Direct Schmid–Leiman Transformations and Rank-Deficient Loadings Matrices

The Schmid–Leiman (S–L; Psychometrika 22: 53–61, 1957) transformation is a popular method for conducting exploratory bifactor analysis that has been used in hundreds of studies of individual differences variables. To perform a two-level S–L transformation, it is generally believed that two separate factor analyses are required: a first-level analysis in which k obliquely rotated factors are extracted from an observed-variable correlation matrix, and a second-level analysis in which a general factor is extracted from the correlations of the first-level factors. In this article, I demonstrate that the S–L loadings matrix is necessarily rank deficient. I then show how this feature of the S–L transformation can be used to obtain a direct S–L solution from an unrotated first-level factor structure. Next, I reanalyze two examples from Mansolf and Reise (Multivar Behav Res 51: 698–717, 2016) to illustrate the utility of ‘best-fitting’ S–L rotations when gauging the ability of hierarchical factor models to recover known bifactor structures. Finally, I show how to compute direct bifactor solutions for non-hierarchical bifactor structures. An online supplement includes R code to reproduce all of the analyses that are reported in the article.     

A Cross-Sectional and Longitudinal Test of the Low Sensitivity to Threat and Affiliative Reward (STAR) Model of Callous-Unemotional Traits Among Spanish Preschoolers

Research on Child and Adolescent Psychopathology - Despite advances in the theoretical and empirical literature, a better understanding of the etiological determinants of callous-unemotional (CU)...     

Correlates of neonatal morbidity: maternal characteristics and family resources

The association of mental characteristics, family resources, and receipt of prenatal care with infant birth weight, gestational age, and Apgar scores was assessed in a retrospective review of an extant computerized data base of admissions to a neonatal intensive care unit (NICU; n = 3,818). Prediction variables included mother's age, marital status, health insurance, parity, prenatal health problems, and prenatal drug use. Less favorable infant birth status was associated with prenatal health problems, being single, not having private insurance, obtaining a prior elective abortion, prenatal drug use, and not receiving prenatal care. Stepwise multiple regressions of predictor variables to birth weight and lack of prenatal care were performed. Low birth weight was associated with being single, receiving no prenatal care, and being Black. Lack of prenatal care was associated with mother's age, high parity, prenatal health problems, and prenatal illicit drug use. Pediatric interventions assessing maternal social and behavioral characteristics are proposed.     

Correlates of parenting on a neonatal intensive care unit: maternal characteristics and family resources

The association of maternal characteristics, family resources, and receipt of prenatal care with parenting behaviors observed in a neonatal intensive care unit (NICU; n = 383) was assessed. The parenting behavior of mothers not receiving prenatal care (n = 128) was compared to that of mothers of the preceding and subsequent admissions (n = 256) by retrospective chart review. Parenting variables included frequency of visits to the NICU and evaluative ratings of parents' involvement with their infant. Parenting of the no-prenatal-care group was significantly less favorable than the control on all comparisons. Factor analysis supported a priori grouping of parenting variables. A stepwise multiple regression of maternal and family characteristics to the factor-derived variable, parenting, showed significant contributions for prenatal drug use and father involvement. Pediatric interventions assessing maternal social and behavioral characteristics are proposed.     

A comparison of the classification capabilities of the 1-dimensional kohonen neural network with two pratitioning and three hierarchical cluster analysis algorithms

Neural Network models are commonly used for cluster analysis in engineering, computational neuroscience, and the biological sciences, although they are rarely used in the social sciences. In this study we compare the classification capabilities of the 1-dimensional Kohonen neural network with two partitioning (Hartigan and Späthk-means) and three hierarchical (Ward's, complete linkage, and average linkage) cluster methods in 2,580 data sets with known cluster structure. Overall, the performance of the Kohonen networks was similar to, or better than, the performance of the other methods.