Homelessness is an important social problem in Canada, and the needs and experiences of women may differ from those of other homeless people. Little research has looked beyond rehousing to examine community integration following homelessness. Predictive models of three distinct facets of community integration for women who have experienced homelessness are presented and tested in this thesis. The first model examines physical integration, which is defined in terms of attaining and retaining stable housing. The second model predicts economic integration, defined in terms of participation in work or education. The third model predicts psychological integration, defined as psychological sense of community in one’s neighbourhood. Data for this research comes from a two-year longitudinal study conducted in Ottawa. Participants were women aged 20 and over (N =101) who were homeless at the study’s outset. Family status was an important predictor of community integration. Women who were accompanied by dependent children were more likely than those unaccompanied by children to be physically, economically and psychologically integrated in their communities. Having access to subsidized housing predicted becoming rehoused and living in one’s current housing for longer. Greater perceived social support predicted living in one’s current housing for longer. Past work history and mental health functioning predicted economic integration. Lower levels of education predicted returns to full-time studies. Living in higher quality housing and having more positive contact with neighbours predicted psychological integration, while living in one’s current housing for longer predicted lower levels of psychological integration. Overall, participants achieved a moderate level of community integration. The majority had been housed for at least 90 days at follow-up. However, only a minority were participating in the workforce or education at follow-up. Participants achieved only a moderate level of psychological integration. Results are discussed in terms of implications for policy and service provision. Improvements in the availability and quality of affordable housing, as well as employment support are recommended. Special attention needs to be paid to providing adequate and effective services for women who are unaccompanied by dependent children. / Fonds québécois de la recherche sur la societé et la culture

Although HIV/AIDS is the 9th leading cause of death in African-American children, 80% of HIV-infected children in the U.S. live into school-age years. This study focuses on associations between HIV illness severity, family factors, and long-term cognitive functioning of these children. Participants included 42 perinatally HIV-infected children (mean age = 72.4 months), 93% of whom were African-American. Mean intellectual functioning was more than one standard deviation below the normative mean; whereas, overall language and attention functioning were generally not different from the normative sample. First, this study described changes in functioning over time and/or between genders. Analyses of variance were conducted for five outcome variables (i.e., full scale IQ, verbal IQ, performance IQ, expressive and receptive one word picture vocabulary test). Expressive language scores increased over time. For receptive language, males’ skills improved significantly over time, while the decline in females’ skills did not reach significance. Second, the associations between Time Two illness severity (i.e., viral load), and Time One familial variables (i.e., adult-to-child ratio in the home, number of caregivers lost to death, number of months since caregiver death), with outcome variables at Time Two (i.e., intellectual, language, and attentional/hyperactivity functioning) were examined. For intellectual and expressive language, only the respective Time One functioning independently contributed a significant amount to Time Two functioning. For receptive language, Time One receptive language and the adult-to-child ratio in the home significantly predicted Time Two functioning. As the number of adults per child increased, there was an improvement in receptive language functioning. For both measures of language, the interaction between Time Two illness severity and Time One months since caregiver death significantly predicted Time Two functioning. With no loss of caregiver, more ill children demonstrated better language abilities than less ill children. Across illness groups, children performed similarly after a recent caregiver death. With greater time since caregiver death, the less ill children performed better than their more ill peers. For attention/hyperactivity, no predictor variables were significant. Interventions that consider family factors, as well as medical information, as potential influences on future child functioning may aid in the battle against this chronic illness.

Longitudinal data arise commonly in many fields including public health studies and survey sampling. Valid inference methods for longitudinal data are of great importance in scientific researches. In longitudinal studies, data collection are often designed to follow all the interested information on individuals at scheduled times. The analysis in longitudinal studies usually focuses on how the data change over time and how they are associated with certain risk factors or covariates. Various statistical models and methods have been developed over the past few decades. However, these methods could become invalid when data possess additional features. First of all, incompleteness of data presents considerable complications to standard modeling and inference methods. Although we hope each individual completes all of the scheduled measurements without any absence, missing observations occur commonly in longitudinal studies. It has been documented that biased results could arise if such a feature is not properly accounted for in the analysis. There has been a large body of methods in the literature on handling missingness arising either from response components or covariate variables, but relatively little attention has been directed to addressing missingness in both response and covariate variables simultaneously. Important reasons for the sparsity of the research on this topic may be attributed to substantially increased complexity of modeling and computational difficulties. In Chapter 2 and Chapter 3 of the thesis, I develop methods to handle incomplete longitudinal data using the pairwise likelihood formulation. The proposed methods can handle longitudinal data with missing observations in both response and covariate variables. A unified framework is invoked to accommodate various types of missing data patterns. The performance of the proposed methods is carefully assessed under a variety of circumstances. In particular, issues on efficiency and robustness are investigated. Longitudinal survey data from the National Population Health Study are analyzed with the proposed methods. The other difficulty in longitudinal data is model selection. Incorporating a large number of irrelevant covariates to the model may result in computation, interpretation and prediction difficulties, thus selecting parsimonious models are typically desirable. In particular, the penalized likelihood method is commonly employed for this purpose. However, when we apply the penalized likelihood approach in longitudinal studies, it may involve high dimensional integrals which are computationally expensive. We propose an alternative method using the composite likelihood formulation. Formulation of composite likelihood requires only a partial structure of the correlated data such as marginal or pairwise distributions. This strategy shows modeling tractability and computational cheapness in model selection. Therefore, in Chapter 4 of this thesis, I propose a composite likelihood approach with penalized function to handle the model selection issue. In practice, we often face the model selection problem not only from choosing proper covariates for regression predictor, but also from the component of random effects. Furthermore, the specification of random effects distribution could be crucial to maintain the validity of statistical inference. Thus, the discussion on selecting both covariates and random effects as well as misspecification of random effects are also included in Chapter 4. Chapter 5 of this thesis mainly addresses the joint features of missingness and model selection. I propose a specific composite likelihood method to handle this issue. A typical advantage of the approach is that the inference procedure does not involve explicit missing process assumptions and nuisance parameters estimation.

Nafion has been the mostly used perfluorosulfonated proton exchange membrane (PEM) in fuel cell. Although a number of problems remain to be resolved on the application of Nafion as a PEM, a less expensive alternative PEM has not been found mainly because of its high proton conductivity. Therefore, much effort has been invested to modify it or find a better and inexpensive material. The exploration of the methods to counter degradation and aging of Nafion is also an important direction of research. In this work, the behavior of PEG in Nafion is investigated with solid state NMR spectroscopy. A series of samples with different PEG sizes and concentrations in Nafion was prepared and the variable temperature proton spectra and longitudinal relaxation times (T1) were measured on two spectrometers. Some interesting findings were made, e.g., the 1H chemical shift of water in concentrated PEG solution decrease while its T1 increase, the higher the concentration of PEG, the larger the increase (of water 1H chemical shift) or decrease (of water 1H T1). These findings provide valuable information on further improving the performance of Nafion in proton conductivity and durability.

The design of reinforced concrete structures for extreme events requires accurate predictions of the ultimate rotational capacity of critical sections, which is dictated by the failure mechanisms of shear, hoop fracture, low-cycle fatigue and longitudinal bar buckling. The purpose of this research is to develop a model for the full compressive behavior of longitudinal steel including the effects of bar buckling. A computational algorithm is developed whereby experimental data can be rigorously modeled. An analytical model is developed from rational mechanics for modeling the complete compressive stress-strain behavior of steel including local buckling effects. The global buckling phenomenon is then investigated in which trends are established using a rigorous computational analysis, and a limit analysis is used to derive simplified design and analysis equations. The derived buckling models are incorporated into wellestablished sectional analysis routines to predict full member behavior, and the application of these routines is demonstrated via an incremental dynamic analysis of a ten-storey reinforced concrete building. The buckling models and the sectional analysis routine compare favorably with experimental data. Design recommendations and topics for further research are presented.

This research, aimed at advancing the theory of environmental hazard adjustment processes by contrasting households from three cities in a high seismic hazard area with households from three other cities in a moderate seismic hazard area. It identified seven types of stakeholders namely, the risk area residents and their families (primary group), the news media, employers, and friends (secondary group), and federal, state, and local governments (tertiary group), and explained why they are relevant to the adoption of seismic hazard adjustments. It also addressed three key attributes— knowledge, trustworthiness, and responsibility for protection—ascribed to these multiple stakeholders and the relationships of these stakeholder attributes with risk perception, hazard intrusiveness, hazard experience, gender, resource adequacy, fatalism and hazard adjustment adoption. It was specifically concerned with the effects of nested interactions due to trust and power differentials among the seven stakeholders, with the self reported adoption of 16 earthquake protective measures at two points in time (1997 and 1999). Some of the key findings indicate that risk perception, gender, fatalism, city activity in earthquake management and demographic characteristics did not significantly predict hazard adjustment adoption. However, all stakeholder characteristics had significant positive correlations with risk perception and hazard adjustment, implying a peripheral route for social influence. Hazard intrusiveness, hazard experience, and stakeholder knowledge, trustworthiness, and responsibility affected the increased adoption of hazard adjustments by households. Particularly important are the peer groups’ (employers, friends and family) knowledge, trustworthiness and responsibility. These findings suggest, hazard managers cannot count only on the federal, state, and local government advisories put out through the news media to affect community decisions and thereby households’ decisions to take protective actions. Instead, hazard managers need to shift focus and work through peer group networks such as service organizations, industry groups, trade unions, neighborhood organizations, community emergency response teams, faith-based organizations, and educational institutions to increase the knowledge, trustworthiness and responsibility of all in the peer group. This will assure higher household hazard adjustment adoption levels, thus facilitating a reduction in post disaster losses and recovery time.

A robust literature documents numerous negative implications of neuroticism for romantic relationships. The current study was the first to demonstrate necessary information regarding how couples can protect against these implications. Given the role of negative affect in the association between neuroticism and relationship difficulties, and given the role of sex in reducing negative affect, the current 8-wave longitudinal study of 72 newlywed couples tested the prediction that sexual frequency would moderate the association between neuroticism and marital satisfaction. Lagged multilevel modeling analyses supported this prediction. Specifically, although neuroticism was negatively associated with changes in marital satisfaction among spouses engaging in less frequent sex over the prior 6 months, neuroticism was unrelated to changes in satisfaction among spouses reporting more frequent sex over the prior 6 months. These findings join others in highlighting the importance of considering the broader context of the relationship to developing a complete understanding of relationship development.

Thesis--University of Michigan. / eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (v. 2, leaves 476-496).

Thesis (Ph. D.)--University of Washington, 2002. / Vita. Includes bibliographical references (p. 133-141).

Not available / text