Shuyan Sun

Meta-analysis of Cohen’s kappa

Cohen’s κ is the most important and most widely accepted measure of inter-rater reliability when the outcome of interest is measured on a nominal scale. The estimates of Cohen’s κ usually vary from one study to another due to differences in study settings, test properties, rater characteristics and subject characteristics. This study proposes a formal statistical framework for meta-analysis of Cohen’s κ to describe the typical inter-rater reliability estimate across multiple studies, to quantify between-study variation and to evaluate the contribution of moderators to heterogeneity. To demonstrate the application of the proposed statistical framework, a meta-analysis of Cohen’s κ is conducted for pressure ulcer classification systems. Implications and directions for future research are discussed.

Multitrait-multimethod analysis of the strengths and difficulties questionnaire in young Asian American children

The Strengths and Difficulties Questionnaire (SDQ) is a widely used psychopathology screening tool that measures children’s emotional symptoms, peer problems, conduct problems, hyperactivity/inattention, and prosocial behavior. Previous psychometric studies of the SDQ focused primarily on older children in Western cultures and suffered from several methodological limitations. This study examined the reliability, factor structure, convergent, and discriminant validity of the SDQ by focusing on young Asian American children and using more rigorous methods. The five-factor structure of the SDQ was confirmed by confirmatory factor analysis. The ω coefficients indicated adequate reliability for all subscales except parent-rated peer problems and conduct problems. The correlated trait–correlated method minus one multitrait–multimethod model provided evidence for convergent validity and discriminant validity of all subscales except for conduct problems relative to hyperactivity/inattention. This study provided new evidence for the psychometric properties of the SDQ in young children and cultural suitability of the SDQ for Asian Americans.

Investigating the Accuracy of Three Estimation Methods for Regression Discontinuity Design.

Regression discontinuity design is an alternative to randomized experiments to make causal inference when random assignment is not possible. This article first presents the formal identification and estimation of regression discontinuity treatment effects in the framework of Rubins causal model, followed by a thorough literature review of three major methods for estimating regression discontinuity effects. The authors conducted a Monte Carlo simulation to compare the accuracy of 3 estimation methods and evaluate the effects of sample sizes, cutoff score locations, and distribution assumptions on the accuracy of parameter estimates. Although all 3 methods can produce reasonably accurate parameter estimates under various manipulated data conditions, extreme cutoff scores tend to introduce large parameter estimate biases when the variance of the outcome variable differs across groups. Implications and directions for further research are discussed.

Outcomes and Processes in the Meyerhoff Scholars Program: STEM PhD Completion, Sense of Community, Perceived Program Benefit, Science Identity, and Research Self-Efficacy

Studies found that African-American Meyerhoff Scholars were 4.8 times more likely to complete STEM PhDs than comparison sample students (study 1) and that perceived benefits of program components fully mediated the relationship between sense of community and both science identity and research self-efficacy (study 2).

Confirming the Multidimensionality of Psychologically Controlling Parenting among Chinese-American Mothers: Love Withdrawal, Guilt Induction, and Shaming.

Despite the theoretical conceptualization of parental psychological control as a multidimensional construct, the majority of previous studies have examined psychological control as a unidimensional scale. Moreover, the conceptualization of shaming and its associations with love withdrawal and guilt induction are unclear. The current study aimed to fill these gaps by evaluating the latent factor structure underlying 18 items from Olsen et al. (2002) that were conceptually relevant to love withdrawal, guilt induction, and shaming practices in a sample of 169 mothers of Chinese-American preschoolers. A multidimensional three-factor model and bi-factor model were specified based on our formulated operational definitions for the three dimensions of psychological control. Both models were found to be superior to the unidimensional model. In addition, results from the bi-factor model and an additional second-order factor model indicated that psychological control is essentially empirically isomorphic with guilt induction. Although love withdrawal and shaming factors were also fairly strong indicators of psychological control, each exhibited important additional unique variability and mutual distinctiveness. Implications for the conceptualization of love withdrawal, guilt induction, and shaming as well as directions for future studies are discussed.

Aptitude-treatment interaction effects on explicit rule learning: A latent growth curve analysis

Finding the match between individuals and educational treatments is the aim of both educators and the aptitude-treatment interaction research paradigm. Using the latent growth curve analysis, the present study investigates the interaction between the type of explicit instructional approaches (deductive vs. explicit-inductive) and the level of foreign language aptitude (high vs. low) in the learning of explicit grammar rules. The results indicate that on the whole the two equally explicit instructional approaches did not differentially affect learning performance. However, when the level of language aptitude, measured by grammatical sensitivity, associative memory, and memory for text (with the last variable being the best measure), was taken into account, low-aptitude learners performed significantly better with the deductive instruction, in the sentence-correction tests. The interaction effects of equally explicit instructional approaches suggest the need for considering aptitude-treatment interaction to maximize learners’ potential for success in second language learning.

The association between sleep dysfunction and psychosis-like experiences among college students

Sleep problems are prominent and pervasive clinical issues experienced by many people with psychotic disorders, often causing distress and functional impairment. Sleep problems are also related to psychosis-like experiences (PLE; non-diagnosable phenomenon such as transient perceptual disturbances, unusual thoughts, periodic suspiciousness) in epidemiological studies. Prior studies in this field have used brief measures that precluded the ability to test (1) whether risk for psychosis-like experiences are related to specific sub-types of sleep disturbance, and (2) whether sleep disturbance is specifically related to clinically significant (i.e., distressing) psychosis-like experiences. The current project examined the relation between specific sleep issues, and PLEs and distress associated with PLEs, in a college sample. Participants (N=420) completed the Prodromal Questionnaire-Brief (PQ-B), which assesses PLEs and associated distress, and the Iowa Sleep Disturbances Inventory - extended version (ISDI-E), which assesses thirteen separate disturbed sleep domains. Symptoms of fragmented sleep, sleep hallucinations, and night anxiety significantly correlated with PLEs, and several sleep domains were significantly associated with PLE-related distress.

A methodological review of statistical methods for handling multilevel non-nested longitudinal data in educational research

As applications of multilevel modelling in educational research increase, researchers realize that multilevel data collected in many educational settings are often not purely nested. The most common multilevel non-nested data structure is one that involves student mobility in longitudinal studies. This article provides a methodological review of three statistical methods for handling student mobility in longitudinal studies: a multilevel approach, a cross-classified approach, and a cross-classified multiple membership approach. The strengths and weaknesses of each approach and the essential differences between the three approaches are discussed. The Early Childhood Longitudinal Study Kindergarten Cohort data are analysed to demonstrate the differences in parameter estimates and statistical inference between the three approaches. Potential applications of the three approaches in educational research and beyond and directions for further methodological investigations are discussed.

Adolescents’ Engagement in Ethnic Harassment: Prejudiced Beliefs in Social Networks and Classroom Ethnic Diversity

Research on ethnic victimization to date has done little to identify the reasons why adolescents victimize their peers due to their ethnic background. To address this limitation, we examined: (1) the extent to which prejudiced attitudes within adolescents’ close and larger social networks determine their engagement in ethnic harassment, and (2) the extent to which classroom ethnic diversity plays a role in any such link. Our sample included 902 Swedish adolescents (Mage = 14.40, SD = .95; 50.3% girls). We found that Swedish adolescents who held negative attitudes toward immigrants or who were surrounded by prejudiced peers were more likely to be involved in ethnic harassment, particularly in classrooms with high ethnic diversity. Adolescents in classrooms with a high anti-immigrant climate were more likely to harass their immigrant peers. These findings suggest that prejudiced beliefs in youth social networks put young people at risk of engaging in ethnic harassment, particularly in ethnically diverse classrooms.

Confirmatory factor analyses of the Parent Attributions Questionnaire among Asian immigrant mothers

The present study examined the factor structure of maternal attributions regarding their caregiving experiences, measured by the Parent Attributions Questionnaire, among Asian immigrant mothers. Chinese and Korean immigrant mothers (N = 333, mean age (M age) = 36.79 years, standard deviation (SD) = 4.79) with preschool children (M age = 4.43 years, SD = 1.31) reported their attributions regarding the reasons for their success and failure at seven daily caregiving tasks, and demographic information. Three structure models were tested: the unidimensional structure model; the internal–external structure model; and the controllable–uncontrollable structure model. Results revealed that the controllability-based structure fitted the data better than the unidimensional or locus-based structure of parental attributions. Metric and partial scalar invariance of the controllability-based model were established between Chinese and Korean immigrant mothers. Asian immigrant mothers focused on whether the causes for the parenting outcomes were controllable or uncontrollable when attributing the causes of their caregiving experiences. The superiority of the controllable–uncontrollable over the internal–external structure model in this sample may reflect Asian immigrant mothers’ Confucian-based heritage cultural emphasis on their responsibility for creating optimal childrearing for their children.