latent growth trajectory class membership on the probability of a binary disease outcome. More generally, this model can be seen as a combination of latent class modeling and conventional mixture modeling. The EM algorithm is used for estimation. As an illustration, a random-coefficient growth model for the prediction of alcohol dependence from three latent classes of heavy alcohol use trajectories among young adults is analyzed.

Model modifications in covariance structure analysis: The problem of capitalization on chance

Article

Jun 1992

In applications of covariance structure modeling in which an initial model does not fit sample data well, it has become common practice to modify that model to improve its fit. Because this process is data driven, it is inherently susceptible to capitalization on chance characteristics of the data, thus raising the question of whether model modifications generalize to other samples or to the population. This issue is discussed in detail and is explored empirically through sampling studies using 2 large sets of data. Results demonstrate that over repeated samples, model modifications may be very inconsistent and cross-validation results may behave erratically. These findings lead to skepticism about generalizability of models resulting from data-driven modifications of an initial model. The use of alternative a priori models is recommended as a preferred strategy.

A New Look At The Statistical Model Identification

Article

Jan 1975

Hirotugu Akaike

The history of the development of statistical hypothesis testing in time series analysis is reviewed briefly and it is pointed out that the hypothesis testing procedure is not adequately defined as the procedure for statistical model identification. The classical maximum likelihood estimation procedure is reviewed and a new estimate minimum information theoretical criterion (AIC) estimate (MAICE) which is designed for the purpose of statistical identification is introduced. When there are several competing models the MAICE is defined by the model and the maximum likelihood estimates of the parameters which give the minimum of AIC defined by AIC = (-2)log-(maximum likelihood) + 2(number of independently adjusted parameters within the model). MAICE provides a versatile procedure for statistical model identification which is free from the ambiguities inherent in the application of conventional hypothesis testing procedure. The practical utility of MAICE in time series analysis is demonstrated with some numerical examples.

Mplus (Version 3.1) [Computer software

Jan 2004

L Muthén
B Muthén

Muthén, L., & Muthén, B. (2004). Mplus (Version 3.1) [Computer software].

Growth Mixture Modeling

Abstract

No full-text available

Recommended publications

Learning Global Models Based on Distributed Data Abstractions.

Estimating nonlinear change models in heterogeneous populations when class membership is unknown: De...

Automatic Kidney Segmentation Using Gaussian Mixture Model on MRI Sequences

Text-Independent Speaker Verification Using Variational Gaussian Mixture Model