Multilevel Modeling Strategy Research Articles

Ambivalence, Political Engagement and Context

Scholars of American politics have generally found a negative relationship between ambivalence and political engagement. This study explores such conclusions and argues that the effect of ambivalence on engagement varies according to electoral context. Using a multi-level modeling strategy, I find that ambivalence has a significant overall effect on political engagement for citizens in the United States but a lesser overall impact for citizens in Great Britain. Yet by allowing the slope and the intercept of ambivalence to vary across parliamentary districts, I find that ambivalence has asymmetrical effects on political engagement within Britain. I conclude by arguing that ambivalence essentially operates in a differential manner across electoral contexts and provide preliminary evidence as to why this is.

Smoking, Ethnic Residential Segregation, and Ethnic Diversity: A Spatio-temporal Analysis

Ethnic residential segregation is a profound social divide in many societies. In the health arena, U.S. work has been influential in demonstrating the impact of ethnic residential segregation on child health outcomes, showing how it can compound other forms of disadvantage. This article builds on and extends this research by examining the transferability of conclusions concerning the health impact of ethnic residential segregation to a non-U.S. context and to the field of health behavior. Using complete adult population data from the 1996 and 2006 New Zealand Censuses of Population geocoded to local and urban area levels, we examine smoking prevalence and cessation in relation to ethnic segregation and diversity. The article employs a repeated cross-sectional multilevel modeling strategy with smoking and cessation as outcome variables. The differential impact of segregation and diversity on smoking behavior by different ethnic groups is considered, taking into account the confounding effect of socioeconomic status and demographic variation. Conclusions suggest that Māori isolation has little overall effect but ethnic diversity has some relevance. Individual ethnic status and area-level deprivation are more important in understanding smoking behavior.

Mothers' differential parenting and children's behavioural outcomes: Exploring the moderating role of family and social context

Exploring an ecological framework, the goal of the present study was to examine the moderating role of contextual risk on the relationship between differential parenting and children's socio‐emotional outcomes. The present study was conducted with a sample of 920 children nested within 397 families (including up to four children per family). Using a multivariate multilevel modelling strategy, we examined a range of contextual risks including maternal personal factors (educational level, family‐average parenting, depression, history of abuse and teen motherhood), parental and family factors (lone‐parent and step‐parent families and marital conflict) and home and community factors (household disorganization neighbourhood quality, collective efficacy and personal safety/victimization). Overall, the moderating role of context was found to be more evident for differential negativity than for differential positivity. Moreover, higher levels of contextual risk were associated with stronger links between differential parenting and child outcomes, with more deleterious effects for the disfavoured child. This same effect was seen when individual risks were aggregated into a cumulative risk index. The exception to this pattern was household disorganization in some cases, children showed more sensitivity to differential parenting at low levels of household disorganization. Copyright © 2012 John Wiley & Sons, Ltd.

Juvenile Penalties for “Lawyering Up”

The presence of counsel for juveniles in the courtroom seems advantageous from a due process perspective, yet some studies suggest that juveniles receive harsher dispositions when represented by an attorney. This study tested whether a “counsel penalty” existed regardless of attorney type and, guided by prior sentencing literature, used a more comprehensive model to determine the influence of extralegal and contextual factors that may amplify the counsel penalty. Utilizing official data from a Northeastern state in a multilevel modeling strategy, this study found that regardless of the type of counsel retained, harsher sentences were received as compared with cases in which a juvenile was not represented by counsel even after controlling for offense type. Moreover, minority youth with public defenders and males with private counsel received harsher sentences while community characteristics did not appear to have a significant influence on sentencing decisions.

Using Integrated Nested Laplace Approximations for the Evaluation of Veterinary Surveillance Data from Switzerland: A Case-Study

SummarySpatiotemporal disease mapping models have been used extensively to describe the pattern of surveillance data. They are usually formulated in a hierarchical Bayesian framework and posterior marginals are not available in closed form. Hence, the standard method for parameter estimation is Markov chain Monte Carlo algorithms. A new method for approximate Bayesian inference in latent Gaussian models using integrated nested Laplace approximations has recently been proposed as an alternative. This approach promises very precise results in short computational time. The aim of the paper is to show how integrated nested Laplace approximations can be used as an inferential tool for a variety of spatiotemporal models for the analysis of reported cases of bovine viral diarrhoea in cattle from Switzerland. Conclusions concerning the problem of under-reporting in the data are drawn via a multilevel modelling strategy. Furthermore, a comparison with Markov chain Monte Carlo methods with regard to the accuracy of the parameter estimates and the usability of both approaches in practice is conducted. Approaches to model choice using integrated nested Laplace approximations are also presented.

Multiscale modeling of ductile failure in metallic alloys

Abstract Micromechanical models for ductile failure have been developed in the 1970s and 1980s essentially to address cracking in structural applications and complement the fracture mechanics approach. Later, this approach has become attractive for physical metallurgists interested by the prediction of failure during forming operations and as a guide for the design of more ductile and/or high-toughness microstructures. Nowadays, a realistic treatment of damage evolution in complex metallic microstructures is becoming feasible when sufficiently sophisticated constitutive laws are used within the context of a multilevel modelling strategy. The current understanding and the state of the art models for the nucleation, growth and coalescence of voids are reviewed with a focus on the underlying physics. Considerations are made about the introduction of the different length scales associated with the microstructure and damage process. Two applications of the methodology are then described to illustrate the potential of the current models. The first application concerns the competition between intergranular and transgranular ductile fracture in aluminum alloys involving soft precipitate free zones along the grain boundaries. The second application concerns the modeling of ductile failure in friction stir welded joints, a problem which also involves soft and hard zones, albeit at a larger scale.

Do Institutions Really Influence Political Participation?

This paper examines the influence of institutions and other contextual variables in a set of individual-level models of political participation, using a multi-level modelling strategy. It uses data from Citizenship Survey of the International Social Survey Programme conducted in 2004, to model relationships in many of the world's democracies. It examines the effects of variables that have been shown to be important in aggregate-level models of turnout, such as the effective number of parties, the distortions in representation associated with the electoral system, and the size of districts. It compares the institutional measures with other contextual variables that arise from rival models of individual-level political participation. The institutional variables have a modest impact on individual-level turnout, but their impact is much less important in relation to other types of participation. For the latter, non-institutional contextual variables arising from models of political participation appear to be more important.

How Does Geography Matter in the Ethnic Labor Market Segmentation Process? A Case Study of Chinese Immigrants in the San Francisco CMSA

With the continuing influx of large numbers of immigrants to the United States, urban labor market segmentation along the lines of race, ethnicity, gender, and class has drawn considerable attention. Using a confidential data set extracted from the United States Decennial Long Form Data 2000 and a multilevel regression modeling strategy, this article presents a case study of Chinese immigrants in the San Francisco metropolitan area. Correspondent with the highly segregated nature of the labor market between Chinese immigrant men and women, different socioeconomic characteristics at the census tract level are significantly related to their occupational segregation. The results of this study suggest that the social process of labor market segmentation is contingent on the immigrant geography of residence and workplace. Whereas the direction and magnitude of the spatial contingency are different between men and women, residency in Chinese immigrant-concentrated areas is perpetuating gender occupational segregation by skill level. Abundant ethnic resources might exist in ethnic neighborhoods and enclaves for certain types of employment opportunities; however, these resources do not necessarily help Chinese immigrant workers, especially women, to move upward in the labor market hierarchy.

Advanced permanent magnet motor drive modeling for automotive application under matlab/simulink environment

This paper proposes an advanced permanent magnet (PM) motor drive modeling for dynamic analysis of automotive control systems with PM motor drives. To enhance the reusability of the proposed PM motor drive model, each component is modeled in an object-oriented manner according to the physical partitioning of a real drive system, and userfriendly and well-organized interfaces are also proposed. These characteristics enable this model to be used as a kind of a programming library. A multi-level modeling strategy is also proposed, which helps make a compromise between model accuracy and simulation speed without structural modification of the system model. The model and modeling strategy developed allow for a comprehensive analysis of each component and dynamic analysis of the total system. System simulation results show the practical effectiveness of the proposed modeling strategy.

U.S. ATTORNEYS AND SUBSTANTIAL ASSISTANCE DEPARTURES: TESTING FOR INTERPROSECUTOR DISPARITY*

An important and highly discretionary component of the federal sentencing guidelines is the downward departure for providing substantial assistance. Critics charge that the substantial assistance departure, which requires a motion by the prosecutor, may produce the type of unwarranted sentencing disparity that the guidelines were intended to eliminate. Research reveals, for example, that jurisdictional variations are evident in the use of substance assistance departures (Johnson, Ulmer, and Kramer, 2008; Nagel and Schulhofer, 1992), and that the likelihood of receiving the departure is affected by legally irrelevant offender characteristics, which include race, ethnicity, and gender (Mustard, 2001). The purpose of this article is to extend this research by exploring the degree to which decisions regarding substantial assistance departures vary across prosecutors. Using data on offenders sentenced in three U.S. district courts and a multilevel modeling strategy, we investigate whether interprosecutor disparity exists in the likelihood of substantial assistance departures and in the criteria that prosecutors use in deciding whether to file a motion for a substantial assistance departure. Findings indicate that significant interprosecutor variation remains after taking into account offender characteristics, case characteristics, and the district in which the case is adjudicated.

Initiative Campaigns

Previous research has found that the campaigns of candidates running for office provide information to voters and can increase turnout. Scholarly research has also found that states with initiatives and referendums appearing on the ballot have higher voter turnout, especially in midterm elections. However, actual initiative campaigns are rarely measured. Drawing on national survey data and state contextual factors, we use a multilevel modeling strategy to test whether Americans are more likely to vote in recent midterm and presidential elections when there is increased spending in initiative and candidate campaigns, as well as more frequent use of direct democracy. The research includes a number of methodological advancements from earlier work on turnout and direct democracy (including a control for endogeneity) by restricting the analysis to initiative states only. The analysis suggests initiative campaigns not only increase individual level turnout but also especially benefit the lower educated.

Race/Ethnicity, Gender and Job Earnings across Metropolitan Areas in the United States: A Multilevel Analysis

Using the US 5 per cent Public Use Microdata Samples (PUMS) and Census Summary Files from the year 2000, and a multilevel linear modelling strategy, this study examines the effects of a range of individual and metropolitan-area characteristics on the job earnings of US-born Whites and Blacks, foreign-born Hispanics and Asians, separately by men and women. The key findings suggest that macroeconomic structure and racial composition have significant influences on interracial and gender earnings disparities, although the direction and magnitude are different among groups. Results also show that all racial and gender groups would benefit from a more evenly distributed labour market among ethnic minorities and immigrants. With respect to gender issues, the earning gaps between racial groups are greater than between men and women. However, compared with men, women workers are more sensitive to macroeconomic conditions, racial residential segregation and labour market segmentation.

How Does Geography Matter in Ethnic Labor Market Segmentation Process? A Case Study of Chinese Immigrants in the San Francisco CMSA

In the context of continuing influxes of large numbers of immigrants to the United States,urban labor market segmentation along the lines of race/ethnicity, gender, and class has drawn considerable growing attention. Using a confidential dataset extracted from the United States Decennial Long Form Data 2000 and a multilevel regression modeling strategy, this paper presents a case study of Chinese immigrants in the San Francisco metropolitan area. Correspondent with the highly segregated nature of the labor market as between Chinese immigrant men and women, different socioeconomic characteristics at the census tract level are significantly related to their occupational segregation. This suggests the social process of labor market segmentation is contingent on the immigrant geography of residence and workplace. With different direction and magnitude of the spatial contingency between men and women in the labor market, residency in Chinese immigrant concentrated areas is perpetuating the gender occupational segregation by skill level. Where as abundant ethnic resources may exist in ethnic neighborhoods and enclaves for certain types of employment opportunities, these resources do not necessarily help Chinese immigrant workers, especially women, to move upward along the labor market hierarchy.

Contextual Influences on Reproductive Health Service Use in Uttar Pradesh, India

This study examines the determinants of the use of four types of reproductive health-care services in Uttar Pradesh, India: contraceptive services, antenatal care, delivery in a medical institution, and services dealing with reproductive tract and sexually transmitted infections. The analysis uses a multilevel modeling strategy to assess the presence of household- and community-level variation in service use. The influence of community-level characteristics and reproductive health-care service attributes on service use is examined. The results highlight strong community-level influences on service use, although the type of community effect varies by service type. The role of some individual and household factors in determining a person's use of services is mediated by the characteristics of the community in which the individual lives. The results demonstrate the need to look beyond individual factors when examining health-care-seeking behavior, and illustrate that there is no singular "community" effect on service use.

A Multilevel Study of Elementary Male Black Students and White Students

Abstract A multilevel modeling strategy was used in this study to identify school characteristics linked to student achievement and the relationship between student achievement and student background, educational expectations, and classroom participation. The school characteristics considered included indicators of school organization, context, and climate. The subjects were 2,356 Black male students and White male students in 76 elementary schools. The results indicated that although classroom participation was a significant predictor of achievement for both groups, the strength of the association was affected by school characteristics for the Black males only. In particular, the ability of Black males to translate academic participation into achievement was weakened in large schools and in schools in which most students were from low socioeconomic backgrounds. It was concluded that in those settings, Black male students may be at a disadvantage.