Rens van de Schoot

A checklist for testing measurement invariance

The analysis of measurement invariance of latent constructs is important in research across groups, or across time. By establishing whether factor loadings, intercepts and residual variances are equivalent in a factor model that measures a latent concept, we can assure that comparisons that are made on the latent variable are valid across groups or time. Establishing measurement invariance involves running a set of increasingly constrained structural equation models, and testing whether differences between these models are significant. This paper provides a step-by-step guide to analysing measurement invariance.

Facing Off with Scylla and Charybdis: A Comparison of Scalar, Partial, and the Novel Possibility of Approximate Measurement Invariance

Measurement invariance (MI) is a pre-requisite for comparing latent variable scores across groups. The current paper introduces the concept of approximate MI building on the work of Muthén and Asparouhov and their application of Bayesian Structural Equation Modeling (BSEM) in the software Mplus. They showed that with BSEM exact zeros constraints can be replaced with approximate zeros to allow for minimal steps away from strict MI, still yielding a well-fitting model. This new opportunity enables researchers to make explicit trade-offs between the degree of MI on the one hand, and the degree of model fit on the other. Throughout the paper we discuss the topic of approximate MI, followed by an empirical illustration where the test for MI fails, but where allowing for approximate MI results in a well-fitting model. Using simulated data, we investigate in which situations approximate MI can be applied and when it leads to unbiased results. Both our empirical illustration and the simulation study show approximate MI outperforms full or partial MI In detecting/recovering the true latent mean difference when there are (many) small differences in the intercepts and factor loadings across groups. In the discussion we provide a step-by-step guide in which situation what type of MI is preferred. Our paper provides a first step in the new research area of (partial) approximate MI and shows that it can be a good alternative when strict MI leads to a badly fitting model and when partial MI cannot be applied.

A gentle introduction to bayesian analysis: applications to developmental research.

Bayesian statistical methods are becoming ever more popular in applied and fundamental research. In this study a gentle introduction to Bayesian analysis is provided. It is shown under what circumstances it is attractive to use Bayesian estimation, and how to interpret properly the results. First, the ingredients underlying Bayesian methods are introduced using a simplified example. Thereafter, the advantages and pitfalls of the specification of prior knowledge are discussed. To illustrate Bayesian methods explained in this study, in a second example a series of studies that examine the theoretical framework of dynamic interactionism are considered. In the Discussion the advantages and disadvantages of using Bayesian statistics are reviewed, and guidelines on how to report on Bayesian statistics are provided.

Personality Types in Adolescence: Change and stability and Links With Adjustment and Relationships : A Five-Wave Longitudinal Study.

We examined change and stability of the 3 personality types identified by Block and Block (1980) and studied their links with adjustment and relationships. We used data from a 5-wave study of 923 early-to-middle and 390 middle-to-late adolescents, thereby covering the ages of 12-20 years. In Study 1, systematic evidence for personality change was found, in that the number of overcontrollers and undercontrollers decreased, whereas the number of resilients increased. Undercontrol, in particular, was found to peak in early-to-middle adolescence. We also found substantial stability of personality types, because 73.5% of the adolescents had the same personality type across the 5 waves. Personality change was mainly characterized by 2 transitions: overcontrol → resiliency and undercontrol → resiliency. The transitional analyses implied that the resilient type serves more often as the end point of personality development in adolescence than do overcontrol and undercontrol. Analyses of the personality type trajectories also revealed that the majority of adolescents who change personality type across 5 years made only 1 transition. Study 2 revealed systematic differences between resilients and overcontrollers in anxiety. Stable resilients were less anxious over time than were stable overcontrollers. Further, change from overcontrol to the resilient type was accompanied by decreases in anxiety, whereas change from the resilient type to overcontrol was accompanied by an increase in anxiety. Similarly, systematic differences between personality types were found in the formation of intimate relationships.

Automatic processes in at-risk adolescents: the role of alcohol-approach tendencies and response inhibition in drinking behavior

AIMS This study examined the association between automatic processes and drinking behavior in relation to individual differences in response inhibition in young adolescents who had just started drinking. It was hypothesized that strong automatic behavioral tendencies toward alcohol-related stimuli (alcohol-approach bias) were associated with higher levels of alcohol use, especially amongst adolescents with relatively weak inhibition skills. DESIGN To test this hypothesis structural equation analyses (standard error of mean) were performed using a zero inflated Poisson (ZIP) model. A well-known problem in studying risk behavior is the low incidence rate resulting in a zero dominated distribution. A ZIP-model accounts for non-normality of the data. SETTING Adolescents were selected from secondary Special Education schools (a risk group for the development of substance use problems). PARTICIPANTS Participants were 374 adolescents (mean age of M = 13.6 years). MEASUREMENTS Adolescents completed the alcohol approach avoidance task (a-AAT), the Stroop colour naming task (Stroop) and a questionnaire that assessed alcohol use. FINDINGS The ZIP-model established stronger alcohol-approach tendencies for adolescent drinkers (P < 0.01) and the interaction revealed a stronger effect of alcohol-approach tendencies on alcohol use in the absence of good inhibition skills (P < 0.05). CONCLUSION Automatically-activated cognitive processes are associated with the drinking behavior of young, at-risk adolescents. It appears that alcohol-approach tendencies are formed shortly after the initiation of drinking and particularly affect the drinking behavior of adolescents with relatively weak inhibition skills. Implications for the prevention of problem drinking in adolescents are discussed.

Prolonged Grief Disorder, depression, and posttraumatic stress disorder are distinguishable syndromes.

BACKGROUND This study examined the distinctiveness of symptoms of Prolonged Grief Disorder (PGD), depression, and posttraumatic stress disorder (PTSD). We compared the fit of a one-factor model with the fit of four hierarchical models in which symptoms formed three distinct correlated higher-order dimensions, and PTSD-items were modeled in different ways. METHODS Self-reported data were available from two samples; 572 mourners recruited via the internet and 408 mourners recruited via healthcare workers. RESULTS In Sample 1, the unitary model did not fit the data. The four hierarchical models all fit better. The model in which PTSD-items constituted four lower-order factors of reexperiencing, avoidance, dysphoria, and hyperarousal fit the data best. The fit was further improved, when one weak PGD-item and one weak PTSD-item were removed, and error-terms of similar items were allowed to correlate. Findings from Sample 1 were replicated in Sample 2. LIMITATIONS This study relied on self-reported data. Not all PGD-criteria and depression-criteria were assessed. CONCLUSIONS This is the first confirmatory factor analysis study showing that symptoms of PGD, depression, and PTSD represent distinguishable syndromes. PGD-symptoms should be addressed in the assessment and treatment of bereaved people seeking treatment.

Health-related quality of life after burns : a prospective multicentre cohort study with 18 months follow-up.

Background: Health-related quality of life (HRQOL) is an important parameter after medical treatments. Knowledge of (predictors of) diminished quality of life can help improve medical outcome. The aim of this study was to quantify health loss in patients with burns and to assess the contribution of injury extent, age, gender, and psychologic factors to HRQOL and speed of recovery. A multicenter prospective cohort design was used to address these aims. Methods: Data were obtained from 260 adults with burns. Patients completed the EQ-5D at 3 weeks, 3, 6, 9, and 18 months after burn and psychologic questionnaires during hospitalization. Patients scores were compared with an age- and gender-weighted norm population. Results: Patients suffered from substantial health losses at short term, but after 18 months the majority reached a HRQOL comparable with the norm population with the exception of patients requiring two or more surgeries. The best predictor of long-term HRQOL and the speed of recovery was the number of surgeries, followed by psychologic problems. Both predicted baseline and trajectories of improvement. Symptoms of traumatic stress were most debilitating over time. Conclusions: Both injury severity and psychologic problems play a pivotal role in reduced HRQOL and the speed of recovery. The number of surgeries seems to give a practically useful indication of the expected recovery speed that could aid in decision making and provides adequate information for patients in the aftermath of their initial surgical treatment. Screening for traumatic stress is recommended.

Reducing bias due to systematic attrition in longitudinal studies : The benefits of multiple imputation

Most longitudinal studies are plagued by drop-out related to variables at earlier assessments (systematic attrition). Although systematic attrition is often analysed in longitudinal studies, surprisingly few researchers attempt to reduce biases due to systematic attrition, even though this is possible and nowadays technically easy. This is particularly true for studies of stability and the long-term prediction of developmental outcomes. We provide guidelines how to reduce biases in such cases particularly with multiple imputation. Following these guidelines does not require advanced statistical knowledge or special software. We illustrate these guidelines and the importance of reducing biases due to selective attrition with a 25-year longitudinal study on the long-term prediction of aggressiveness and delinquency.

Analyzing small data sets using Bayesian estimation: the case of posttraumatic stress symptoms following mechanical ventilation in burn survivors

Background The analysis of small data sets in longitudinal studies can lead to power issues and often suffers from biased parameter values. These issues can be solved by using Bayesian estimation in conjunction with informative prior distributions. By means of a simulation study and an empirical example concerning posttraumatic stress symptoms (PTSS) following mechanical ventilation in burn survivors, we demonstrate the advantages and potential pitfalls of using Bayesian estimation. Methods First, we show how to specify prior distributions and by means of a sensitivity analysis we demonstrate how to check the exact influence of the prior (mis-) specification. Thereafter, we show by means of a simulation the situations in which the Bayesian approach outperforms the default, maximum likelihood and approach. Finally, we re-analyze empirical data on burn survivors which provided preliminary evidence of an aversive influence of a period of mechanical ventilation on the course of PTSS following burns. Results Not suprisingly, maximum likelihood estimation showed insufficient coverage as well as power with very small samples. Only when Bayesian analysis, in conjunction with informative priors, was used power increased to acceptable levels. As expected, we showed that the smaller the sample size the more the results rely on the prior specification. Conclusion We show that two issues often encountered during analysis of small samples, power and biased parameters, can be solved by including prior information into Bayesian analysis. We argue that the use of informative priors should always be reported together with a sensitivity analysis.

Automatic processes and the drinking behavior in early adolescence: a prospective study

BACKGROUND This study examined the bi-directional prospective link between automatic alcohol-approach tendencies and alcohol use in a group of young adolescents (mean age = 13.6 years). The adolescents in the present study were assumed to be at-risk of early alcohol use and later problem drinking. It was hypothesized that alcohol use and automatic approach tendencies would reinforce one another particularly in the absence of well-developed inhibition skills. METHODS A total of 347 adolescents (N = 279 at follow-up) from special secondary education, a risk group for the development of substance use problems, participated in the study. Automatic approach tendencies were assessed with the alcohol-approach avoidance task, inhibition skills were assessed with the Stroop task, and alcohol used was measured using a self-report measure. RESULTS Zero-inflated Poisson analysis revealed a significant effect of automatic approach tendencies predicting alcohol use 6 months later, although only for adolescents with weaker inhibition skills. CONCLUSIONS Automatic approach tendencies predict future drinking behavior of young adolescents with relatively weak inhibition skills. The findings of the present study have important implications for alcohol interventions for adolescents. Results are discussed in terms of risk factors for the development of problematic alcohol use in young adolescents.