Multilevel Structural Modeling Research Articles

ASIAN BUSINESS CYCLE SYNCHRONIZATION

AbstractThis paper develops a multilevel structural factor model to study international output comovement and its underlying driving forces. Our method combines a structural vector autoregression with a multilevel factor model, which helps us understand the economic meaning of the estimated factors. Using quarterly data of real GDP growth covering 9 emerging Asian economies and G‐7 countries, we estimate a global supply factor, a global demand factor, and group supply and demand factors for each group of the economies. We find that although the role of the global factors has intensified over the past 15 years for most of the economies, output fluctuations in Asia have remained less synchronized with the global factor than those in the industrial countries. The Asian regional factors have become increasingly important in tightening the interdependence within the region over time. Therefore, although emerging Asian economies cannot ‘decouple’ completely from the advanced economies, they have, nonetheless, sustained a strong independent cycle among themselves. We also find that synchronized supply shocks contributed more to the observed synchronization in output fluctuations among the Asian economies than demand shocks. This points to the role of productivity enhancement and transmission of other supply shocks through, for example, vertical trade integration, rather than dependence on external demand, as the primary source of business cycle synchronization in emerging Asia.

The global long-range order of quasi-periodic patterns in Islamic architecture

Three decades after their discovery, the unique long-range structure of quasicrystals still poses a perplexing puzzle. The fact that some ancient Islamic patterns share similar quasi-periodic symmetries has prompted several scientists to investigate their underlying geometry and construction methods. However, available structural models depend heavily on local rules and hence they were unable to explain the global long-range order of Islamic quasi-periodic patterns. This paper shows that ancient designers, using simple consecutive geometry, have resolved the complicated long-range principles of quasi-periodic formations. Derived from these principles, a global multi-level structural model is presented that is able to describe the global long-range translational and orientational order of quasi-periodic formations. The proposed model suggests that the position of building units, locally and globally, is defined by one framework, and not tiled based on local rules (matching, overlapping or subdividing). In this way, quasi-periodic formations can grow rapidly ad infinitum without the need for any defects or mismatches. The proposed model, which presents a novel approach to the study of quasi-periodic symmetries, will hopefully provide a deeper understanding of the structure of quasicrystals at an atomic scale, allowing scientists to achieve improved control over their composition and structure.

Person-Organization Value Congruence: How Transformational Leaders Influence Work Group Effectiveness

Using multilevel structural equations modeling, we examine the extent to which the influence of transformational leadership on work group effectiveness flows through follower perceptions of person-organization or person-supervisor value congruence. Results indicate that the group-level effect of transformational leadership on work group effectiveness was fully accounted for by the group-level impact of transformational leadership on follower perceptions of person-organization value congruence, not by its impact on follower perceptions of person-supervisor value congruence. These results are discussed in the context of leadership as a “sense-making” process and the practical barriers faced by transformational leaders in modern organizations.

Asian Business Cycle Synchronisation

This paper develops a multi-level structural factor model to study international output comovement and its underlying driving forces. Our method combines a structural VAR with a multi-level factor model, which helps us understand the economic meaning of the estimated factors. Using quarterly data of real GDP growth covering nine emerging Asian economies and G-7 countries, we estimate a global supply factor, a global demand factor, and group supply and demand factors for each group of the economies. We find that, while the role of the global factors has intensified over the past fifteen years for most of the economies, output fluctuations in Asia have remained less synchronised with the global factor than those in the industrial countries. The Asian regional factors have become increasingly important in tightening the interdependence within the region over time. Thus while emerging Asian economies cannot decouple completely from the advanced economies, they have nonetheless sustained a strong independent cycle among themselves. We also find that synchronised supply shocks contributed more to the observed synchronisation in output fluctuations among the Asian economies than demand shocks. This points to the role of productivity enhancement through vertical trade integration, rather than dependence on external demand, as the primary source of business cycle synchronisation in emerging Asia.

Examining relationships among elementary schools' contexts, leadership, instructional practices, and added-year outcomes: a regression discontinuity approach

This study proposes and tests a multilevel structural model of school context, composition, and school leadership on school instructional practices and outcomes in elementary schools in a western state in the United States. We focus on direct and indirect relationships implied in our proposed model using an “added year of schooling” in reading and math as our primary school-level outcomes. Added-year effects, which result from a regression discontinuity design, represent a relatively new approach for describing how school factors influence outcomes. Our results suggest that, net of context and composition factors, improvement-focused school leadership directly affected subsequent school instructional practices and, in turn, instructional practices affected added-year outcomes. We discuss the findings in terms of their theoretical and practical implications for conducting further educational effectiveness research.

Preparing for prison?

American schools increasingly define and manage the problem of student discipline through a prism of crime control. Most theoretical explanations fail to situate school criminalization in a broader structural context, to fully explain its spatio-temporal variations, and to specify the processes and subjectivities that mediate between structural and legal forces and the behavior of school actors. A multilevel structural model of school criminalization is developed which posits that a troubled domestic economy, the mass unemployment and incarceration of disadvantaged minorities, and resulting fiscal crises in urban public education have shifted school disciplinary policies and practices and staff perceptions of poor students of color in a manner that promotes greater punishment and exclusion of students perceived to be on a criminal justice `track'.

Personnalité, caractère et tempérament : la structure translinguistique des traits

We review progress on an important scientific issue--how attributes of personality and character can best be organized and structured. We explain the rationale for studies of the language of personality, and then review the most salient findings from lexical studies of person-descriptors in 16 languages. Using a wide range of criteria for the value of a structural model, we compare prominent lexically derived models. We conclude that one- and two-factor structures are not only the most parsimonious but also the most easily replicated across cultures. Although structures with more factors (e.g. the Big Five) have a predictive advantage, there is currently uncertainty over whether a lexical model of five, six, or seven factors is more cross-culturally generalizable. It will be desirable to develop a multi-level structural model with specific facets as well as independent, broad factors, and a structural model that goes beyond conventionally defined personality traits to include a wider range of individual differences.

People are variables too: Multilevel structural equations modeling.

The article uses confirmatory factor analysis (CFA) as a template to explain didactically multilevel structural equation models (ML-SEM) and to demonstrate the equivalence of general mixed-effects models and ML-SEM. An intuitively appealing graphical representation of complex ML-SEMs is introduced that succinctly describes the underlying model and its assumptions. The use of definition variables (i.e., observed variables used to fix model parameters to individual specific data values) is extended to the case of ML-SEMs for clustered data with random slopes. Empirical examples of multilevel CFA and ML-SEM with random slopes are provided along with scripts for fitting such models in SAS Proc Mixed, Mplus, and Mx. Methodological issues regarding estimation of complex ML-SEMs and the evaluation of model fit are discussed. Further potential applications of ML-SEMs are explored.

Leadership and Individual Principal-Teacher Relationships in Schools

Transformational approaches to leadership have increasingly been advocated for schools. Research evidence suggests that the effect of leadership on student learning outcomes is mediated by school conditions such as goals, structure, people, and school culture. Using the combined approach of multilevel analysis and structural modeling, this study investigated the relationships of leadership with school learning culture. Data were collected from a random sample of 373 teachers. The results suggested relationships between leadership and school learning culture did exist, and they highlight the importance of individual principal-teacher relationships in schools.

A Model-Based Approach to Validating Education Indicators Using Multilevel Structural Equation Modeling

This article considers an approach to validating the selection of education indicators by incorporating them into a multilevel structural model and using the estimates from that model to engage in policy-relevant simulations. Multilevel structural equation modeling was applied to a subsample of the first follow-up of the National Education Longitudinal Study of 1988 ( National Center for Education Statistics, 1988 ) to demonstrate the potential of this approach. Focus of attention was on science education indicators. A within-school model of science achievement was linked to a between-school model of the academic press of the school. Separate estimation of these models revealed adequate fit to the data after minor modifications. The multilevel model also showed adequate fit to the data. Utilizing the reduced form of the full multilevel model, predictive validity of the model was studied by gauging movements in various outcome indicators as a function of changes in policy-relevant input indicators. The article closes with a discussion of the limitations of the proposed modeling approach, the potential for future model development, and the implications of this approach for quantitative modeling within the domain of education policy.