L'Université, comme lieu de formation du savoir et de formation au savoir, paraît dédiée au culte d’une science désintéressée. La logique de son fonctionnement, sa rationalité, semble ainsi étrangère aux modalités économiques d’organisation des sociétés de marché. Pourtant, depuis le Moyen Âge, l’Université forme des professionnels. À l’heure de sa démocratisation, et dans un contexte d’économie de la connaissance, est-il alors certain qu’elle s’oppose frontalement à la rationalisation économique des règles sociales ? Ne faut-il pas plutôt penser qu’elle a désormais un rôle décisif à jouer, en bonne intelligence avec la logique économique ? Sa capacité à (re)distribuer aux individus des chances d’accéder à des positions socio-professionnelles en fait une institution exemplaire de la manière dont le problème de la justice sociale se pose dans des sociétés développées. / University considered in the first place as an Institution of and to knowledge seems purely devoted to Science. The logic of its function, its rationality, thus appears to be unfamiliar to the economical method of the organization of the market society. Yet since the Middle Age University trains professionals. Now in these democratic times and in such context of economical knowledge, is it normal that it shows fierce resistance to the economical rationalization of social rules? Hence mustn’t we rather think that she has from now on to take a crucial decision, to cope with economical logic? Its ability to give to individuals access to social professional positions makes of this institution an example in the way how developed societies are face with social justice.

Neste estudo concernente ao paricá (Schizolobium amazonicum), objetivou-se avaliar a interação entre diferentes fontes de explantes e a concentração de fitorreguladores sobre a indução de calos, por meio da abordagem de modelos de regressão linear generalizados. Os explantes foram extraídos de plântulas germinadas in vitro. Foram utilizados os segmentos apical, cotiledonar, nodal e de raiz para indução com 2,4-diclorofenoxiacético (2,4-D) de 0,0 a 2,1 mg L-1 e, foi ainda testado o segmento hipocotiledonar para a indução com a combinação de ácido indol-3-butírico (AIB) e 6-benzilaminopurina (BAP), ambos de 0,0 a 1,0 mg L-1. Foi utilizado o meio Woody Plant Medium (WPM) acrescido de 0,3% de sacarose, 0,1% de polivinilpirrolidone (PVP) e 0,6% de ágar. Os parâmetros avaliados foram o percentual de calos, o percentual de cobertura do explante por calos e a razão da massa seca pela massa fresca. Como resultados, obteve-se 100% de calogênese nos explantes em contato com as menores concentrações aplicadas, exceto os segmentos de raiz. Também se reporta que as maiores chances de cobertura por calos foram observadas no explante nodal, tendo-se utilizado 2,4-D, e no explante cotiledonar, ao se empregar a combinação de AIB com BAP. Os modelos ajustados selecionados se mostraram uma boa alternativa de inferência por permitirem a avaliação e estudo de variáveis independentes com distribuição diferente da normal. / In this study concerning the parica (Schizolobium amazonicum), it was aimed to evaluate the interaction among different explant sources and plant growth regulators concentration on callus induction, through the approach of generalized linear models. The explants were extracted from seedlings germinated in vitro. Apical, cotyledon, root and nodal segments were used for induction with 2,4-dichlorophenoxyacetic acid (2,4-D) from 0,0 to 2,1 mg L-1 and, was still tested hypocotyledonary segments for induction with indol-3-butyric acid (IBA) and 6-benzylaminopurine (BAP) combined from 0,0 to 1,0 mg L-1. It was used Woody Plant Medium (WPM) supplemented with 0,3% sucrose, 0,1% polyvinylpyrrolidone (PVP) and 0,6% agar. It was evaluated the callus percentage, the explant coverage percentage by callus and the ratio of dry mass by fresh mass. As a result, it was obtained 100% callus formation in that explants in contact with the lower concentrations applied, except for the root segments. It also reports that the higher chances of callus coverage were observed at the nodal explant, having been used 2,4-D, and the cotyledon explants, by using the IBA with BAP combination. The adjusted models selected proved itself a good inference alternative for allowing the evaluation and study of independent variables with distribution other than normal.

Estudos que comprovem a superioridade de um fármaco em relação a outros já existentes no mercado são de grande interesse na prática clínica. Através deles a Agência Nacional de Vigilância Sanitária (ANVISA) concede registro a novos produtos, que podem curar mais rápido ou aumentar a probabilidade de cura dos pacientes, em comparação ao tratamento padrão. É de suma importância que os testes de hipóteses controlem a probabilidade do erro tipo I, ou seja, controlem a probabilidade de que um tratamento não superior seja aprovado para uso; e também atinja o poder de teste regulamentado com o menor número de indivíduos possível. Os testes de hipóteses existentes para esta finalidade ou desconsideram o tempo até que o evento de interesse ocorra (reação alérgica, efeito positivo, etc) ou são baseados no modelo de riscos proporcionais. No entanto, na prática, a hipótese de riscos proporcionais pode nem sempre ser satisfeita, como é o caso de ensaios cujos riscos dos diferentes grupos em estudo se igualam com o passar do tempo. Nesta situação, o modelo de chances proporcionais é mais adequado para o ajuste dos dados. Neste trabalho desenvolvemos e investigamos dois testes de hipóteses para ensaios clínicos de superioridade, baseados na comparação de curvas de sobrevivência sob a suposição de que os dados seguem o modelo de chances de sobrevivências proporcionais, um sem a incorporação da fração de cura e outro com esta incorporação. Vários estudos de simulação são conduzidos para analisar a capacidade de controle da probabilidade do erro tipo I e do valor do poder dos testes quando os dados satisfazem ou não a suposição do teste para diversos tamanhos amostrais e dois métodos de estimação das quantidades de interesse. Concluímos que a probabilidade do erro tipo I é subestimada quando os dados não satisfazem a suposição do teste e é controlada quando satisfazem, como esperado. De forma geral, concluímos que é imprescindível satisfazer as suposições dos testes de superioridade. / Studies that prove the superiority of a drug in relation to others already existing in the market are of great interest in clinical practice. Based on them the Brazilian National Agency of Sanitary Surveillance (ANVISA) grants superiority drugs registers which can cure faster or increase the probability of cure of patients, compared to standard treatment. It is of the utmost importance that hypothesis tests control the probability of type I error, that is, they control the probability that a non-superior treatment is approved for use; and also achieve the test power regulated with as few individuals as possible. Tests of hypotheses existing for this purpose or disregard the time until the event of interest occurrence (allergic reaction, positive effect, etc.) or are based on the proportional hazards model. However, in practice, the hypothesis of proportional hazards may not always be satisfied, as is the case of trials whose risks of the different study groups become equal over time. In this situation, the proportional odds survival model is more adequate for the adjustment of the data. In this work we developed and investigated two hypothesis tests for clinical trials of superiority, based on the comparison of survival curves under the assumption that the data follow the proportional survival odds model, one without the incorporation of cure fraction and another considering cure fraction. Several simulation studies are conducted to analyze the ability to control the probability of type I error and the value of the power of the tests when the data satisfy or not the assumption of the test for different sample sizes and two estimation methods of the quantities of interest. We conclude that the probability of type I error is underestimated when the data do not satisfy the assumption of the test and it is controlled when they satisfy, as expected. In general, we conclude that it is indispensable to satisfy the assumptions of superiority tests.

Cette thèse porte sur la mesure et la compréhension des inégalités des chances en santé c'est-à-dire aux inégalités attribuables à des facteurs ne relevant pas de la responsabilité individuelle, tel que le milieu d'origine. Nous portons un intérêt spécifique à la contribution des comportements de santé dans la construction de ces inégalités. Nous développons notre analyse à travers trois axes : (i) la mesure de l'importance respective des conditions de vie dans l'enfance, du niveau d'éducation et des comportements de santé dans l'explication des inégalités de santé ; (ii) l'analyse des mécanismes en jeu dans la transmission intergénérationnelle des comportements de santé avec l'exemple du tabagisme et des habitudes de soins ; (iii) la mesure des différences entre pays européens dans les inégalités des chances en santé. Les analyses empiriques combinent des données prospectives d'une cohorte britannique ainsi que des données rétrospectives issues d'une enquête française et d'une enquête européenne. Les résultats soulignent la contribution aux inégalités de santé des conditions de vie dans l'enfance et du niveau d'éducation de façon directe et de façon indirecte par les comportements de santé. / This thesis focuses on the measurement and the understanding of inequality of opportunity in health which are inequalities related to factors beyond the individual responsability, such as the individual's social background. We focus on the contribution of health-related behaviors in the construction of these inequalities. Our analysis is based on three topics: (i) the measure of the respective contribution of early-life conditions, education and lifestyles to health inequality ; (ii) the analysis of the intergenerational transmission of health-related behaviors with the example of smoking and health care habits ; (iii) the measure of cross-country differences in inequality of opportunity in health with a European perspective. Empirical analysis are conducted with both prospective data using a British cohort and retrospective data using a French study and a European study. The results emphasize the contribution of early-life conditions and education to health inequality both directly and indirectly through lifestyles.

Cette thèse se concentre sur la mesure des dissemblances dans la distribution des attributs économiques, et sur les implications pour l'inégalité des chances. L'égalité des chances a gagné en popularité pour définir l'objectif de la distribution d'une vaste gamme de résultats économiques entre les groupes sociaux. Cette thèse est motivée par le fait que l'évaluation des politiques publiques fondée sur l'égalité des chances s'appuie toujours sur des comparaisons de dissemblance entre des distributions conditionnelles, et nous proposons des critères empiriques pour vérifier ces comparaisons. Dans le premier chapitre, nous caractérisons axiomatiquement le pre-ordre de dissemblance permettant de classer les distributions conditionnelles au groupe d'origine, que sont définies sur des classes de résultats discrètes. Lorsque les classes sont permutables, nous démontrons que la dissemblance est rationalisée par un ordre de majorisation de matrices et mis en œuvre en vérifiant l'inclusion des zonotopes . Lorsque les classes sont ordonnées nous fondons le jugement de dissemblance sur un nombre fini de comparaisons au sein de la majorisation au sens de Lorenz entre les proportions des groupes, vérifiées à des étapes différentes de cumulation de la population agrégée. Dans le deuxième chapitre, on examine la pertinence du pre-ordre de dissemblance pour étudier la ségrégation au niveau individuel. On obtient une caractérisation complète d'une famille bien définie d'indicateurs de ségrégation et nous étudions l'un d'eux, l'indice d'exposition de Gini, en utilisant des données italiennes. Le dernier chapitre présente un critère d'égalisation des chances. L'égalité des chances est atteinte lorsqu'il n'y a pas de consensus, selon une classe de préférences donnée, sur l'identité du groupe défavorisé. Nous utilisons les changements de (manque de) consensus sur l'existence et l'étendue du désavantage pour caractériser le critère d'égalisation des chances. Les restrictions nécessaires, autant que des procédures possibles d'agrégations, sont également discutées. Nous démontrons que ce critère est identifié selon la classe de préférences représentées par les fonctions d'utilité dépendantes du rang et on obtient des résultats innovants d'inférence sur la dominance stochastique inverse qui nous permettent de tester ce critère. Deux applications sur des données françaises illustrent l'impact en termes d'égalisation de chances des politiques éducatives qui ont lieu tôt dans la vie des étudiants.

Les outils d'évaluation, utilisés en France pour éclairer les décideurs sur les politiques de transports en milieu urbain, n'intègrent pas pleinement la dimension sociale de ces politiques. Or, cet enjeu devient primordial. En s'appuyant sur l'approche théorique de Amartya Sen, l'objectif de notre thèse est de construire une méthode d'évaluation permettant de rendre compte de cette dimension, en termes d'égalité des chances d'accès aux activités de reproduction sociale. Pour ce faire, nous construisons un outil identifiant, au niveau infra-communal, les groupes sociaux et les quartiers riches ou pauvres de l'espace urbain. Puis, nous construisons un indicateur d'accessibilité à un panier de biens. Cet indicateur mesure, pour un mode de déplacement donné, pour chaque service du panier de biens et depuis le lieu de résidence, le temps maximal d'accès au nombre d'activités pour 1 000 habitants. Cet indicateur prend en compte les différents sous-systèmes de l'espace urbain.<br /><br />La méthode est déclinée sur l'agglomération lyonnaise. Nous montrons, tout d'abord, l'existence entre les individus et entre les quartiers d'inégalité de chances d'accès au panier de biens en 1999. Ces inégalités procèdent principalement d'un inégal accès à l'automobile (trois fois plus performante que les transports collectifs en termes de temps d'accès au panier de biens), mais aussi de la répartition hétérogène des activités et de la qualité de l'offre en transports collectifs. L'évolution de la localisation des activités entre 1990 et 1999 met en évidence une aggravation des inégalités d'accès entre les quartiers pauvres et les quartiers riches. Tout se passe comme si les populations aisées façonnaient la division fonctionnelle du sol et la localisation des activités, au détriment des populations les plus fragiles. Nous montrons enfin qu'une amélioration forte de l'offre en transports collectifs a des impacts limités qui souvent peinent à compenser les pertes d'accessibilité dues à l'évolution de la localisation des activités.

The thesis consists of an introduction and three studies, which are comparative sociological and institutional analyses of a number of countries' publicly regulated vocational education and training systems at upper secondary and post-secondary level. Official regulation of programme content and curricula - the main empirical material - is interpreted, and focus is directed on aspects which distinguish between, among others, general and vocational elements of teaching, learning and training. The analyses employ concepts from educational studies, historical sociology and institutional sociology with the aim of conceptual elaboration and illumination of similarities and differences between cases. Study I compares upper secondary vocational education and training programmes in Denmark, Finland, Norway and Sweden. The programmes are analysed and compared with respect to scope, sequence and selection of curricular units of subject matter and periods of training. Programmes combine general and vocational elements in various ways, thereby offering choices for obtaining both the occupation-related qualification, as well as higher education eligibility. The combination of curricular units with additional subject matter seems to, in some cases, lead to emerging learning aims. Study II focuses on vocational education and training systems at upper secondary level in Denmark, Norway and Sweden, and seeks to identify differences and similarities by conceptualising features of ideal types along different dimensions. It furthermore develops a method of analysis and visualisation which enables more accurate understanding of how various learning aims are designed within educational programmes. Some of the results indicate that the resources necessary for attaining the full, intended educational qualification are not guaranteed in any of the three cases; however, this applies particularly to apprenticeship-based programmes, where it can be attained only after training contracts with status-like features, resembling parts of occupational jurisdictions, are seized within sheltered circumstances. In the school-based programmes, a larger proportion of resources are guaranteed, but training is shorter and less vocationally oriented. Vocational education and training programmes in all three systems orient learning not only towards vocational, but also general learning aims. Study III compares admission criteria to post-secondary vocational education and training programmes in Denmark and Sweden, focusing on the Danish erhvervsakademi and the Swedish yrkeshögskola. Options, as elements of life chances, for individuals holding upper secondary vocational education qualifications who wish to apply for admission to these programmes, differ in terms of scope between the two cases. In the Swedish case, options are wide and mostly structured by course-based admission criteria; while in the Danish case, options are more narrow and structured by qualification-based admission criteria. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 2: Manuscript. Paper 3: Manuscript.</p>

Resumo não disponível.

Resumo não disponível.

Estudos que comprovem a superioridade de um fármaco em relação a outros já existentes no mercado são de grande interesse na prática clínica. Através deles a Agência Nacional de Vigilância Sanitária (ANVISA) concede registro a novos produtos, que podem curar mais rápido ou aumentar a probabilidade de cura dos pacientes, em comparação ao tratamento padrão. É de suma importância que os testes de hipóteses controlem a probabilidade do erro tipo I, ou seja, controlem a probabilidade de que um tratamento não superior seja aprovado para uso; e também atinja o poder de teste regulamentado com o menor número de indivíduos possível. Os testes de hipóteses existentes para esta finalidade ou desconsideram o tempo até que o evento de interesse ocorra (reação alérgica, efeito positivo, etc) ou são baseados no modelo de riscos proporcionais. No entanto, na prática, a hipótese de riscos proporcionais pode nem sempre ser satisfeita, como é o caso de ensaios cujos riscos dos diferentes grupos em estudo se igualam com o passar do tempo. Nesta situação, o modelo de chances proporcionais é mais adequado para o ajuste dos dados. Neste trabalho desenvolvemos e investigamos dois testes de hipóteses para ensaios clínicos de superioridade, baseados na comparação de curvas de sobrevivência sob a suposição de que os dados seguem o modelo de chances de sobrevivências proporcionais, um sem a incorporação da fração de cura e outro com esta incorporação. Vários estudos de simulação são conduzidos para analisar a capacidade de controle da probabilidade do erro tipo I e do valor do poder dos testes quando os dados satisfazem ou não a suposição do teste para diversos tamanhos amostrais e dois métodos de estimação das quantidades de interesse. Concluímos que a probabilidade do erro tipo I é subestimada quando os dados não satisfazem a suposição do teste e é controlada quando satisfazem, como esperado. De forma geral, concluímos que é imprescindível satisfazer as suposições dos testes de superioridade. / Studies that prove the superiority of a drug in relation to others already existing in the market are of great interest in clinical practice. Based on them the Brazilian National Agency of Sanitary Surveillance (ANVISA) grants superiority drugs registers which can cure faster or increase the probability of cure of patients, compared to standard treatment. It is of the utmost importance that hypothesis tests control the probability of type I error, that is, they control the probability that a non-superior treatment is approved for use; and also achieve the test power regulated with as few individuals as possible. Tests of hypotheses existing for this purpose or disregard the time until the event of interest occurrence (allergic reaction, positive effect, etc.) or are based on the proportional hazards model. However, in practice, the hypothesis of proportional hazards may not always be satisfied, as is the case of trials whose risks of the different study groups become equal over time. In this situation, the proportional odds survival model is more adequate for the adjustment of the data. In this work we developed and investigated two hypothesis tests for clinical trials of superiority, based on the comparison of survival curves under the assumption that the data follow the proportional survival odds model, one without the incorporation of cure fraction and another considering cure fraction. Several simulation studies are conducted to analyze the ability to control the probability of type I error and the value of the power of the tests when the data satisfy or not the assumption of the test for different sample sizes and two estimation methods of the quantities of interest. We conclude that the probability of type I error is underestimated when the data do not satisfy the assumption of the test and it is controlled when they satisfy, as expected. In general, we conclude that it is indispensable to satisfy the assumptions of superiority tests.