Heining Cham

Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis

Little research has examined factors influencing statistical power to detect the correct number of latent classes using latent profile analysis (LPA). This simulation study examined power related to interclass distance between latent classes given true number of classes, sample size, and number of indicators. Seven model selection methods were evaluated. None had adequate power to select the correct number of classes with a small (Cohens d = .2) or medium (d = .5) degree of separation. With a very large degree of separation (d = 1.5), the Lo–Mendell–Rubin test (LMR), adjusted LMR, bootstrap likelihood ratio test, Bayesian Information Criterion (BIC), and sample-size-adjusted BIC were good at selecting the correct number of classes. However, with a large degree of separation (d = .8), power depended on number of indicators and sample size. Akaikes Information Criterion and entropy poorly selected the correct number of classes, regardless of degree of separation, number of indicators, or sample size.

Advanced nonlinear latent variable modeling: Distribution analytic LMS and QML estimators of interaction and quadratic effects

Interaction and quadratic effects in latent variable models have to date only rarely been tested in practice. Traditional product indicator approaches need to create product indicators (e.g., x 1 2, x 1 x 4) to serve as indicators of each nonlinear latent construct. These approaches require the use of complex nonlinear constraints and additional model specifications and do not directly address the nonnormal distribution of the product terms. In contrast, recently developed, easy-to-use distribution analytic approaches do not use product indicators, but rather directly model the nonlinear multivariate distribution of the measured indicators. This article outlines the theoretical properties of the distribution analytic Latent Moderated Structural Equations (LMS; Klein & Moosbrugger, 2000) and Quasi-Maximum Likelihood (QML; Klein & Muthén, 2007) estimators. It compares the properties of LMS and QML to those of the product indicator approaches. A small simulation study compares the two approaches and illustrates the advantages of the distribution analytic approaches as multicollinearity increases, particularly in complex models with multiple nonlinear terms. An empirical example from the field of work stress applies LMS and QML to a model with an interaction and 2 quadratic effects. Example syntax for the analyses with both approaches is provided.

Multilevel Modeling: Current and Future Applications in Personality Research

Traditional statistical analyses can be compromised when data are collected from groups or multiple observations are collected from individuals. We present an introduction to multilevel models designed to address dependency in data. We review current use of multilevel modeling in 3 personality journals showing use concentrated in the 2 areas of experience sampling and longitudinal growth. Using an empirical example, we illustrate specification and interpretation of the results of series of models as predictor variables are introduced at Levels 1 and 2. Attention is given to possible trends and cycles in longitudinal data and to different forms of centering. We consider issues that may arise in estimation, model comparison, model evaluation, and data evaluation (outliers), highlighting similarities to and differences from standard regression approaches. Finally, we consider newer developments, including 3-level models, cross-classified models, nonstandard (limited) dependent variables, multilevel structural equation modeling, and nonlinear growth. Multilevel approaches both address traditional problems of dependency in data and provide personality researchers with the opportunity to ask new questions of their data.

Propensity scores as a basis for equating groups: basic principles and application in clinical treatment outcome research.

A propensity score is the probability that a participant is assigned to the treatment group based on a set of baseline covariates. Propensity scores provide an excellent basis for equating treatment groups on a large set of covariates when randomization is not possible. This article provides a nontechnical introduction to propensity scores for clinical researchers. If all important covariates are measured, then methods that equate on propensity scores can achieve balance on a large set of covariates that mimics that achieved by a randomized experiment. We present an illustration of the steps in the construction and checking of propensity scores in a study of the effectiveness of a health coach versus treatment as usual on the well-being of seriously ill individuals. We then consider alternative methods of equating groups on propensity scores and estimating treatment effects including matching, stratification, weighting, and analysis of covariance. We illustrate a sensitivity analysis that can probe for the potential effects of omitted covariates on the estimate of the causal effect. Finally, we briefly consider several practical and theoretical issues in the use of propensity scores in applied settings. Propensity score methods have advantages over alternative approaches to equating groups particularly when the treatment and control groups do not fully overlap, and there are nonlinear relationships between covariates and the outcome.

Estimating Latent Variable Interactions With Nonnormal Observed Data: A Comparison of Four Approaches

A Monte Carlo simulation was conducted to investigate the robustness of 4 latent variable interaction modeling approaches (Constrained Product Indicator [CPI], Generalized Appended Product Indicator [GAPI], Unconstrained Product Indicator [UPI], and Latent Moderated Structural Equations [LMS]) under high degrees of nonnormality of the observed exogenous variables. Results showed that the CPI and LMS approaches yielded biased estimates of the interaction effect when the exogenous variables were highly nonnormal. When the violation of nonnormality was not severe (normal; symmetric with excess kurtosis < 1), the LMS approach yielded the most efficient estimates of the latent interaction effect with the highest statistical power. In highly nonnormal conditions, the GAPI and UPI approaches with maximum likelihood (ML) estimation yielded unbiased latent interaction effect estimates, with acceptable actual Type I error rates for both the Wald and likelihood ratio tests of interaction effect at N ≥ 500. An empirical example illustrated the use of the 4 approaches in testing a latent variable interaction between academic self-efficacy and positive family role models in the prediction of academic performance.

Estimating Interaction Effects With Incomplete Predictor Variables

The existing missing data literature does not provide a clear prescription for estimating interaction effects with missing data, particularly when the interaction involves a pair of continuous variables. In this article, we describe maximum likelihood and multiple imputation procedures for this common analysis problem. We outline 3 latent variable model specifications for interaction analyses with missing data. These models apply procedures from the latent variable interaction literature to analyses with a single indicator per construct (e.g., a regression analysis with scale scores). We also discuss multiple imputation for interaction effects, emphasizing an approach that applies standard imputation procedures to the product of 2 raw score predictors. We thoroughly describe the process of probing interaction effects with maximum likelihood and multiple imputation. For both missing data handling techniques, we outline centering and transformation strategies that researchers can implement in popular software packages, and we use a series of real data analyses to illustrate these methods. Finally, we use computer simulations to evaluate the performance of the proposed techniques.

Impacts of the CARE for Teachers program on teachers’ social and emotional competence and classroom interactions

Understanding teachers’ stress is of critical importance to address the challenges in today’s educational climate. Growing numbers of teachers are reporting high levels of occupational stress, and high levels of teacher turnover are having a negative impact on education quality. Cultivating Awareness and Resilience in Education (CARE for Teachers) is a mindfulness-based professional development program designed to promote teachers’ social and emotional competence and improve the quality of classroom interactions. The efficacy of the program was assessed using a cluster randomized trial design involving 36 urban elementary schools and 224 teachers. The CARE for Teachers program involved 30 hr of in-person training in addition to intersession phone coaching. At both pre- and postintervention, teachers completed self-report measures and assessments of their participating students. Teachers’ classrooms were observed and coded using the Classroom Assessment Scoring System (CLASS). Analyses showed that CARE for Teachers had statistically significant direct positive effects on adaptive emotion regulation, mindfulness, psychological distress, and time urgency. CARE for Teachers also had a statistically significant positive effect on the emotional support domain of the CLASS. The present findings indicate that CARE for Teachers is an effective professional development both for promoting teachers’ social and emotional competence and increasing the quality of their classroom interactions.

Propensity Score Analysis With Missing Data

Propensity score analysis is a method that equates treatment and control groups on a comprehensive set of measured confounders in observational (nonrandomized) studies. A successful propensity score analysis reduces bias in the estimate of the average treatment effect in a nonrandomized study, making the estimate more comparable with that obtained from a randomized experiment. This article reviews and discusses an important practical issue in propensity analysis, in which the baseline covariates (potential confounders) and the outcome have missing values (incompletely observed). We review the statistical theory of propensity score analysis and estimation methods for propensity scores with incompletely observed covariates. Traditional logistic regression and modern machine learning methods (e.g., random forests, generalized boosted modeling) as estimation methods for incompletely observed covariates are reviewed. Balance diagnostics and equating methods for incompletely observed covariates are briefly described. Using an empirical example, the propensity score estimation methods for incompletely observed covariates are illustrated and compared. (PsycINFO Database Record

Balancing scientific accuracy and participant burden: testing the impact of sampling intensity on diurnal cortisol indices

Abstract Despite the increasing popularity of incorporating salivary cortisol measurement into health and social science research, relatively little empirical work has been conducted on the number of saliva samples across the day required to capture key features of the diurnal cortisol rhythm, such as the diurnal cortisol slope, the area under the curve (AUC), and the cortisol awakening response (CAR). The primary purpose of this study is to compare slope, AUC, and CAR measures obtained from an intensive sampling protocol with estimates from less intensive protocols, to identify sampling protocols with minimal participant burden that still provide reasonably accurate assessment of each of these measures. Twenty-four healthy adults provided samples four times in the first hour awake, and then every hour throughout the rest of the day until bedtime (M = 17.8 samples/day; SD = 2.0), over two consecutive days (N = 862 total samples). We compared measures calculated from this maximum intensity protocol to measures calculated from two to six sampling points per day. Overall, results show that salivary cortisol protocols with two fixed samples (waking and bedtime) and three additional daily samples, closely approximates the full cortisol decline (slope). Abbreviated sampling protocols of total cortisol exposure across the day (AUC), however, were not well approximated by reduced sampling protocols. CAR measures based on only two samples, including waking cortisol and a second sample measured at a fixed time point between 30 and 60 min after waking, provided a measure of the CAR that closely approximated CAR measures obtained from 3 or 4 sampling points.

Long-term effects of the Family Bereavement Program on spousally bereaved parents: Grief, mental health problems, alcohol problems, and coping efficacy.

This study reports on the findings from a 6-year follow-up of a randomized trial of the Family Bereavement Program (FBP) on the outcomes for spousally bereaved parents. Spousally bereaved parents (N = 131) participated in the trial in which they were randomly assigned to receive the FBP (N = 72) or literature control (N = 59). Parents were assessed at four time points: pretest, posttest, and 11-month and 6-year follow-up. They reported on mental health problems, grief, and parenting at all four time periods. At the 6-year follow-up, parents reported on additional measures of persistent complex bereavement disorder, alcohol abuse problems, and coping efficacy. Bereaved parents in the FBP as compared to those in the literature control had lower levels of symptoms of depression, general psychiatric distress, prolonged grief, and alcohol problems, and higher coping efficacy (for mothers) at the 6-year follow-up. Multiple characteristics of the parent (e.g., gender, age, and baseline mental health problems) and of the spousal death (e.g., cause of death) were tested as moderators of program effects on each outcome, but only 3 of 45 tests of moderation were significant. Latent growth modeling found that the effects of the FBP on depression, psychiatric distress, and grief occurred immediately following program participation and were maintained over 6 years. Mediation analysis found that improvement in positive parenting partially mediated program effects to reduce depression and psychiatric distress, but had an indirect effect to higher levels of grief at the 6-year follow-up. Mediation analysis also found that improved parenting at the 6-year follow-up was partially mediated by program effects to reduce depression and that program effects to increase coping efficacy at the 6-year follow-up was partially mediated through reduced depression and grief and improved parenting. FBP reduced mental health problems, prolonged grief, and alcohol abuse, and increased coping efficacy of spousally bereaved parents 6 years later. Mediation pathways for program effects differed across outcomes at the 6-year follow-up.