The Optimal Starting Model to Search for the Accurate Growth Trajectory in Latent Growth Models

Kim, Minjung; Hsu, Hsien-Yuan; Kwok, Oi‐Man; Seo, Sunmi

doi:10.3389/fpsyg.2018.00349

Cited by 6 publications

(4 citation statements)

References 33 publications

(56 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These models typically have two latent variables, representing individual intercepts and slopes, thereby allowing each unit to have its own trajectory. Often, linear trajectories are assumed, but higher order polynomials can also be modeled (Kim et al, 2018). Fixed effects variants of these models that control for time-invariant unobserved heterogeneity can be specified by allowing the time-varying covariates to correlate with the latent trajectory components.…”

Section: Introductionmentioning

confidence: 99%

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.

Andersen¹

2022

Psychological Methods

View full text Add to dashboard Cite

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 residuallevel models place potentially strong assumptions on the initial conditions, i.e., 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 amongst 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 topics, essentially whenever the process under investigation is changing over the course of the observation period. This paper 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.

show abstract

Section: Introductionmentioning

confidence: 99%

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.

Andersen¹

2022

Psychological Methods

View full text Add to dashboard Cite

show abstract

“…Estimation procedures for the interaction models are prone to convergence problems [ 62 ]. To minimize convergence problems, we estimated sets of starting values for residual variances and other terms from a collection of sub-models that were together approaching a saturated model [ 62 ]. In addition, convergence problems increased with an increasing number of interaction terms.…”

Section: Methodsmentioning

confidence: 99%

The role of frailty in the relationships between social relationships and health outcomes: a longitudinal study

Mehrabi,

Béland

2024

BMC Public Health

View full text Add to dashboard Cite

Background Socially isolated older adults incur increased risks of adverse health outcomes, though the strength of this association is unclear. We examined whether changes in physical frailty moderated the associations between changes in social relationships and changes in health outcomes among older adults. Methods This longitudinal study is based on three waves of the FRéLE study among 1643 Canadian community-dwelling older adults aged 65 years and older over 2 years. We performed latent growth curve modelling (LGMs) to assess changes with the assumption of missing not at random, adjusting for time-invariant covariates. We used the latent moderated structural equations (LMS) to test the interactions in LGMs. Social relationships were measured by social participation, social networks, and social support from different social ties. Frailty was assessed using the five components of the phenotype of frailty. Results The results revealed that changes in frailty moderated changes in social participation (β = 3.229, 95% CI: 2.212, 4.245), social contact with friends (β = 4.980, 95% CI: 3.285, 6.675), and social support from friends (β = 2.406, 95% CI: 1.894, 2.917), children (β = 2.957, 95% CI: 1.932, 3.982), partner (β = 4.170, 95% CI: 3.036, 5.305) and extended family (β = 6.619, 95% CI: 2.309, 10.923) with changes in cognitive function and depressive symptoms, but not with chronic diseases. These results highlight the beneficial role of social relationships in declining depressive symptoms and improving cognitive health among older adults experiencing increases in frailty. Conclusions The findings suggest that changes in social support have a protective and compensatory role in decreasing depressive symptoms and enhancing cognitive health among older adults with increasing frailty. Public health policy and strategies should consider the impact of social support on multiple health outcomes among older adults with increasing frailty. Further experimental studies and interventions are warranted to extend findings on the relationships between social relationships and health outcomes, targeting frail older adults. Future studies may also consider other health-related risk factors that may impact the associations between social relationships and health outcomes among older adults.

show abstract

“…For both growth models (total analysis sample; subgroup of first-time mothers) the best fitting unconditional model (without predictors) was built using stepwise backward model selection ( Kim et al, 2018 ) relying on the following criteria for model evaluation: ΔAIC (Akaike information criterion), ΔBIC (Bayesian information criterion), ΔCFI (comparative fit index), ΔRMSEA (root mean squared error of approximation), ΔSRMR (standardized root mean square residual). Afterwards, this best fitting growth trajectory was extended by including all a priori defined predictors as time-invariant covariates, resulting in the conditional growth model ( Figure 1 ).…”

Section: Methodsmentioning

confidence: 99%

Development of Mothers' Health Literacy: Findings From the KUNO-Kids Study

Pawellek¹,

Köninger²,

Melter³

et al. 2023

HLRP: Health Literacy Research and Practice

View full text Add to dashboard Cite

Background and Objective: The aim of this study was to analyze the longitudinal development of health literacy (HL) in a large cohort of new mothers in Germany and to investigate which determinants are associated with the initial HL level and with change over time. Methods: Longitudinal data from 1,363 mothers participating with their child in the KUNO-Kids Health Study was used; data were collected at birth of the child (baseline), after 6 and 12 months, using interviews and self-report questionnaires. The HL of mothers was assessed with the health care scale of the European Health Literacy Survey Questionnaire, which has 16 items on accessing, understanding, appraising, and applying health information in the health care setting. Latent growth curve models were used to analyze average trajectories and predictors of HL in the total sample and in the subgroup of first-time mothers. Key Results: HL values increased from baseline ( M = 35.46, standard deviation [ SD ] = 7.34) over 6 months ( M = 37.31, SD = 7.31) to 12 months ( M = 38.01, SD = 7.41). The increase was statistically significant in the total sample (1.188, standard error [SE] = 0.087, p < .001) and in the subgroup of first-time mothers (1.357, SE = 0.113, p <. 001), with a steeper trajectory for mothers with lower HL at baseline. Several personal and situational variables were associated with HL at baseline (e.g., education, child health) and with its development (e.g., number of children). Conclusion: Overall, new mothers became slightly more health literate during their child's first year of life. However, some groups of mothers could benefit from support in developing HL skills even before childbirth. [ HLRP: Health Literacy Research and Practice . 2023;7(1):e39–e51. ] Plain Language Summary: We investigated how health literacy (that is the ability to find, understand, and apply health information) develops in new mothers in Germany. Mothers told us that this ability slightly improved during the child's first year of life. Some mothers still have difficulties in dealing with health information; these mothers should be supported even before childbirth.

show abstract

The Optimal Starting Model to Search for the Accurate Growth Trajectory in Latent Growth Models

Cited by 6 publications

References 33 publications

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.

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.

The role of frailty in the relationships between social relationships and health outcomes: a longitudinal study

Development of Mothers' Health Literacy: Findings From the KUNO-Kids Study

Contact Info

Product

Resources

About