The minority stress theory suggests LGBs experience greater stress levels due to their sexual minority identities; thus, they are more prone to psychological distress. Poor mental health is linked to internalized homophobia and heterosexism. However, affirmative social support may mitigate the stress response via the buffering hypothesis. My model posits that LGBs are more likely to report perceived stress; however, affirmative social support can mitigate stress. I investigated the relationship between perceived stress and sexual minority identity. I explored the relationship between heterosexism, emotional support and perceived stress and the moderating role of social support in my LGB sample. I conducted a hierarchical linear regression to test my model, which accounted for 29% of the variance in perceived stress. Heterosexism and emotional support were significantly associated with perceived stress. I failed to find a moderating role of emotional support. Limitations, strengths, future research and implications are discussed.

The Support Vector Machine (SVM) classification method has recently gained popularity due to the ease of implementing non-linear separating surfaces. SVM is an optimization problem with the two competing goals, minimizing misclassification on training data and maximizing a margin defined by the normal vector of a learned separating surface. We develop and implement new SVM models based on previously conceived SVM with L_1-Norm regularization with ramp loss error terms. The goal being a new SVM model that is both robust to outliers due to ramp loss, while also easy to implement in open source and off the shelf mathematical programming solvers and relatively efficient in finding solutions due to the mixed linear-integer form of the model. To show the effectiveness of the models we compare results of ramp loss SVM with L_1-Norm and L_2-Norm regularization on human organ microbial data and simulated data sets with outliers.

Abstract Carbon nanotubes (CNTs) were grown catalytically by a chemical vapor deposition method and characterized by a range of techniques. Fe, Ru and Co catalysts supported on the carbon nanotubes were prepared and investigated for their performances in the Fischer-Tropsch synthesis. CNTs were synthesized in a quartz tubular reactor at atmospheric pressure and at temperatures of 700°C over iron supported on CaCO3 using C2H2 as carbon source. Prior to CNT synthesis, the iron catalyst was first reduced under the same conditions (700°C and atmospheric pressure) in a flow of 5% H2 balanced in Argon. The catalyst, for the preparation of the CNTs, was prepared by the incipient wetness impregnation. The purification of the CNTs was performed with 30 wt % HNO3. Characterization of the CNTs using TEM, SEM, HRTEM, BET and TPR revealed that the crude product contained solely CNTs, catalysts particles and support, while no amorphous carbon was observed. The purified product is comprised of an interwoven matrix of tubes that were shown to be multi-walled (MWCNTs). CNT supported FT based catalysts were also prepared by an incipient wetness impregnation method and tested in a plug flow reactor in Fischer-Tropsch synthesis. The TEM images of the different FT catalysts supported on CNTs revealed that the catalyst particles are well dispersed on the surface of the CNTs. The catalyst particles were very iii small, and some residual Fe catalyst material, not removed by the acid treatment, could clearly be seen on the surface of the CNTs. The reduction and metal dispersion properties of the catalysts were investigated through TPR and chemisorption techniques. A TPR study showed three reduction steps for Co catalysts, and addition of Ru to the catalyst decreased the reduction temperature of the catalysts. Gasification of the CNTs was noted to occur at temperatures higher than 600°C. The effect of metal catalyst loading and promoters on the activity and selectivity of CNT supported FT synthesis catalysts was studied under condition of 275°C, 8 bar, CO/H2 = 1/2 and different flow rates. The FT catalysts supported on carbon nanotubes displayed a high CO conversion and excellent stability with time on stream in the Fischer-Tropsch synthesis. Fe catalysts displayed the lowest methane selectivity compared to all other FT synthesis catalysts used in this study.

MSc (Med), Community Paediatrics, Faculty of Health Sciences, University of the Witwatersrand / Background: Although breastfeeding is a key child survival strategy, breastfeeding practices in South African health institutions can generally be categorised as being poor. The global Baby-Friendly Hospital Initiative (BFHI), led by UNICEF and the WHO, aims to create a health care environment that promotes breastfeeding as the norm. This study aimed to document practices around breastfeeding support and compliance with the BFHI’s “10 steps to successful breastfeeding” in both baby-friendly accredited and non-accredited facilities. Methods: This was a cross-sectional study, conducted in nine facilities in Gauteng, including tertiary, secondary and districts hospitals, and midwife obstetric units. Convenience sampling was used. Study questionnaires were based on the generic BFHI assessment tool, but were modified to include more detailed investigation of HIV related factors. The questionnaire was verbally administered on-site to 165 mothers of well infants, and 65 nursing staff. Results: Suboptimal practices were identified in both baby-friendly and non-baby-friendly accredited facilities, but more so in the latter. None of the facilities passed all of the 10 BFHI steps. None of the baby-friendly certified institutions achieved a score sufficient to be still deemed baby-friendly. Steps that required advice and support from nursing staff, viz. step 5 (showing mothers how to breastfeed), step 8 (encouraging breastfeeding on demand), and step 10 (breastfeeding support after discharge from the facility), were particularly poorly done in the majority of facilities. Baby-friendly certified institutions were significantly better than non-accredited facilities for steps 2 (training), 4 (initiating breastfeeding within 1 hour), and 7 (rooming-in). Conclusion: Although baby-friendly accredited facilities generally performed better than non-accredited facilities, their performance failed to justify maintenance of their accredited status based on this assessment. However, some positive practices were sustained over time. There is a dire need for greater attention to be directed to the promotion of good breastfeeding practices by health professionals and institutions.

Thesis (D.B.A.)--Boston University / PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / Quality management has become a key competitive factor (Garvin, 1988). Product design quality requires incorporating the wants and needs of the customer base into new product development, and product conformance quality relies upon identifying and resolving flaws found in the products. Collecting this feedback requires that the boundary between the customer and the firm be spanned, and applying the feedback requires that it be communicated to the functional group that can effect product improvement. Through its constant interaction with customers, the customer support group is well positioned to sense this customer feedback and communicate it to the product development group. This dissertation examined the strategic role of the customer support organization within software companies as a quality assurance agent. While traditionally viewed as a rectifier of external quality failures, the research model proposed that customer support could also serve appraisal and preventive quality assurance roles through its boundary spanning activities. A small sample, comparative case study tested hypotheses about customer support's boundary spanning and provided for grounded theory building. The research revealed three primary findings. First, service delivery designs that included higher levels of customer contact allowed the service agents to sense a broader range of customer issues. Second, product quality increased when the customer data were collected in a detailed format structured to meet the practices of the development group. Third, product quality also increased when the customer support group had more involvement in product management decisions throughout the product development cycle. Through grounded theory-building, the research identified contextual factors that fostered a preventative quality assurance role for customer support. / 2031-01-01

Fung Chi Tim. / Thesis (M.Sc.)--Chinese University of Hong Kong, 1993. / Includes bibliographical references (leaves 53). / Abstract / Chapter I. --- Introduction --- p.1 / Chapter II. --- Objective --- p.4 / Chapter III. --- Methodology / Decision Methodology --- p.6 / Sensitivity Analysis and Re-iteration --- p.14 / Ease of use of the system --- p.22 / Chapter IV. --- Test Result --- p.26 / Chapter V. --- Future Research Direction --- p.28 / Chapter VI. --- Conclusion --- p.32 / Chapter VII. --- Appendix A Description of System --- p.35 / Chapter VIII. --- Appendix B User's Guide --- p.44 / Figures --- p.50 / References --- p.53

Kin-hong Wong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1992. / Includes bibliographical references (leaves 67-70). / Chapter Chapter 1 --- Introduction --- p.1 / Chapter Chapter 2 --- Literature Review --- p.4 / Chapter 2.1 --- Modelling --- p.4 / Chapter 2.1.1 --- Integer Programming Model --- p.4 / Chapter 2.1.2 --- Network Flow Model --- p.6 / Chapter 2.1.3 --- Graph Colouring Model --- p.8 / Chapter 2.2 --- Heuristic Approach --- p.10 / Chapter 2.2.1 --- Lesson Selection Strategy --- p.11 / Chapter 2.2.2 --- Time Slot Selection Strategy --- p.12 / Chapter 2.2.3 --- Conflict handling Strategy --- p.13 / Chapter 2.3 --- Remarks --- p.15 / Chapter Chapter 3 --- Project Overview --- p.16 / Chapter 3.1 --- The Work Flow --- p.16 / Chapter 3.2 --- System Objectives --- p.17 / Chapter 3.3 --- An Incremental Construction Approach --- p.18 / Chapter 3.4 --- A Decision Support System --- p.19 / Chapter Chapter 4 --- Task Analysis --- p.21 / Chapter 4.1 --- Three-Stage Task Model --- p.21 / Chapter 4.1.1 --- Conflict Free (CF) Stage --- p.22 / Chapter 4.1.2 --- Conflict Avoidance (CA) Stage --- p.23 / Chapter 4.1.3 --- Conflict Resolution (CR) Stage --- p.24 / Chapter 4.2 --- System Supports --- p.24 / Chapter 4.2.1 --- Supports for CF Stage --- p.24 / Chapter 4.2.2 --- Supports for CA Stage --- p.25 / Chapter 4.2.3 --- Supports for CR Stage --- p.25 / Chapter Chapter 5 --- System Design --- p.27 / Chapter 5.1 --- User Interface --- p.28 / Chapter 5.2 --- Data Model --- p.28 / Chapter 5.2.1 --- Timetable Structure --- p.30 / Chapter 5.2.2 --- Resource Lists --- p.31 / Chapter 5.2.3 --- Lesson List --- p.33 / Chapter 5.2.4 --- History Log --- p.34 / Chapter 5.3 --- Operational Tools --- p.35 / Chapter 5.3.1 --- Elementary Scheduling Functions --- p.35 / Chapter 5.3.2 --- AUTO Function --- p.36 / Chapter 5.3.3 --- FLAG/COMMIT/ABORT --- p.37 / Chapter 5.4 --- Analytical Tools --- p.39 / Chapter 5.4.1 --- Availability Table --- p.39 / Chapter 5.4.2 --- Operation History --- p.39 / Chapter 5.4.3 --- Slot Finder --- p.40 / Chapter 5.4.4 --- Scheduling Priority --- p.42 / Chapter 5.5 --- Remarks --- p.43 / Chapter Chapter 6 --- System Evaluation --- p.45 / Chapter 6.1 --- Case 1 --- p.46 / Chapter 6.2 --- Case 2 --- p.46 / Chapter 6.3 --- Case 3 --- p.47 / Chapter 6.4 --- Remarks --- p.48 / Chapter Chapter 7 --- Further Works --- p.50 / Chapter 7.1 --- Algorithmic Tools --- p.50 / Chapter 7.2 --- Post-Scheduling Room Assignment --- p.52 / Chapter 7.3 --- Input and Output Processors --- p.53 / Chapter 7.4 --- Electronic Tutorial Subsystem --- p.54 / Chapter 7.5 --- Task-Model-Based Help Subsystem --- p.54 / Chapter Chapter 8 --- Conclusions --- p.56 / Chapter Appendix A --- Database Definition --- p.58 / Chapter Appendix B --- Conflict Resolution Algorithm --- p.61 / Chapter Appendix C --- System Specification --- p.63 / References --- p.67

Objectives: The current study aimed to profile the personality styles, expectations and psychological adaptation of Clinical Psychology Trainees. It also aimed to look at the relationship between these variables. Design: A cross-sectional postal questionnaire study, employing between group and correlational analyses. Methods: A sample of 364 psychologists in clinical training (57% response rate) from 15 UK clinical psychology training courses participated in the study. They completed questionnaires of personality, psychological adaptation, social support and an expectations measure specifically designed for the study. Results: The majority of psychologists in clinical training who participated in the study were well adjusted in terms of personality, did not experience extensive problems with psychological adaptation, and had the majority of their expectations met. A significant sub group reported personality adjustment problems and problems with self esteem, anxiety, depression and work adjustment. Low self esteem was present in just under a quarter of the sample. Personality adjustment was found to be related to expectations and psychological adaptation. Trainee psychologists with poorer personality adjustment were less likely to have their expectations met, especially with regard to the impact of training on their life, and were more likely to suffer from poor psychological adaptation, particularly in terms of low self esteem, anxiety, depression and work adjustment problems. Self esteem was related to discrepancies in actual and ideal social support. Some differences were found between year groups. Gender and age were not related to personality adjustment, psychological adaptation or expectations. Conclusions and Implications: The findings were discussed in terms of the interpretation of personality style. Implications for clinical psychology training and the profession of clinical psychology were considered.

Bullying among school children is a social phenomenon that is now recognised as a widespread and serious problem across the globe. While decades of research have generated valuable insights as regards prevalence, main correlates and detrimental health consequences, many questions and gaps remain. For instance, it is unclear why the great majority of peer bystanders not intervene to support victims in a bullying conflict despite holding anti-bullying beliefs. Furthermore, great efforts have been made in the area of peer support and anti-bullying initiatives however there is still no intervention that has shown to be effective long term and cross-culturally. This thesis consists of two empirical studies. To advance knowledge of factors that influence pupils' victim support behaviour, the first study examined the role of perceived friend and peer consequences in predicting intentions to three types of help: provide emotional support, help to stop the bully and get adult support. Structural equation modelling revealed that perceived friend consequences were significantly associated with each of the victim support behaviours studied. Additionally, perceived peer reactions predicted intentions to get adult help. These findings suggest that friends play a more important role than peers in affecting victim support. Some significant gender effects emerged, showing that the overall pattern of associations held for boys, but not for girls. The findings highlight the concerns children hold with regard to their (dis)approving views related to victim support. Outcomes further suggest that victim defending should not be regarded as a broad homogeneous construct. The second study assessed the effectiveness of a cross-age teaching of social issues intervention (CATS) on enhancing pupils' knowledge on three victim support behaviours, and their awareness of the value of helping. In small cooperative groups older pupils were invited to step into the tutor role to prepare a lesson and teach it to two years younger tutees. An experimental-control group design was employed to test participants' performance at three time points over a six to eight week period. CATS tutors significantly improved their knowledge and awareness of the provictim behaviours studied while no positive changes were evident for participants in the control group. Furthermore, children who participated in the project expressed high satisfaction with the intervention. Based on the positive findings it was concluded that CATS is a viable technique for enhancing pupils' knowledge and awareness on prosocial topics. Helping children to see the value of supporting victims of bullying, in any of the ways studied, could help them avoid anticipating negative reactions from friends and peers, and in turn make it more likely that they would choose to help if the need arose.

Heart failure (HF) is a common cause of hospitalisation and mortality across industrialised countries. The number of hospitalisations and deaths attributed to heart failure is increasing, and this trend is predicted to continue. Numerical and in-vitro simulations of the human cardiovascular system constitute the basic tools for enhancing diagnostic and therapeutic technologies for HF and this would in turn, have significant effects on morbidity,mortality, and healthcare expenditure. Mechanical Circulatory Support (MCS) as a destination therapy for HF is rising significantly as it provides a cost-effective alternative to long-term treatment and cardiac transplantation. However, long-term versatility is far from ideal and incidence of transient and permanent neurological events is still high. To this end, evolution of MCS devices calls for more sophisticated design and evaluation methods. The purpose of this work is to develop a numerical model and to implemented a novel in-vitro model of the cardiovascular system with the intention of evaluating the performance characteristics of a purposely selected centrifugal pump impeller for the treatment of both Class III and IV HF conditions when placed in series with the heart at two different anatomic locations: Ascending Aorta and Descending Aorta. An existing lumped-parameter model of the CV system, that included models for the heart, the pulmonary and the systemic circulatory loops by adapting a modified version of the fourth-element Windkessel model was enhanced by dividing the systemic circulation into six parallel vascular beds, and by including an autoregulatory system to control both pressures and volumes throughout the system. As part of the novelty of the present work, a volume reflex loop was included with the purpose of simulating volume overload conditions, as commonly found in HF conditions, and obtaining a more realistic analysis of volume displacement, while using a MCS device. The in-vitro model implemented in this work adopted most of the features included in the mathematical counterpart with the purpose of validating the numerical results. As a result of the combination of models and proper optimisation of the system parameters, predictions of pathophysiological trends and MCS usage are satisfactorily obtained. The models implemented in this work offer a valuable tool for the selection and performance evaluation of MCS devices for the treatment of HF conditions.