William M. van der Veld

Testing Structural Equation Models or Detection of Misspecifications

Assessing the correctness of a structural equation model is essential to avoid drawing incorrect conclusions from empirical research. In the past, the chi-square test was recommended for assessing the correctness of the model but this test has been criticized because of its sensitivity to sample size. As a reaction, an abundance of fit indexes have been developed. The result of these developments is that structural equation modeling packages are now producing a large list of fit measures. One would think that this progression has led to a clear understanding of evaluating models with respect to model misspecifications. In this article we question the validity of approaches for model evaluation based on overall goodness-of-fit indexes. The argument against such usage is that they do not provide an adequate indication of the “size” of the models misspecification. That is, they vary dramatically with the values of incidental parameters that are unrelated with the misspecification in the model. This is illustrated using simple but fundamental models. As an alternative method of model evaluation, we suggest using the expected parameter change in combination with the modification index (MI) and the power of the MI test.

Fit of Different Models for Multitrait-Multimethod Experiments

In the past, several models have been developed for the estimation of the reliability and validity of measurement instruments from multitrait-multimethod (MTMM) experiments. Suggestions have been made for additive, multiplicative and correlated uniqueness models, whereas recently Coenders and Saris (2000) suggested a procedure to test these models against one another. In this article, the different models suggested for the analysis of MTMM matrixes have been compared for their fit to 87 data sets collected in the United States (Andrews, 1984; Rodgers, Andrews, & Herzog, 1992), Austria (Koltringer, 1995), and the Netherlands (Scherpenzeel & Saris, 1997). As most variables are categorical, the analysis has been carried out on the basis of polychoric-polyserial correlation coefficients and of Pearson correlations. The fit of the models based on polychoric correlations is much worse than the fit of models based on product moment correlations, but in both cases a model that assumes additive method effects fits most data sets better than the other models, including the so-called multiplicative models.

Psychometric properties of the Positive Mental Health Scale (PMH-scale)

BackgroundIn recent years, it has been increasingly recognized that the absence of mental disorder is not the same as the presence of positive mental health (PMH). With the PMH-scale we propose a short, unidimensional scale for the assessment of positive mental health. The scale consists of 9 Likert-type items.MethodsThe psychometric properties of the PMH-scale were tested in a series of six studies using samples from student (n = 5406), patient (n = 1547) and general (n = 3204) populations. Factorial structure and measurement equivalence were tested with the measurement invariance testing. The factor models were analysed with the maximum likelihood procedure. Internal consistency was examined using Cronbach’s alpha, test-retest reliability, convergent and divergent validity was examined by Pearson correlation. Sensitivity to (therapeutic) change was examined with the t-test.ResultsResults confirmed unidimensionality, scalar invariance across samples and over time, high internal consistency, good retest-reliability, good convergent and discriminant validity as well as sensitivity to therapeutic change.ConclusionsThese findings suggest that the PMH-Scale indeed measures a single concept and allows us to compare scores over groups and over time. The PMH-scale thus is a brief and easy to interpret instrument for measuring PMH across a large variety of relevant groups.

Validating the Beck Depression Inventory-II in Indonesia's general population and coronary heart disease patients

This study assesses the validity and determines the cut-off point for the Beck Depression Inventory-II (the BDI-II) among Indonesians. The Indonesian version of the BDI-II (the Indo BDI-II) was administered to 720 healthy individuals from the general population, 215 Coronary Heart Disease (CHD) patients, and 102 depressed patients. Confirmatory factor analysis indicated factorial similarity across the three samples. Significant correlations between the Indo BDI-II and other self-report measures related to depression demonstrated construct validity of the Indo BDI-II. Furthermore, there was a highly significant difference in the Indo BDI-II scores between depressed patients and non-depressed participants. Internal consistency and re-test reliability of the Indo BDI-II were acceptable. The receiver operating characteristic (ROC) curve indicated that the cut-off point of the Indo BDI-II for a mild severity of depression in Indonesian population should be 17. We conclude that the Indo BDI-II is a valid measure of depression, both in the Indonesian general population and in CHD patients.

A Cluster Randomized Controlled Trial Testing the Effectiveness of Houvast: A Strengths-Based Intervention for Homeless Young Adults

Objective: To test the effectiveness of Houvast: a strengths-based intervention for homeless young adults. Method: A cluster randomized controlled trial was conducted with 10 Dutch shelter facilities randomly allocated to an intervention and a control group. Homeless young adults were interviewed when entering the facility and when care ended. Repeated-measures analyses and logistic regression analyses were conducted by the principle of intention-to-treat framework (N = 251). Results: Improvements were demonstrated on quality of life; satisfaction with family relations, finances, and health; employed or in school; depression; care needs; autonomy; competence, and resilience in both conditions. A higher proportion of homeless young adults who received care according to Houvast were still receiving care at follow-up and successfully completed the trajectory compared to those who received care as usual. Conclusion: Homeless young adults seem to benefit from service provision in general. Further research on the effectiveness of Houvast is needed after sufficient model fidelity has been achieved.

Representation of the Cattell–Horn–Carroll Theory of Cognitive Abilities in the Factor Structure of the Dutch-Language Version of the WAIS-IV

The Cattell–Horn–Carroll (CHC) theory of cognitive abilities has been guiding in the revision of the Wechsler Adult Intelligence Scale–Fourth edition (WAIS-IV). Especially the measurement of fluid reasoning (Gf) is improved. A total of five CHC abilities are included in the WAIS-IV subtests. Using confirmatory factor analysis, a five-factor model based on these CHC abilities is evaluated and compared with the four index scores in the Dutch-language version of the WAIS-IV. Both models demonstrate moderate fit, preference is given to the five-factor CHC model both on statistical and theoretical grounds. Evaluation of the WAIS-IV according to CHC terminology enhances uniformity, and can be important when interpreting possible sources of index discrepancies. To optimally assemblage CHC and WAIS-IV, more knowledge of the interaction of abilities is needed. This can be done by incorporating intelligence testing in neuropsychological assessment. Using this functional approach contributes to a better understanding of an individual’s cognitive profile.

Severe fatigue in type 1 diabetes: Exploring its course, predictors and relationship with HbA1c in a prospective study

AIMS To prospectively identify the course of severe fatigue, its predictors and the relationship with HbA1c in patients with type 1 diabetes. METHODS 214 adult patients completed questionnaires on fatigue severity and fatigue-related factors at baseline. HbA1c was retrieved from medical records. After 43months, fatigue severity and HbA1c were reassessed in 194 patients. A logistic regression analysis was used to determine predictors of severe fatigue at follow-up with various cognitive-behavioral and clinical factors as potential predictors. The relationship between fatigue and HbA1c was investigated in a sub-analysis by differentiating between patients with suboptimal glucose control [HbA1c>7% (53mmol/mol)] and optimal glucose control [HbA1c⩽7% (53mmol/mol)]. RESULTS The prevalence of severe fatigue was 40% at baseline and 42% at follow-up. In three out of four severely fatigued patients at baseline (76%), severe fatigue persisted over time. More depressive symptoms, more pain, sleep disturbances, lower self-efficacy concerning fatigue, less confidence in diabetes self-care, more fatigue severity at baseline and more diabetes complications predicted severe fatigue at follow-up. Over time, HbA1c at baseline was positively associated with fatigue severity at follow-up in both groups (suboptimal glucose control: r=.18, p<.05; optimal glucose control: r=.09, p<.05). CONCLUSIONS About three quarters of fatigued[corrected] patients with type 1 diabetes suffer from persistent fatigue. Aside from the number of diabetes complications, no clinical factors explained the persistence of fatigue. HbA1c and fatigue were weakly associated in a sub-analysis. Since the strongest predictors of severe fatigue were cognitive-behavioral factors, behavioral interventions might be effective in decreasing fatigue.

Efficacy of Art Therapy in Individuals With Personality Disorders Cluster B/C: A Randomized Controlled Trial

Multidisciplinary treatment programs for patients with personality disorders (PDs) often include art therapy, but the efficacy of this intervention has hardly been evaluated. The objective of this study is to evaluate the effects of an art therapy intervention on psychological functioning of patients with a PD. In this randomized controlled trial, 57 adult participants diagnosed with a PD cluster B/C (SCID-II) were randomly assigned to either weekly group art therapy (1.5 hours, 10 weeks) or a waiting list group. Outcome measures OQ45, AAQ-II, and SMI were assessed at baseline, at post-test (10 weeks after baseline), and at follow-up (5 weeks after post-test). The results show that art therapy is an effective treatment for PD patients because it not only reduces PD pathology and maladaptive modes but it also helps patients to develop adaptive, positive modes that indicate better mental health and self-regulation.

Vocabulary development in children with hearing loss: The role of child, family, and educational variables

In the present study we examined the effect of hearing status on reading vocabulary development. More specifically, we examined the change of lexical competence in children with hearing loss over grade 4-7 and the predictors of this change. Therefore, we used a multi-factor longitudinal design with multiple outcomes, measuring the reading vocabulary knowledge in children with hearing loss from grades 4 and 5, and of children without hearing loss from grade 4, for 3 years with two word tasks: a lexical decision task and a use decision task. With these tasks we measured word form recognition and (in)correct usage recognition, respectively. A GLM repeated measures procedure indicated that scores and growth rates on the two tasks were affected by hearing status. Moreover, with structural equation modeling we observed that the development of lexical competence in children with hearing loss is stable over time, and a childs lexical competence can be explained best by his or her lexical competence assessed on a previous measurement occasion. If you look back, differences in lexical competence among children with hearing loss stay unfortunately the same. Educational placement, use of sign language at home, intelligence, use of hearing devices, and onset of deafness can account for the differences among children with hearing loss.

Personality and Psychopathology Mapping the MMPI-2-RF on Cloninger’s Psychobiological Model of Personality

This study investigates the relationship between the Minnesota Multiphasic Personality Inventory-2-Restructured Form (MMPI-2-RF) and the Temperament and Character Inventory (TCI) in a combined data set (N = 491) of patients with a broad range of psychiatric disorders (n = 286) as well as alcohol use disorder (n = 205). We examined bivariate correlations between both measures. The MMPI-2-RF scales relate to the TCI dimensions as was hypothesized, and relationships between both measurements were largely similar for psychiatric patients and alcohol-dependent patients. Theoretical and clinical implications are considered.