Projects and programmes are inherently complex; the interaction of people, systems, processes and data within a dynamic environment creates an intricate network of agents whose behaviour can be unpredictable and unexpected. The management of this complexity is ordinarily concerned with the implementation of tools and techniques to ensure that projects are completed within the desired cost and time, at the agreed level of performance and quality – this is often referred to as the †̃iron triangleâ€TM. However, the impact of a dynamic external environment on the †̃softâ€TM boundaries of the project domain can lead to extreme difficulty in attempting to forecast or predict outcomes and system behaviours. This thesis contends that there is a clear desideratum for a new paradigm in project management practice and research that moves beyond the traditionalist (reductionist) approach to one that embraces, rather than attempts to simplify complexity. The research described in this thesis seeks to uncover the characteristics of complexity, in the context of projects and programmes, in an attempt to uncover if complexity is a factor in the determination of †̃valuableâ€TM outcomes. Subsequently, and through the theoretical lens of complexity theory, this research seeks to highlight the importance of our understanding and treatment of complexity in the execution and management of projects and programmes. The research further seeks to demonstrate how complexity thinking may inform a more sophisticated understanding of how projects, programmes and portfolios delivered successfully (Ziadat, 2017). The context of the research is the oil and gas (O & G) engineering sector in the Middle East and North Africa (MENA) region. A two stage qualitative and quantitative methodology is applied, based on deductive reasoning. The first stage involves the development of a questionnaire and a series of unstructured interviews to gain an understanding of the practical consideration that emerges from the literature review. The second stage of the research involves the application of meta-analysis to study the correlation between the complexity factors identified in the first stage, aiming for heterogeneity, identification of patterns and directing to achieve robust conclusions by using sensitivity analysis. The thesis proposes a new model of complexity factors for oil & gas engineering projects in the MENA region. The model is designed to facilitate the analysis of the project complexity landscape and to define requirements for oil & gas organisations involved with the delivery of projects and programmes to cope with different complexity factors within and across the MENA region. The outcomes include substantial relationship between technical and health, safety & environment complexity factors and project performance despite the mediation of project management complexity factors, yet the organizational complexity factors can be observed at a significant level when project management in complexity factors are considered as a mediator in the model (Ziadat, 2016).

Orientador: Antonio José Maria Cataneo / Resumo: Introdução: A paralisia de pregas vocais caracteriza-se como uma das principais complicações e sequelas após a tireoidectomia. A técnica convencional compreende a identificação visual do trajeto do nervo para sua preservação e a neuromonitorização do nervo laríngeo recorrente (NLR) durante a cirurgia de tireóide visa a redução do risco de paralisias por injúria ao NLR, porém sua utilidade ainda é discutível. Objetivos: Determinar através de revisão sistemática se o uso da neuromonitorização do NLR durante tireoidectomias realmente apresenta benefícios comparado á técnica convencional de identificação somente do nervo durante a cirurgia, especialmente em termos de redução do risco de paralisia de prega vocal, seja transitória ou permanente. Materiais e Métodos: A pesquisa nas principais bases de dados, MEDLINE via PUBMED, LILACS, SCOPUS, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL); clinicaltrials.gov e fontes adicionais foi realizada em Maio de 2018. Uma revisão sistemática e meta-análise foram feitas usando o modelo de efeitos fixos. O GRADE foi usado para classificar a qualidade das evidências. Os desfechos primários avaliados foram o total de paralisias do NLR por nervos em risco, paralisias transitórias e permanentes e o tempo de cirurgia comparando-se as duas técnicas. Como desfechos secundário avaliou-se o tempo de cirurgia. Foi realizada análise de subgrupos de pacientes de alto e baixo risco de injuria do NLR. Também avaliamos como análise de subgr... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: Vocal cord paralysis is characterized as one of the main complications and sequels after thyroidectomy. The conventional technique comprises the visual identification of the nerve path for its preservation and recurrent laryngeal nerve (RLN) neuromonitoring during thyroid surgery aims to reduce the risk of paralysis due to injury to the RLN, but its usefulness is still debatable. Objectives: To determine through systematic review if the use of neuromonitoring of the RLN during thyroidectomies actually presents benefits compared to the conventional technique of only identification of the nerve during surgery, especially in terms of reducing the risk of vocal fold paralysis, whether transient or permanent. Materials and Methods: The search in the main databases, MEDLINE, LILACS, SCOPUS, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL); clinicaltrials.gov and additional sources was performed in May 2018. A systematic review and meta-analysis was done using the fixed-effects model. GRADE was used to classify the quality of the evidence. The primary endpoints evaluated were the total nerve paralysis for nerves at risk, transient and permanent paralyzes, and surgery time comparing the two techniques. As secondary outcomes, the time of surgery was evaluated. We also evaluated as subgroup analysis the transient and permanent paralysis in a group of patients with high and low risk of RLN injury. Results: Four randomized clinical trials involving 1.379 pat... (Complete abstract click electronic access below) / Mestre

Submitted by TAMIRES FERREIRA SIQUEIRA (tamy_170@yahoo.com.br) on 2018-08-03T16:10:50Z No. of bitstreams: 1 Tese Mestrado Tamires.1.pdf: 1168639 bytes, checksum: b2029c52ab3322c660f0abd08f81dd9e (MD5) / Approved for entry into archive by Sulamita Selma C Colnago null (sulamita@btu.unesp.br) on 2018-08-06T13:19:27Z (GMT) No. of bitstreams: 1 siqueira_tm_me_bot.pdf: 1168639 bytes, checksum: b2029c52ab3322c660f0abd08f81dd9e (MD5) / Made available in DSpace on 2018-08-06T13:19:27Z (GMT). No. of bitstreams: 1 siqueira_tm_me_bot.pdf: 1168639 bytes, checksum: b2029c52ab3322c660f0abd08f81dd9e (MD5) Previous issue date: 2018-07-12 / Introdução: A paralisia de pregas vocais caracteriza-se como uma das principais complicações e sequelas após a tireoidectomia. A técnica convencional compreende a identificação visual do trajeto do nervo para sua preservação e a neuromonitorização do nervo laríngeo recorrente (NLR) durante a cirurgia de tireóide visa a redução do risco de paralisias por injúria ao NLR, porém sua utilidade ainda é discutível. Objetivos: Determinar através de revisão sistemática se o uso da neuromonitorização do NLR durante tireoidectomias realmente apresenta benefícios comparado á técnica convencional de identificação somente do nervo durante a cirurgia, especialmente em termos de redução do risco de paralisia de prega vocal, seja transitória ou permanente. Materiais e Métodos: A pesquisa nas principais bases de dados, MEDLINE via PUBMED, LILACS, SCOPUS, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL); clinicaltrials.gov e fontes adicionais foi realizada em Maio de 2018. Uma revisão sistemática e meta-análise foram feitas usando o modelo de efeitos fixos. O GRADE foi usado para classificar a qualidade das evidências. Os desfechos primários avaliados foram o total de paralisias do NLR por nervos em risco, paralisias transitórias e permanentes e o tempo de cirurgia comparando-se as duas técnicas. Como desfechos secundário avaliou-se o tempo de cirurgia. Foi realizada análise de subgrupos de pacientes de alto e baixo risco de injuria do NLR. Também avaliamos como análise de subgrupos as paralisias transitórias e permanentes em grupo de pacientes de alto e baixo risco de injúria do NLR. Resultados: Quatro ensaios clínicos randomizados envolvendo 1.379 pacientes submetidos à tireoidectomia utilizando a neuromonitorização ou com a identificação somente do NLR, sendo 2.605 o número total de nervos em risco foram avaliados. A neuromonitorização diminuiu o risco do total paralisias (transitórias e permanentes) de prega vocal em tireoidectomias (RR 0,64; IC 95% de 0,43 a 0,95, I 2 = 30%), assim como diminuiu o risco de paralisias transitórias (RR 0,60; IC 95% de 0,38 a 0,95, I 2 =10%), porém não diminuiu o risco de paralisias permanentes (RR 0,84; IC 95% de 0,36 a 1,93, I 2 = 0%). A análise da média do tempo de cirurgia foi conduzida em dois estudos, entretanto como a heterogeneidade entre os estudos foi maior que 75% a metanálise não foi apropriada para estabelecer a diferença do tempo médio de cirurgia entre os grupos. Quanto a análise de subgrupos, observou-se que a neuromonitorização não diminuiu o risco de paralisias transitórias (RR 0,61; IC 95% de 0,28 a 1,31, I 2 = 0%) e nem o risco de paralisias permanentes de prega vocal em tireoidectomias de baixo risco (RR 1,03; IC 95% de 0,21 a 5,07, I 2 não aplicável). Assim como também não diminuiu o risco de paralisias transitórias (RR 0,75; IC 95% de 0,18 a 3,07, I 2 = 71%) e de paralisias permanentes de prega vocal em tireoidectomias de alto risco (RR 0,69; IC 95% de 0,27 a 1,79, I 2 = 4%). Conclusões: Essa meta-análise utilizando apenas ensaios clínicos randomizados demonstrou que a neuromonitorização do nervo laríngeo recorrente em tireoidectomias possivelmente reduz ligeiramente o risco de paralisia transitória das pregas vocais, mas provavelmente faz pouca ou nenhuma diferença no risco de paralisia permanente. / Introduction: Vocal cord paralysis is characterized as one of the main complications and sequels after thyroidectomy. The conventional technique comprises the visual identification of the nerve path for its preservation and recurrent laryngeal nerve (RLN) neuromonitoring during thyroid surgery aims to reduce the risk of paralysis due to injury to the RLN, but its usefulness is still debatable. Objectives: To determine through systematic review if the use of neuromonitoring of the RLN during thyroidectomies actually presents benefits compared to the conventional technique of only identification of the nerve during surgery, especially in terms of reducing the risk of vocal fold paralysis, whether transient or permanent. Materials and Methods: The search in the main databases, MEDLINE, LILACS, SCOPUS, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL); clinicaltrials.gov and additional sources was performed in May 2018. A systematic review and meta-analysis was done using the fixed-effects model. GRADE was used to classify the quality of the evidence. The primary endpoints evaluated were the total nerve paralysis for nerves at risk, transient and permanent paralyzes, and surgery time comparing the two techniques. As secondary outcomes, the time of surgery was evaluated. We also evaluated as subgroup analysis the transient and permanent paralysis in a group of patients with high and low risk of RLN injury. Results: Four randomized clinical trials involving 1.379 patients undergoing thyroidectomy using neuromonitoring or with identification only of the RLN, of which 2.605 was the total number of nerves at risk evaluated. Neuromonitoring reduced the risk of total (transient and permanent) vocal fold paralysis after thyroidectomies (RR 0.64, 95% CI 0.93 to 0.95, I 2 = 30%), as well as decreased the risk of transient paralysis (RR 0.60, 95% CI 0.38 to 0.95, I 2 = 10%), but did not decrease the risk of permanent paralysis (RR 0.84, 95% CI 0.36 to 1.93, I2 = 0%). The analysis of the mean time of surgery was conducted in two studies, however, the heterogeneity between the studies was greater than 75% and the meta-analysis was not appropriate to establish the difference in the mean time of surgery between the groups. As the analysis of subgroups, it was observed that neuromonitoring did not reduce the risk of transient paralysis (RR 0.61, 95% CI 0.28 to 1.31, I 2 = 0%) and neither the risk of permanent paralysis in low-risk thyroidectomies (RR 1.03, 95% CI 0.21 to 5.07, I2 not applicable). As it also did not reduce the risk of transient paralysis (RR 0.75, 95% CI 0.18 to 3.07, I2 = 71%) and permanent vocal fold paralysis in high-risk thyroidectomies (RR 0, 69, 95% CI 0.27 to 1.79, I 2 = 4%). Conclusions: This meta-analysis using only randomized clinical trials has demonstrated that recurrent laryngeal nerve neuromonitoring in thyroidectomies possibly slightly reduces the risk of transient vocal fold paralysis but probably makes little or no difference in the risk of permanent paralysis.

The aim of this paper is to analyze the variation in empirical estimates of the income and price elasticities of energy demand. The evidence presented, through a meta-analysis, allows identification of the weighted average of the income and price elasticities, shows that the estimates are very heterogeneous, that there is publication bias, and that factors such as region, energy sector, among others, affect its volatility. The evidence also indicates that income elasticity in Latin America is greater than in the OECD countries, and that the price elasticity of energy demand is lower in Latin America than in the OECD countries. Therefore, continued economic growth in Latin America will be accompanied by a growth in energy demand. Moreover, the establishment of a tax in Latin America, under the current elasticities, is less effective and will be insufficient to control the increase in energy consumption. / El objetivo de este artículo es analizar la variación de las elasticidades ingreso y precio de la demanda de energía. La evidencia presentada, con un metaanálisis, permite identificar la media ponderada de estas elasticidades ingreso y precio, muestra que las estimaciones son muy heterogé- neas, que existe sesgo de publicación y que algunos factores como la región, el sector del consumo de energía, entre otros, inciden en su volatilidad. La evidencia también indica que la elasticidad ingreso en América Latina es mayor que aquella de los países de la OCDE y, simultáneamente, que la elasticidad precio de la demanda de energía es menor en América Latina que en los países de la OCDE. Así, un crecimiento económico continuo en América Latina vendrá acompañado de un crecimiento de la demanda de energía y que el establecimiento de un impuesto en América Latina, bajo las actuales elasticidades, es menos efectivo y en general sería insuficiente para controlar el aumento del consumo de energía.

Dans les essais cliniques randomisés (ECR) en cancérologie, un critère de substitution est une mesure biologique utilisée à la place d’un critère cliniquement pertinent pour le patient, par exemple la survie globale (SG), qui doit permettre de prédire l’effet attendu du traitement. Des critères alternatifs à la SG, par exemple la survie sans progression, sont de plus en plus fréquemment utilisés en tant que critère de jugement principal dans les ECR. En pratique cependant, les capacités de substitution à la SG de ces critères ne sont pas systématiquement évaluées. Nous avons dressé un état des lieux des critères de substitution validés en cancérologie à partir d’une revue systématique de la littérature. Par la suite, nous avons évalué par une approche méta-analytique des critères de substitution dans le contexte des sarcomes des tissus mous en situation avancée et du cancer du sein en situation adjuvante. Les résultats n’ont pas permis de définitivement valider de critères de substitution à la SG dans ces indications. La SG doit donc rester le critère de jugement principal des ECR, même si certains critères alternatifs restent informatifs dans des évaluations plus précoces (phase II, analyse de futilité), sous réserve que les données de survie continuent à être recueillies. Ce travail fournit des informations clés pour le développement des ECR en cancérologie afin notamment de sélectionner au mieux les critères de jugement de l’efficacité thérapeutique. / In cancer randomized controlled trials (RCT), a surrogate endpoint is intended to substitute a clinically relevant endpoint, e.g. overall survival (OS), and it is supposed to predict treatment effect. Alternative endpoints, for example progression-free survival, are increasingly being used in place of OS as primary efficacy endpoints in RCTs. In practice however, the surrogate properties of these endpoints are not systematically assessed. We performed a systematic literature review to identify surrogate endpoints validated in oncology. We next conducted MAs to evaluate surrogate endpoints in two cancer settings: advanced soft-tissue sarcoma and adjuvant breast cancer. Results could not definitely validate surrogate endpoints in these indications. OS must remain the primary efficacy endpoint in these settings, even though alternative endpoints may provide valuable input in earlier phase studies (phase II trials, futility analyses). This work provides key information for the design of cancer RCTs, in particular for the choice of primary endpoints to assess treatment efficacy.

A prevalência da obesidade está aumentando em um ritmo alarmante, por isso estratégias de prevenção e controle da obesidade são importantes. Com isto em mente, foi realizada uma revisão sistemática com metanálise para avaliar o efeito dos programas de intervenções que incentivassem a prática de atividade física e a alimentação saudável, identificando os mais apropriados para a prevenção e controle da obesidade na população escolar. Foram incluídos estudos controlados randomizados disponíveis nas seguintes bases de dados eletrônicas: Pubmed, Lilacs, Embase, Scopus, Web of Science e Cochrane Library. A qualidade dos estudos foi avaliada levando em conta os critérios de sigilo de alocação propostos pela Colaboração Cochrane e da escala de Jadad. Com o auxílio do programa Review Manager versão 5.1 foram obtidas as seguintes estatísticas: medidas de sumário baseadas na diferença de medias padronizadas (DMP) e no odds ratio (OR). A obtenção dos respectivos intervalos de confiança (IC) seguiu um modelo de efeitos fixos ou randômicos, dependendo da heterogeneidade observada. Os resultados foram apresentados por meio de gráficos Forest Plot. Foram identificados 995 estudos, sendo que somente 26 foram incluídos na metanálise. Considerando as diferentes características das intervenções e dos desfechos dos estudos, foram obtidas 13 metanálises. Os resultados das metanálises das intervenções isoladas com atividade física apresentaram efeito estatisticamente significativo na redução da dobra cutânea tricipital DMP: -0,43 (IC95%: -0,51; -0,34), (P<0,01) e do percentual de gordura corporal com DMP: -0,13 (IC95%: -0,27; 0,00), (P=0,05). Também, houve efeito estatisticamente significativo das intervenções isoladas com educação nutricional na redução do sobrepeso com OR: -0,77 (IC95%: 0,66; 0,90), (P<0,01). Além disso, os escolares que receberam intervenções combinadas com atividade física e educação nutricional apresentaram redução significativa da obesidade com OR: 0,56 (IC95%: 0,43; 0,74), (P<0,01), do sobrepeso com OR: 0,75 (IC95%: 0,63; 0,89), (P<0,01) e do índice de massa corporal, com DMP: -0,37 (IC95%: -0,63; -0,12), (P<0,01). Não houve efeito significativo das intervenções na redução da circunferência da cintura e da dobra cutânea bicipital. Através desta metanálise concluiu-se que as intervenções combinadas com atividade física e educação nutricional tiveram os efeitos mais positivos como estratégia para a prevenção e controle da obesidade em escolares, do que quando aplicadas isoladamente. / The prevalence of obesity is increasing an alarming rate. Because of that strategies for the prevention and control of obesity are important. With that in mind, a systematic review with meta-analysis was conducted to evaluate the effects of intervention programs stimulating physical activity and healthy eating, and identifying those more suitable for the prevention and control of obesity among school-age students. We included randomized controlled trials available in the following electronic databases: Pubmed, Lilacs, Embase, Scopus, Web of Science and Cochrane Library. The quality of these studies was evaluated taking into account the criteria of allocation concealment proposed by the Cochrane Collaboration and the Jadad Scale. Summary statistics based on the standardized mean difference (SMD) and on the odds ratio (OR) were estimated using Review Manager version 5.1. Confidence intervals (CI) were obtained using fixed or randomized effects models, according to the heterogeneity observed in the data. Results were presented using Forest Plots. Among the 995 studies identified only, 26 were included in the meta-analysis. Considering the different characteristics of interventions and the study outcomes, 13 meta-analyses were obtained. The results of the meta-analyses focusing on isolated interventions based on physical activity showed a statistically significant effect on the reduction of triceps skinfold SMD: -0.43 (CI 95%: -0.51;-0.34), (P< 0.01) and body fat percentage with SMD: -0.13 (CI 95%: -0.27; 0.00), (P= 0.05). There was also a statistically significant effect of the isolated interventions isolated based on nutritional education on the reduction of overweight with OR: -0.77 (CI 95%: 0.66; 0.90), (P< 0.01). Besides that, students who received an intervention combining physical activities and nutritional education presented a significant reduction of obesity with OR: 0.56 (CI 95%: 0.43; 0.74), (P< 0.01), overweight with OR: 0.75 (CI 95%: 0.63; 0.89), (P< 0.01) and Body Mass Index with SMD: -0.37 (CI 95%: -0.63; -0.12), (P< 0.01). We did not find a significant effect of interventions on the reduction of waist circumference and biceps skinfold. In closing, these meta-analyses indicated that combined interventions based on physical activity and nutritional education had a more positive effect as a feasible strategy for the prevention and control of obesity in school-age students when compared to isolated.

Considerando a hipótese de que o excesso de utilização de novas tecnologias e a escassez de recursos de saúde pode apresentar desfechos similares quanto às taxas de baixo peso ao nascer (BPN), foi analisada a associação entre a posição social, obtida pelo grau de instrução, da ocupação profissional e da renda maternos e as taxas de baixo peso ao nascimento. Trata-se de revisão sistemática de literatura e metanálise, nas quais foram incluídos estudos de coorte e transversais, disponíveis na base de dados bibliográficos MEDLINE, utilizando a estratégica de busca previamente definida com a inclusão dos descritores: “socioeconomic factors”, “infant, low birth weight”, “cohort studies”, “crosssectional studies”. A metanálise foi realizada através do Programa STATA 10.0, utilizando o comando metan. Identificaram-se 729 estudos, tendo sido incluídos na metanálise 12. Para a obtenção das medidas de sumário de efeito foi utilizado o modelo de efeito aleatório, e os seus resultados foram apresentados por intermédio dos gráficos Forest Plot. O instrumento empregado na avaliação da qualidade dos estudos foi a Escala de Newcastle-Ottawa. O resultado, num formato de pontuação, mostrou que a maioria dos estudos incluídos apresentou alta qualidade. Os resultados das metanálises mostram uma associação entre a condição socioeconômica da mãe e a proporção de BPN. O estrato social mais elevado, quando identificado, pelo grau de instrução apresenta um risco de BPN de 0,67 do risco do estrato social mais baixo. Esse risco relativo é de 0,68, quando o nível social é identificado pela ocupação profissional e 0,61 quando a identificação é feita usando o nível de renda. Esses achados indicam que uma melhor condição socioeconômica protege em cerca de 30% para o risco de BPN. No estrato social médio, os resultados não foram significativos para o grau de instrução e para a ocupação, não sendo possível determinar sua relação com o BPN; já para a variável renda, a magnitude de efeito foi de 0,81, demonstrando que pertencer ao estrato social médio protege em 19% para o risco de baixo peso ao nascimento. Na análise do viés de publicação, foi utilizado o Teste de Egger. Os resultados apresentados por intermédio dos gráficos Funnel Plot, não demonstraram presença de viés de publicação, exceto na análise do grau de instrução materno médio (P= 0,027). Para o recálculo do tamanho de efeito desta análise, foi utilizado o Método Trim and Fill. Na primeira avaliação, este foi de 0,86 (IC 95%: 0,69 – 1,06) e, na segunda, de 0,71 (IC 95%: 0,56 – 0,88). Os resultados obtidos demonstram que a hipótese de similaridades entre extremos da distribuição social em relação taxa de BPN não foi confirmada e que, apesar de décadas de investigações e intervenções na área da saúde materno-infantil, as desigualdades sociais permanecem como um importante fator de impacto sobre os desfechos perinatais como o BPN. / Considering the hypothesis that an excessive use of new technologies and the lack of heath resources can present similar outcomes concerning low weight rates at birth (LWB), it was analyzed the association between the social standing, which was obtained from the maternal schooling, profession and income, and the low weight rates at birth. It deals with a literature and meta-analysis systematic revision, in which cohort and cross-sectional studies were included, available at the MEDLINE bibliographic database, using the previously defined strategical search including the descriptors: socio-economic factors, infant low birth weight, cohort studies, and cross-sectional studies. The meta-analysis was done through the STATA Program 10.00, using the metan command. Seven hundred and twenty-nine studies were identified, which were included in the meta-analysis 12. In order to obtain summary measures of effect, the odd ration effect model was used and its results were presented through Forest Plot graphics. The instrument employed to assess quality was the Newcastle- Ottawa Scale. The result, in a punctuation format, showed that the majority of studies included were of high quality. The meta-analysis results revealed a relationship between the upper class maternal socio- economical condition and the birth weight. For schooling it was 0.67; for profession, 0.68; and for income 0.61. These findings show that a better socioeconomical condition protects in 30% the LWB risk. In the middle classes, the results were not meaningful for schooling and occupation, being not possible to determine its relationship with the LWB; therefore, the income variant magnitude effect was 0.81, showing that belonging to the middle class protects in 19% against the low weight risk at birth. In the publication bias analysis, the Egger Test was used. The results obtained from the Funnel Plot graphics showed no publication bias, except for the maternal schooling average (P=0.027). For the effect size recalculation of this analysis, it was used the Trim and Fill Method. In the first assessment, it was 0.86 (IC 95%: 0.69 - 1.06) and in the second, 0.71 (IC 95%; 0.56 - 0.88). The results obtained showed that similarity hypotheses between the extremes from the social distribution in relation to the LWB rate was not confirmed and that, despite the decades of investigation and intervention in the maternal infant health area, the social inequalities remain as an important impact factor over the prenatal outcomes like the LWB.

The motivation for subjective well-being research rather than Gross Domestic Product (GDP) is becoming important to the roles of many governments across the globe and so identifying the strongest correlates of subjective well-being is vital as a starting point to informing policies that support subjective well-being. This thesis investigated the correlates of subjective well-being. Chapter 1 introduced the topic and has been divided into two parts: section 1 explores the motivation for subjective well-being research and section 2 presents the conceptual model of subjective well-being. Chapter 2 gave the rationale for the methodological approaches taken to investigate factors that are associated with subjective well-being. Also, the methods chapter presented limitations of the data used. Chapter 3 explored the determinants of subjective well-being in representative samples of nations; and the results obtained in chapter 3 led to three systematic reviews and meta-analyses (Chapter 4, 5 and 6). Chapter 4 conducted a systematic review and meta-analysis of the association between income inequality and subjective well-being to test the general assumption that people's subjective well-being can be increased by tackling income inequality and investigated inconsistencies of previous studies reporting a negative, positive or no association between income inequality and subjective well-being. Chapter 5 carried out a systematic review and meta-analysis of the association between health status and subjective well-being because the results of the empirical study conducted in chapter 3 suggest that health status is positively associated with subjective well-being. Chapter 6 conducted a systematic review and meta-analysis of the association between financial satisfaction and subjective well-being as the results of the empirical study conducted in chapter 3 suggest that financial satisfaction is positively associated with subjective well-being. Chapter 7 discussed the results, highlighted the need for further studies and policy directions and concluded. Taken altogether these studies suggest that: (1) subjective well-being is important to informing policies that support subjective well-being, (2) they might be circumstances where income inequality may not be associated with people's subjective well-being, (3) health status and financial satisfaction are positively associated with subjective well-being and the magnitude of the association is affected by key operational and methodological factors, (4) life satisfaction might be preferred to happiness as a measure of subjective well-being because it may better captures the influence of health status and financial satisfaction, (5) government policies that support subjective well-being measures should consider using self-reported health status and financial satisfaction amongst factors that are correlated with people's subjective well-being, (6) the association between health status and subjective well-being and the link between financial satisfaction and subjective well-being are medium and further research is required to identify other strongest correlates of subjective well-being.

Charles University in Prague Faculty of Social Sciences Institute of Economic Studies DISSERTATION THESIS ABSTRACT Essays on Natural Resource Richness, Economic Growth and Institutional Quality Author: Ayaz Zeynalov Supervisor: Doc. Roman Horvath, Ph.D Academic Year: 2016/2017 Abstract This dissertation consists of three empirical papers on natural resource, economic growth and institutional quality. The first paper analyzes possible publication bias and the reason for contradictory findings in the natural resource literature, the second paper examines the effect of natural resource exports on manufacturing performance in the 15 former Soviet Union countries, and the last addresses whether similarities in country income size and at the institutional level encourage increased amounts of bilateral trade between countries. An introductory chapter puts these three papers into perspective. In the first paper, I analyze 43 studies providing 605 different regression specifica- tions and found that approximately 40% report a negative and statistically significant effect, another 40% report no effect, and the remaining 20% report a positive and statistically significant effect of natural resources on economic growth. The findings show that including interaction between natural resources and institutional quality, controlling...

The impact of inequality on economic growth has become a topic of broad and current interest. Multiple researches investigated the issue but the disparity of opinions and empirical results is huge. The present thesis revises the pri- mary literature through a meta-analytical approach applying Bayesian Model Averaging (BMA) estimation technique. We examine 562 estimates collected from 58 studies published between 1991 and 2015. I find the evidence of the publication bias presence in the literature. The authors of primary studies tend to report preferentially negative and significant estimates. The BMA results suggest that the effect of inequality on growth is not straightforward and is likely not linear. A single pattern for inequality/growth relationship is not fea- sible since the results vary across used income inequality measures, estimation methods and data structure and quality. JEL Classification D31, O10, C11, C82 Keywords meta-analysis, inequality, economic growth, Bayesian model averaging, publication bias Author's e-mail 23376990@fsv.cuni.cz Supervisor's e-mail zuzana.havrankova@fsv.cuni.cz