Modelling nonlinear moderation effects with local structural equation modelling (LSEM): A non-technical introduction

Understanding the differential strength of effects in the presence of a third variable, known as a moderation effect, is a common research goal in many psychological and behavioural science fields. If structural equation modelling is applied to test effects of interest, the investigation of differential strength of effects will typically ask how parameters of a latent variable model are influenced by categorical or continuous moderators, such as age, socio-economic status, personality traits, etc. Traditional approaches to continuous moderators in SEMs predominantly address linear moderation effects, risking the oversight of nonlinear effects. Moreover, some approaches have methodological limitations, for example, the need to categorise moderators or to pre-specify parametric forms of moderation. This tutorial introduces local structural equation modelling (LSEM) in a non-technical way. LSEM is a nonparametric approach that allows the analysis of nonlinear moderation effects without the above-mentioned limitations. Using an empirical dataset, we demonstrate the implementation of LSEM through the R-sirt package, emphasising its versatility in both exploratory analysis of nonlinear moderation without prior knowledge and confirmatory testing of hypothesised moderation functions. The tutorial also addresses common modelling issues and extends the discussion to different application scenarios, demonstrating its flexibility.

Job characteristics and personality change in young adulthood: A 12-year longitudinal study and replication

OBJECTIVE

Personality changes are related to successfully performing adult occupational roles which require teamwork, duty, and managing stress. However, it is unclear how personality development relates to specific job characteristics that vary across occupations.

METHOD

We investigated whether 151 objective job characteristics, derived from the Occupational Information Network (O*NET), were associated with personality levels and changes in a 12-year longitudinal sample followed over the school to work transition. Using cross-validated regularized modeling, we combined two Icelandic longitudinal datasets (total N = 1054) and constructed an individual-level, aggregated job characteristics score that maximized prediction of personality levels at baseline and change over time.

RESULTS

The strongest association was found for level of openness (0.25), followed by conscientiousness (0.16) and extraversion (0.14). Overall, aggregated job characteristics had a stronger prediction for personality intercepts (0.14) than slopes (0.10). These results were subsequently replicated in a U.S. sample using levels of the Big Five as the dependent variable. This indicates that associations between job characteristics and personality are generalizable across life stages and nations.

CONCLUSIONS

Our findings suggest that job titles are a valuable resource that can be linked to personality to better understand factors that influence psychological development. Further work is needed to document the prospective validity of job characteristics across a wider range of occupations and age.

Basic or Adaptation: The Assessment and Heritability of a Brief Measure of Agency

The interpersonal circumplex describes two major axes of personality that guide much of social behavior. Agency, one half of the interpersonal circumplex, refers to relatively stable behavioral patterns that center on self-focused dominance and assertiveness. Past empirical work on agency tends to treat the dimension as a characteristic adaptation, rather than a basic component of personality, in part due to the relatively large gender difference in agency with masculine individuals tending to behave more agentic. However, the psychometric overlap between agency and the most closely linked big five dimension, extraversion, is not well-established, and no behavior genetic work has documented evidence concerning the role of genetic and environmental influences. It is unclear whether agency is more similar to a personality trait, with no evidence of shared environmental influence and moderate heritability, or a characteristic adaptation, with some evidence for shared environmental influence and possibly lower heritability. We used the Midlife Development in the United States study to examine agency, big five, and generativity with replication and robustness check (Nnon-twins = 5,194; Ntwins = 1,914; NMilwaukee = 592). Results indicated that agency was higher in men (d = -.24), moderately heritable (44.4%), strongly correlated with extraversion (r = .51), moderately correlated with generativity (r = .36), and that approximately 40% of the variance in agency was shared with the big five. Agency also changed strongly with extraversion and openness, but less so generativity. Altogether, these results indicate that agency functions similar to other basic personality dimensions but is not clearly a dispositional trait.

Estimating Local Structural Equation Models

Local structural equation models (LSEM) are structural equation models that study model parameters as a function of a moderator. This article reviews and extends LSEM estimation methods and discusses the implementation in the R package sirt. In previous studies, LSEM was fitted as a sequence of models separately evaluated as each value of the moderator variables. In this article, a joint estimation approach is proposed that is a simultaneous estimation method across all moderator values and also allows some model parameters to be invariant with respect to the moderator. Moreover, sufficient details on the main estimation functions in the R package sirt are provided. The practical implementation of LSEM is demonstrated using illustrative datasets and an empirical example. Moreover, two simulation studies investigate the statistical properties of parameter estimation and significance testing in LSEM.

Personality and limitations in instrumental activities of daily living in old age: Reciprocal associations across 12 years

Personality traits have been reported to predict difficulties in performing instrumental activities of daily living (IADL) in old age, such as preparing meals or shopping. However, little is known about the reciprocal effects on personality. In this study, we examined bidirectional relationships between personality traits and the capacity to perform IADL using four waves of longitudinal data from 3540 older adults (aged 65 years and older) from the Health and Retirement Study. We applied a random-intercept cross-lagged panel model to separate between- and within-person effects across time and compared it to a traditional cross-lagged panel model. At the between-person level, higher neuroticism and lower conscientiousness were associated with more IADL limitations. Within individuals across time, increases in neuroticism and decreases in conscientiousness and extraversion were associated with increases in IADL limitations 4 years later. In contrast, increases in IADL limitations only predicted increases in neuroticism and decreases in extraversion. These results indicate that some personality traits affect and are affected by limitations in functional capacities in old age. Results of the within-person model build a strong foundation for future personality interventions as a pathway to maintain high functioning in old age.

Detecting Idiographic Personality Change

Personality changes across the lifespan, but strong evidence regarding the mechanisms responsible for personality change remains elusive. Studies of personality change and life events, for example, suggest that personality is difficult to change. But there are two key issues with assessing personality change. First, most change models optimize population-level, not individual-level, effects, which ignores heterogeneity in patterns of change. Second, optimizing change as mean-levels of self-reports fails to incorporate methods for assessing personality dynamics, such as using changes in variances of and correlations in multivariate time series data that often proceed changes in mean-levels, making variance change detection a promising technique for the study of change. Using a sample of N = 388 participants (total N = 21,790) assessed weekly over 60 weeks, we test a permutation-based approach for detecting individual-level personality changes in multivariate time series and compare the results to event-based methods for assessing change. We find that a non-trivial number of participants show change over the course of the year but that there was little association between these change points and life events they experienced. We conclude by highlighting the importance in idiographic and dynamic investigations of change.