Latent Curve Modeling Research Articles

Finding the Optimal Number of Persons (N) and Time Points (T) for Maximal Power in Dynamic Longitudinal Models Given a Fixed Budget

Planning longitudinal studies can be challenging as various design decisions need to be made. Often, researchers are in search for the optimal design that maximizes statistical power to test certain parameters of the employed model. We provide a user-friendly Shiny app OptDynMo available at https://shiny.psychologie.hu-berlin.de/optdynmo that helps to find the optimal number of persons (N) and the optimal number of time points (T) for which the power of the likelihood ratio test (LRT) for a model parameter is maximal given a fixed budget for conducting the study. The total cost of the study is computed from two components: the cost to include one person in the study and the cost for measuring one person at one time point. Currently supported models are the cross-lagged panel model (CLPM), factor CLPM, random intercepts cross-lagged panel model (RI-CLPM), stable trait autoregressive trait and state model (STARTS), latent curve model with structured residuals (LCM-SR), autoregressive latent trajectory model (ALT), and the latent change score model (LCS).

The Hitchhiker’s guide to longitudinal models: A primer on model selection for repeated-measures methods

Longitudinal data are becoming increasingly available in developmental neuroimaging. To maximize the promise of this wealth of information on how biology, behavior, and cognition change over time, there is a need to incorporate broad and rigorous training in longitudinal methods into the repertoire of developmental neuroscientists. Fortunately, these models have an incredibly rich tradition in the broader developmental sciences that we can draw from. Here, we provide a primer on longitudinal models, written in a beginner-friendly (and slightly irreverent) manner, with a particular focus on selecting among different modeling frameworks (e.g., multilevel versus latent curve models) to build the theoretical model of development a researcher wishes to test. Our aims are three-fold: (1) lay out a heuristic framework for longitudinal model selection, (2) build a repository of references that ground each model in its tradition of methodological development and practical implementation with a focus on connecting researchers to resources outside traditional neuroimaging journals, and (3) provide practical resources in the form of a codebook companion demonstrating how to fit these models. These resources together aim to enhance training for the next generation of developmental neuroscientists by providing a solid foundation for future forays into advanced modeling applications.

Bidirectional associations between sleep duration and emotion dysregulation across adolescence.

Throughout adolescence, both sleep and emotion regulation abilities undergo significant developmental changes. The maturational systems that govern sleep and emotion regulation are closely intertwined leading many researchers to posit a mutually reinforcing relationship. Although there is support for a bidirectional relationship among adults, empirical support for reciprocal relationships among adolescents is lacking. Given the notable developmental changes and instability that occur throughout adolescence, this is a critical period to examine whether sleep and emotion regulation abilities may be reciprocally related. Using a latent curve model with structured residuals, this study examined within-person reciprocal associations between sleep duration and emotion dysregulation among 12,711 Canadian adolescents (Mage = 14.30 years; 50% female). Participants self-reported their sleep duration and emotion dysregulation each year for 3 years beginning in Grade 9. After accounting for underlying developmental trajectories, the results did not support a bidirectional relationship between sleep duration and emotion dysregulation from one year to the next. However, there was evidence of contemporaneous associations between the residuals at each wave of assessment (r = -.12) such that less sleep than expected was concurrently associated with higher-than-expected deviations in emotion dysregulation, or, conversely, that reporting greater emotion dysregulation than expected was associated with lower-than-expected sleep duration. In contrast to previous findings, the between-person associations were not supported. Taken together, these results indicate that the relationship between sleep duration and emotion dysregulation is primarily a within-person process rather than reflecting differences between individuals and likely operates on a more proximal timescale. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Genomewide analysis of Episodic Memory trajectories in The Mexican Health and Aging Study (MHAS)

AbstractBackgroundGenetic characterization of age‐related memory changes can help identify population subgroups at‐risk of memory decline and dementia. The majority of genetic studies examining memory trajectories in old age have used data from non‐Hispanic White populations. To our knowledge, there are no reports of genetic factors underlying memory function over time in the Mexican population.MethodUsing a previously described latent curve model approach, we estimated episodic memory trajectories in a longitudinal sample from the Mexican Health and Aging Study. We conducted GWAS analyses in a sub‐sample of 6,343 participants. Analyses were stratified by memory stability (Stables, n = 4,437) and APOE status. Three independent cohorts were used for replication purposes: two Non‐Hispanic White samples from the Alzheimer’s Disease Genetic Consortium and the Religious Orders Study, and a Caribbean‐Hispanic sample from the Washington Heights Inwood Community Aging Project.ResultThe strongest genome‐wide significant association was found for an intronic variant in the NR2F1‐AS1 gene (rs555528825, p = 1.8×10−9) among APOE non‐E4 carriers in the “Stables” group. NR2F1 gene is an evolutionarily conserved long non‐coding RNA that enhances neuronal cell maturation and regulates transcription of neuronal genes. SNP variants located less than 50Kb apart from the identified signal showed also nominally significant associations in the three replication datasets (p = 3.9×10−4, p = 1.8×10−4, and p = 0.006 respectively). Additional replication efforts using the UK biobank recourse are ongoing.ConclusionOur study nominates novel genetic variants associated with longitudinal changes in episodic memory performance using data from the Mexican Health and Aging Study, a representative and comprehensive longitudinal study that includes genetic data on a sub‐sample of participants 50 years and older.

Studying Between-Subject Differences in Trends and Dynamics: Introducing the Random Coefficients Continuous-Time Latent Curve Model with Structured Residuals

The recently proposed continuous-time latent curve model with structured residuals (CT-LCM-SR) addresses several challenges associated with longitudinal data analysis in the behavioral sciences. First, it provides information about process trends and dynamics. Second, using the continuous-time framework, the CT-LCM-SR can handle unequally spaced measurement occasions and describes processes independently of the length of the time intervals used in a given study. Third, it is a hierarchical model. Thus, multiple subjects can be analyzed simultaneously. However, subjects might also differ in dynamics and trends. Therefore, in the present paper, we extend the CT-LCM-SR to capture these differences as well. This newly proposed random coefficients continuous-time latent curve model with structured residuals (RC-CT-LCM-SR) is introduced theoretically and technically. Additionally, we provide an illustrative example with data from the Health and Retirement Study (HRS), and we show how the RC-CT-LCM-SR can be used to study multiple sources of between-subject differences over time.

Assessing cross-lagged associations between depression, anxiety, and binge drinking in the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) study

BackgroundBetween 20 % and 30 % of teens suffer from depression or anxiety before reaching adulthood, and up to half also use or misuse alcohol. Although theories suggest bidirectional links between harmful alcohol use (e.g., binge drinking) and internalizing symptoms (i.e., depression and anxiety), empirical evidence to-date has been mixed. Systematic reviews have attributed mixed findings to limitations in study design, such as the utilization of between-person analyses and the focus on unidirectional effects. The goal of this study was to address these limitations by assessing bidirectional within-person associations between internalizing symptoms and binge drinking over the course of 5 years in the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) sample, a large cohort recruited at ages 12–21 and followed annually on substance use and psychiatric functioning. MethodsWe used latent curve models with structured residuals to examine within-person lagged associations between depression, anxiety, and past month counts of binge drinking using NCANDA data (N = 831). Analyses were supplemented with post-hoc power simulations. ResultsWe found marginal evidence linking binge drinking with subsequent depression symptoms one year later among females. We found no evidence that depression or anxiety predicted subsequent binge drinking despite sufficient power. ConclusionsSocial and cognitive consequences of binge drinking may predict later depression symptoms in adolescence and young adulthood for young women, though there was little evidence favoring self-medication models for binge drinking. We note several moderating variables and common factor mechanisms that may better explain this link.

Development and practice of psychoeducation to maintain and promote social support exchanges with family and friends

This study aimed to develop a psychoeducational practice to maintain and promote support exchanges with family and friends for college students. We developed four sessions, based on previous studies, and they were conducted once weekly as part of a psychology class at a women’s university in Tokyo. Through the psychoeducational practice, psycho-behavioral indicators regarding social support exchanges—self-efficacy regarding support exchanges, perceived and provision of social support from family and friends, and social skills—were measured thrice (pre- and post-psychoeducation, and at the 1-month follow-up). The intervention group (n = 33) received psychoeducational sessions and the contrast group (n = 35) included students who did not take this class. To describe the measurement changes over three points in each group, a latent curve model was tested. The results showed that some indicators in the intervention group were more likely to improve after attending the sessions when compared to the contrast group. Concerning perceived social support from friends, a temporary improvement was only observed post-psychoeducation. As such, further examinations with additional data are suggested.

Longitudinal trajectories of approaches to learning among Chinese preschoolers: Family predictors and academic outcomes

Considering the beneficial effects associated with approaches to learning (ATL) and the need to better understand its occurrence and development, this study examined the trajectories of ATL, as well as family predictors and academic outcomes in a longitudinal study in China. It followed 417 preschoolers (55.9% males; M=42 months, SD=4 months) for three years and noted three main findings. First, latent curve modeling revealed an initially above-average level and linear increase trajectories of ATL with interindividual variations. Second, although socioeconomic status (SES) predicted children's ATL in early kindergarten, home learning environment (HLE) trajectories had a uniquely positive effect on ATL trajectories. Third, ATL trajectories predicted an improvement in language skills and cognitive skills. These results help to effectively guide ATL for improving children's academic outcomes.

Incivility and psychological safety in youth sport: the reciprocal effects and its impact on well-being and social outcomes

ABSTRACT Youth can often be the subject of rude and discourteous behaviors in their sport setting, given their susceptibility to the social environments. Incivility refers to insensitive behavior that exhibits a lack of respect for others, namely, disrespectful and rude behaviors. Incivility is a significant issue in youth sport since it negatively influences teams and individuals. The present study aims to investigate 1) how incivility and psychological safety are associated with each other over time and 2) the influence of the initial value of psychological safety and the change in psychological safety on youth athletes’ well-being and social outcomes. Three-wave time-lagged data collection was employed, and the present study included 283 youth athletes who completed the survey three times. The hypotheses were tested in SEM with cross-lagged panel and growth latent curve modeling. The results showed that coach and teammate incivility were significantly associated with the change in psychological safety. In contrast, the initial value of psychological safety was a significant antecedent of the subsequent coach and teammate incivility, well-being, and social outcomes. Lastly, the change in psychological safety was significantly associated with youth athletes’ well-being and social outcomes. The findings suggest that incivility and psychological safety were reciprocally associated, and psychological safety rather than incivility was a significant predictor of youths’ well-being and social outcomes. The present study found a mechanism underlying the relationship between incivility, psychological safety, and essential outcomes (i.e., well-being and social outcomes) in youth sport.

Does pressure to gain social media attention have consequences for adolescents’ friendship closeness and mental health? A longitudinal examination of within-person cross-lagged relations

Many adolescents face pressure when it comes to securing social media attention in the form of views, comments and/or likes on their posted content. The purpose of this study was to examine how this pressure impacts adolescents’ current relationships with friends in addition to their mental health over time. Participants were Canadian adolescents (Time 1 n = 345; Mage = 17.29; 80.6% female) who reported on their felt pressure to gain social media attention, friendship closeness and internalizing symptoms in 3 surveys approximately 4 months apart (from August 2020 to June 2021). We used latent curve modeling with structured residuals (LCM-SR) to model the lagged relations between the aforementioned variables, while also controlling for time spent on social media and number of likes received. LCM-SR builds on multivariate latent curve modeling and autoregressive latent trajectory and allows for the simultaneous testing of between- and within-person stability and change over time. In line with our first hypothesis, results demonstrated that at time points when adolescents experienced more pressure to gain social media attention than usual, their friendship closeness decreased at the next time point. Social media pressure was not a significant predictor of internalizing symptoms, however. Results emphasize the importance of teasing apart within- and between-person effects when examining impacts of adolescent social media use. They also highlight the importance of targeting felt pressures to gain social media attention in order to support healthy adolescent relationships.

The reciprocal relations between well-being and maternal and peer warmth in adolescents involved in the juvenile justice system.

Although justice system involvement increases the risk of negative outcomes for adolescents, many justice-involved youth desist from crime as adults (Sampson & Laub, 2005). There are few studies examining predictors of positive development in justice-involved adolescents. In the current study, we assess the influence of maternal and peer warmth on the development of well-being in adolescents involved in the US justice system over the course of 5years. Participants included 1216 adolescent males who experienced their first arrest. Interviews were given every year for 5 years. Well-being was measured using the EPOCH questionnaire (Kern et al., 2016) and relationship warmth was measured using a scale adapted from Conger et al. (1994). Hypotheses were tested using latent curve models with structured residuals. Baseline levels of well-being were associated with maternal (β = 0.49, p < .001) and peer warmth, β = 0.52, p < .001. When an individual's maternal warmth was higher than predicted given their maternal warmth trajectory, their subsequent well-being was higher than expected given their well-being trajectory, b = 0.07, p < .001. When an individual's peer warmth was higher than predicted, their subsequent well-being was higher than expected, b = 0.06, p < .001. These relations were reciprocal, such that well-being also predicted increased maternal and peer warmth. These findings suggest that increasing maternal or peer warmth may have cascading effects on the well-being of justice-involved adolescents. Interventions for justice-involved youth may benefit from targeting factors that increase positive development for these youth.

Confidence in the therapist and confidence in the treatment predict symptomatic improvement week by week in therapy: A latent curve modeling approach.

Previous research suggests that common relationship factors are composed of two overarching factors, "Confidence in the therapist" and "Confidence in the treatment." The aim of this naturalistic process-outcome study was to investigate the reciprocal relationships between these two constructs and patients' symptom level across treatment. The sample consisted of 587 patients who were admitted to an inpatient program and treated with psychotherapy for a range of mental health disorders, such as chronic depression, anxiety disorders, and eating disorders. Our data consisted of weekly measures of symptomatic distress (Patient Health Questionnaire) and the common relationship factors were measured weekly using a newly developed scale. Latent curve modeling with structured residuals was used to investigate the between- and within effects of week-to-week changes in the two components as predictors of subsequent symptom level. An increase in both relationship factors predicted a decrease in subsequent levels of symptoms at the within-patient level, and the other way around, but the two relationship factors did not systematically relate to one another at the within-patient level over the course of treatment. Our findings indicate that patients' perceptions of the therapist as a person and their appraisal of the treatment, are important, different predictors of therapeutic change. Furthermore, they support prior research demonstrating a reciprocal relationship between common relationship factors and symptomatic distress and add to existing common factor theory by exploring the role of two central relationship dimensions and using a method which examines reciprocal relationships and within-patient effects simultaneously. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

A multilevel structured latent curve model for disaggregating student and school contributions to learning

A multilevel structured latent curve model for disaggregating student and school contributions to learning

Equivalent approaches to dealing with unobserved heterogeneity in cross-lagged panel models? Investigating the benefits and drawbacks of the latent curve model with structured residuals and the random intercept cross-lagged panel model.

Panel models in structural equation modeling that combine static and dynamic components make it possible to investigate reciprocal relations while controlling for time-invariant unobserved heterogeneity. Recently, the latent curve model with structured residuals and the random-intercept cross-lagged panel model were suggested as "residual-level" versions of the more traditional autoregressive latent trajectory and dynamic panel models, respectively. Their main benefit is that they allow for a more straightforward interpretation of the trajectory factors. It is not widely known, however, that the residual-level models place potentially strong assumptions on the initial conditions-that is, the process that was occurring before the observation period began. If the process under investigation is not both stationary and at equilibrium then the residual-level models are not appropriate. They then do not control for all time-invariant unobserved heterogeneity and can result in biased cross-lagged and autoregressive estimates. I demonstrate this using the problem behavior of cigarette smoking among adolescents: Because the mean and variance of this process changes as a young person's smoking behavior develops, early stages of this process should not be examined using the residual-level models. This issue potentially exists for a wide variety of psychological and sociological subjects, essentially whenever the process under investigation is changing over the course of the observation period. This article discusses strategies to help researchers decide which model to use when, and compares some of their relative advantages and drawbacks. An amendment to the residual-level models is suggested in which the latent individual effects are allowed to covary with the initial residuals. This makes the residual-level models robust to violations of the assumptions surrounding the initial conditions, while retaining their other beneficial aspects. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Trajectories of Perceived Technological Impairment and Psychological Distress in Adolescents.

Fears that digital technologies harm adolescents’ mental health abound; however, existing research is mixed. This study examined how perceived technological impairment (i.e., perceptions of digital technology interfering with daily life) related to psychological distress across five years in adolescence. A latent curve model with structured residuals was applied to disentangle between-from within-person associations, in which it was tested whether (a) adolescents who increased in their perceptions of technological impairment over time also increased in psychological distress (between-person) and (b) if an adolescent who reported greater perceptions of technological impairment relative to their underlying trajectory at one wave consequently reported greater distress at the subsequent wave (within-person). These associations were tested in a sample of 2104 adolescents (Mage = 12.36; 52% girls; 48% Non-White). Perceived technological impairment and psychological distress increased together over time. Girls and older adolescents (13–15 at baseline) reported greater initial levels of perceived impairment. Younger adolescents (9–12 at baseline) increased more steeply in perceived impairment over time. There was no evidence of longitudinal within-person associations. The findings suggest that although there is evidence of between-person associations in which increases in perceived technological impairment coincide with increases in psychological distress, the absence of within-person associations cautions against a cause-and-effect narrative between digital technology use and mental health.

Latent Curve Detrending for Disaggregating Between-Person Effect and Within-Person Effect

Longitudinal data analysis is gaining attention from researchers because it enables us to examine both between- and within-person effects simultaneously. Traditionally, centering has been used with multilevel models to estimate these two effects. However, recent studies found that centering could not disaggregate the between-person and within-person effects when a time-varying predictor shows time trends. This article develops methods for disaggregating the two effects in the presence of time trends using the latent curve model. The proposed methods reveal the link between centering and detrending, which are often seen as different preprocessing for different purposes. Two simulations are conducted to assess and compare the performance of the proposed and existing models. The results show that models with a slope factor behind a predictor can unbiasedly estimate the between- and within-person effects. Also, models with latent between-person predictors can unbiasedly estimate the between-person effect, while those with observed ones suffer from bias.

Expectant parents’ emotions evoked by pregnancy: A longitudinal dyadic analysis of couples in the Swedish Pregnancy Panel

RationaleHolistic antenatal care requires knowledge of individuals’ emotional response to pregnancy. Little is known about how a pregnant woman and her partner influence each other emotionally during a pregnancy. ObjectiveThis study examines six discrete emotions that expectant couples experience during pregnancy, how these emotions change mid-to late-pregnancy, and whether the partners’ emotional responses influence each other. MethodsA longitudinal dyadic study where pregnant women and their partners (1432 couples) rated the extent to which the pregnancy evoked joy, strength, security, worry, shame, and anger at pregnancy week 12–19, 22–24, and 36. Latent curve models with structured residuals identify levels of and change in these emotions over time, while accounting for between- and within-couple variance. ResultsPregnancy evoked mainly joy, strength, security, and worry, and lower levels of anger and shame. Pregnant women and partners felt similar levels of joy, strength, and security, but pregnant women felt more worry, shame, and anger. There was little to no mean-level change in all six measured emotions evoked by pregnancy (between-couple change), and no reciprocal effects between the partners (within-couple change). ConclusionsEmotions in mid-pregnancy were also felt in late pregnancy. Furthermore, the pregnant woman and her partner have individual emotional trajectories. The results can assist healthcare professionals and researchers target interventions to expectant mothers and partners, specifically by understanding emotional response to pregnancy as a stable confound and by not approaching the couple as one emotional unit.

Does income moderate basic relationship processes?

This study explores whether household income moderates the predictive association from adaptive processes (positive and negative interactions and commitment), enduring vulnerabilities (psychological distress), and stressors (financial strain) to future relationship satisfaction? Theory and research have long conceptualized socioeconomic status as a predictor of couple relations, but recent work questions whether socioeconomic status may moderate basic couple relationship processes. This study used data from a U.S. national sample of 927 adults aged 18-34 years in a cohabiting (marital or nonmarital) different-sex partnership (66% female; 22% non-White; 47% earned a high school diploma or GED as their highest education credential) surveyed five times at 4-to 6-month intervals. A series of latent curve models with structured residuals were used to examine between- and within-person associations. Robust between-persons associations emerged consistent with prior literature (e.g., those with more positive and less negative interactions, higher commitment, lower psychological distress, and less financial strain reported higher relationship satisfaction). One robust longitudinal association emerged at the within-person level: higher than typical negative interactions predicted intraindividual decreases in future relationship satisfaction. Within-person associations were more evident in the cross-section: at times when positive interactions and commitment were higher than one's own average and negative interactions and psychological distress were lower than average, relationship satisfaction was also higher than average. Income did not moderate any links with future relationship satisfaction. Results suggest that basic longitudinal processes in relationships operate consistently across income level.

A comparison of models for inferring longitudinal reciprocal relationships between constructs: A case example with internalizing and externalizing problems

AbstractDevelopmental researchers commonly utilize longitudinal data to decompose reciprocal and dynamic associations between repeatedly measured constructs to better understand the temporal precedence between constructs. Although the cross‐lagged panel model (CLPM) is commonly used in developmental research, it has been criticized for its potential to produce biased estimates due to the fact of ignoring the trait‐like, time‐invariant nature of stability in constructs across people and aggregating the between‐ and within‐person effects together as a single estimate. Recently, a growing set of alternatives have emerged, but the estimates across CLPM and alternatives have rarely been compared in developmental research. The primary purpose of this article is to (a) provide a three‐component framework to help developmental researchers to select and specify the most appropriate model, and (b) illustrate how models differ in estimates with an empirical example investigating the reciprocal associations between internalizing and externalizing problems in school‐aged children. Methods: We specified CLPM, Random‐Intercept Cross‐lagged Panel Model (RI‐CLPM), Latent Curve Model with Structured Residuals (LCM‐SR), Latent Change Score (LCS), and a Random‐Mean LCS using four waves of data from ECLS‐K: 2011 (N = 8779). Results: The CLPM provided the most evidence of significant cross‐lagged paths but the poorest fit to the data compared to other models. Alternative models had excellent fit and found either only negative temporal precedence from internalizing to externalizing problems or simply no evidence of prospective within‐person relations between internalizing and externalizing problems.

Women and Men are the Barometers of Relationships: Testing the Predictive Power of Women’s and Men’s Relationship Satisfaction

There is a longstanding belief in relationship science and popular opinion that women are the barometers in mixed-gender relationships such that their perceptions about the partnership carry more weight than men's in predicting future relationship satisfaction, but this idea has yet to be rigorously tested. We analyze data from two studies to test within-person links between men's and women's relationship satisfaction on their own and their partner's next-day and next-year satisfaction. Study 1 combined nine daily diary datasets from Canada and the United States with 901 mixed-gender couples who provided 29,541 daily reports of relationship satisfaction. Study 2 analyzed five annual waves of data from the German Family Panel (pairfam) that surveyed 3,405 mixed-gender couples who provided 21,115 relationship satisfaction reports. Latent curve models with structured residuals (LCM-SR) revealed that in both studies, men's and women's relationship satisfaction significantly predicted their own and their partner's relationship satisfaction, with no gender differences in the magnitude of these effects. Results underscore the interdependence of romantic partners' satisfaction and indicate that both men and women jointly shape romantic relationship satisfaction.