College student and Amazon's Mechanical TURK (MTURK) samples are regularly utilized in trauma research. Recent literature, however, has criticized these samples for not being generalizable to the general U.S. population. Measurement invariance (MI) using confirmatory factor analyses (CFA), is rarely utilized in trauma research, even though the analysis can determine whether groups are invariant across factor structure, factor loadings, item intercepts, and residual error variances on a given measure of PTSD symptom severity. The purpose of this study was to determine whether college student (n = 255) and MTURK (n = 316) samples are invariant on the PCL-5. Model fit indices indicated the 7-factor Hybrid model was the best fitting model, but the 6-factor anhedonia model was the most parsimonious model. Both models demonstrated equivalence in factor structures (configural invariance), factor loadings (metric invariance), intercepts (scalar invariance), and residuals (strict invariance), indicating MTURK and college student samples are similar in regards to PTSD symptom severity. These findings provide evidence that these groups can be combined in future studies to increase sample size for trauma research. Only the Anhedonia factor exhibited mean differences between groups, which may be related to true differences between college students and MTURK survey-takers. Thus, there is …

The purpose of this study was twofold: (a) assess cycling coaches' trauma-related competencies, as measured by trauma knowledge (i.e., trauma-specific education, familiarity with post-traumatic stress [PTS]), stigma of persons with PTS (i.e., fear/dangerousness, help/interact, forcing treatment, negative emotions), and interpersonal skills (i.e., self-reported emotional intelligence, perceived quality of coach-athlete relationships); and (b) examine the influence that trauma knowledge and stigma of persons with PTS has on coaching efficacy specific to coaching trauma-impacted athletes (i.e., trauma-informed coaching efficacy), after controlling for general coaching experience. Descriptive statistics indicated the majority of coaches had no trauma-specific education, a high degree of familiarity with PTS, and a low level of stigma via four attribution variables. Moreover, participants highly appraised their own emotional intelligence, the quality of their coach-athlete relationships, and their trauma-informed coaching efficacy. A hierarchical regression analyses indicated that familiarity with PTS helped to explain additional variance in trauma-informed coaching efficacy over and above demographic and general coach experience variables. The study establishes trauma-informed coaching as a distinct area of research and highlights the need for improved continuing education opportunities for coaches related to psychological trauma and PTS.

In response to concerns about using only standardized multiple-choice assessments, some school districts have moved to using alternative ratings of student achievement with authentic assessments. However, such assessments are often limited in terms of the psychometric validity data supporting their use. The present study mixed quantitative and qualitative methods to examine the validity, development, and refinement of an authentic rating approach currently in use in middle school mathematics classes from a large suburban school district in the Southwest United States. A sample of teachers (n = 14), rated students (n = 110) using a pilot rubric of 187 items. Analyses resulted in a 32-item rubric with 20 themes and 9 factors. Results from a G-study revealed the facet that best explained variance in student scores was the interaction between raters and assessment units, as well as students and assessment units. As part of the development of the assessment, a content validity exercise revealed 18% of the rubric items as below average quality. Findings highlight the need to enhance contextualization of rubrics, use a strategy of assessment that includes contextualized and decontextualized assessment, and to investigate the role of utilization deficiency in explaining low student scores.

This study investigated the association between Chinese immigrant mothers' reports of parenting stress and two sources of parenting support: fathers' support and social support. Five independent-samples t-tests were first computed to examine whether there were discrepancies between Chinese immigrant mothers' and fathers' perceptions of fathers' involvement across five domains of fathers' involvement in early child care: (1) fathers' warmth and attunement; (2) control and process responsibility; (3) emotional involvement; (4) indirect care; and (5) positive engagement. Then regression analyses were conducted to understand how and to what extent the five domains of father involvement reported by mothers and mothers' perceived social support influenced Chinese immigrant mothers' parenting stress. In addition, an indirect effects of mothers' self-efficacy on the relation between fathers' involvement and mothers' parenting stress was also tested. The results showed that Chinese mothers reported significantly greater level of fathers' emotional involvement than the same type of involvement reported by Chinese immigrant fathers. This study also found that fathers' emotional involvement reported by Chinese immigrant mothers was a significant predictor of Chinese immigrant mothers' parenting stress. However, mothers' perceived social support was a stronger predictor, accounting for more variance in their parenting stress than their reports of fathers' involvement …

Previous research has individually assessed parallel analysis and minimum average partial for factor retention in exploratory factor analysis using ordinal variables. The current study is a comprehensive simulation study including the manipulation of eight conditions (type of correlation matrix, sample size, number of variables per factor, number of factors, factor correlation, skewness, factor loadings, and number of response categories), and three types of retention methods (minimum average partial, parallel analysis, and empirical Kaiser criterion) resulting in a 2 × 2 × 2 × 2 × 2 × 3 × 3 × 4 × 5 design that totals to 5,760 condition combinations tested over 1,000 replications each. Results show that each retention method performed worse when utilizing polychoric correlation matrices. Moreover, minimum average partials are quite sensitive to factor loadings and overall perform poorly compared to parallel analysis and empirical Kaiser criterion. Empirical Kaiser criterion performed almost identical to parallel analysis in normally distributed data; however, performed much worse under highly skewed conditions. Based on these findings, it is recommended to use parallel analysis utilizing principal components analysis with a Pearson correlation matrix to determine the number of factors to retain when dealing with ordinal data.