Yuk Fai Cheong

Measurement and structural models for children's problem behaviors

This article considers an analytic strategy for measuring and modeling child and adolescent problem behaviors. The strategy embeds an item response model within a hierarchical model to define an interval scale for the outcomes, to assess dimensionality, and to study how individual and contextual factors relate to multiple dimensions of problem behaviors. To illustrate, the authors analyze data from the primary caregiver ratings of 2,177 children aged 9-15 in 79 urban neighborhoods on externalizing behavior problems using the Child Behavior Checklist 4-18 (T. M. Achenbach, 1991a). Two subscales, Aggression and Delinquency, are highly correlated, and yet unidimensionality must be rejected because these subscales have different associations with key theoretically related covariates.

Teaching As a Nonroutine Task: Implications for the Management of Schools.

Using contingency theory as a frame work for analysis, this article sought to accomplish three main objectives: (a) to examine workplace conditions that make teaching in high schools nonroutine; (b) to investigate the hypothesis that when the work of teachers becomes nonroutine, organic forms of management arise in high schools to coordinate and control instruction; and (c) to investigate whether organic forms of management can reasonably be expected to enhance the effectiveness of teachers by promoting job-related learning. The findings suggest that perceived variability in students and the disciplinary specialization of teachers affect the extent to which teachers report their work as nonroutine. The findings also demonstrate that nonroutine teaching is associated with the development of organic management in high schools. However, there is little evidence to support the idea that organic management of instruction leads to more job-related learning by teachers. The implications of these findings for research on teaching, the organizational design of schools, and school effectiveness are discussed.

Synthesizing Results from the Trial State Assessment

Using data collected under the Trial State Assessment (TSA) of the National Assessment of Educational Progress (NAEP), this article describes and illustrates a two-stage statistical model for investigating state-to-state variation in mathematics achievement. At the first stage, within each state, a two-level hierarchical linear model is estimated via maximum likelihood. At the second stage, results are combined across states using Bayesian estimation implemented via the Gibbs sampler. The results reveal considerable state-to-state heterogeneity in mathematics proficiency, but most heterogeneity is explainable on the basis of covariates defined on students, teachers, and schools. The findings suggest that interest in state comparisons might productively focus on state differences in policy-relevant correlates of proficiency rather than on state differences in mean proficiency. The analytical approach can be applied in other cases where data are dense at the lower level of a hierarchy but thin at the higher level.

Psychometric Analysis of Computer Science Help-Seeking Scales.

The purpose of this studywas to develop scales to assess instrumental help seeking, executive help seeking, perceived benefits of help seeking, and avoidance of help seeking and to examine their psychometric properties by conducting factor and reliability analyses. As this is the first attempt to examine the latent structures underlying the measured items, the authors conducted exploratory factor analyses. In addition, they also examined the relationship between the help-seeking scales and motivation and achievement constructs frequently used in the study of academic motivation. Results supported the continued use and development of the newscales, which can be adapted to assess help-seeking behavior across varied academic domains.

Motivation and Academic Help-Seeking in High School Computer Science

The primary objective of this study was to determine the degree to which academic motivation predicted the executive help-seeking, instrumental help-seeking, perceived benefits of help-seeking, and avoidance of help-seeking of high school students enrolled in computer science (n = 314). Task goals were positively associated with instrumental help-seeking and perceiving the benefits of help-seeking and negatively associated with executive help-seeking; performance-avoid goals were negatively associated with instrumental help-seeking and positively associated with avoiding help-seeking. Controlling for motivation and computer science competence, girls were more likely to seek instrumental help and to perceive the benefits of help-seeking, and African American students were more likely to seek help than were White students or Asian American students. Despite possessing equal computer science skills, girls reported lower self-efficacy, self-concept, self-efficacy for self-regulation, and value than did boys.

Efficiency and Robustness of Alternative Estimators for Two- and Three-Level Models: The Case of NAEP.

This article investigates the efficiency and robustness of alternative estimators of regression coefficients for three-level data. To study student achievement, researchers might formulate a standard regression model or a hierarchical model with a two- or three-level structure. Having chosen the model, the researchers might employ either a model-based or a robust estimator of the standard errors. A simulation study showed that, as expected, the hierarchical model analyses produced more efficient point estimates than did analyses that ignored the covariance structure in the data, even when the normality assumption was violated. When samples were fairly large, the three-level analyses produced sound standard errors. In contrast, single-level analysis yielded seriously biased standard errors for coefficients defined at level 3 and level 2; and two-level analysis yielded biased standard errors for coefficients defined at level 2. These biases in standard error estimates were largely corrected by robust variance estimation. Implications of the results for analyzing NAEP and other large-scale surveys such as the Early Childhood Longitudinal Study (ECLS) and the Third International Mathematics and Science Study (TIMSS) are discussed.

Using Explanatory Item Response Theory Modeling to Investigate Context Effects of Differential Item Functioning for Students With Disabilities

To address whether or not modifications in test administration influence item functioning for students with disabilities on a high-stakes statewide problem-solving assessment, a sample of 868 students (with and without disabilities) from 74 Georgia schools were randomly assigned to one of three testing conditions (resource guide, calculator, or standard administration). The authors describe a conceptual framework for investigating measurement invariance, specifically differential item functioning within the context of assessing students with disabilities. Specifically, we illustrate how the hierarchical generalized linear model, an explanatory model that incorporates item response models into hierarchical models in multilevel settings, can be used to explore issues of measurement invariance traditionally addressed using descriptive item response models, that is, the many-facet Rasch model. Results obtained from the two approaches are reported and discussed.

Multilevel Rasch Models

Over the past few years, several studies have investigated and demonstrated the relationships between generalized linear mixed models (GLIMM) and item response modeling. Some benefits associated with this GLIMM-based modeling framework include the modeling of nested structure of data, such as examinees nested within schools (Kamata, 2001), of multidimensional measures (Cheong & Raudenbush, 2000), and of wider class of item response models, such as 2PL item response model (Rijmen et al., 2003).

Women's age at first marriage and postmarital agency in Egypt

Early - or child - marriage (before age 18) may diminish womens ability to exercise agency, or their capacity to act upon their goals. Using a propensity score adjustment approach, we analyzed data from 2394 married women ages 35-49 years who participated in the 2006 Egypt Labor Market Panel Survey (ELMPS). We examined whether womens first marriage at age 18 or older was associated with their post-marital agency, measured in terms of their influence in family decisions, freedom of movement in public spaces, and unfavorable views about intimate partner violence against wives. In bivariate analyses, womens age at first marriage was positively associated with their decision-making and more equitable gender attitudes. However, once we controlled for selection into age-at-first-marriage groups, there were no significant differences between the two age-at-first-marriage groups in any dimension of womens agency. We examined the sensitivity of the non-significant age-at-first-marriage effects to possible violations of the strong ignorability assumption and the results did not alter our conclusions. The assumption that womens age at first marriage is a proxy for their post-marital agency, as defined here, warrants further study.

Community Economic Status and Intimate Partner Violence Against Women in Bangladesh: Compositional or Contextual Effects?

In this research, we used a multi-level contextual-effects analysis to disentangle the household- and community-level associations between income and intimate partner violence (IPV) against women in Bangladesh. Our analyses of data from 2,668 women interviewed as part of the World Health Organization (WHO) multi-country study on women’s health and domestic violence against women showed that household income was negatively associated with women’s risk of experiencing IPV. Controlling for residence in a low-income household, living in a low-income community was not associated with women’s risk of experiencing IPV. These results support a household-level, not community-level, relationship between income and IPV in Bangladesh.