Cette recherche a pour objet l'analyse des politiques de la réforme éducative en Haïti, de 1979 à 2013. Nous considérons, en effet, une action publique en éducation comme une figure exemplaire des politiques publiques. Nous adoptons une approche analytique flexible suivant quatre grandes étapes cycliques, à savoir : la mise à l'agenda, la programmation, la mise en œuvre et l'évaluation. Dans cet ordre d'idées, nous développons notre argumentation en deux grandes parties. Dans la première partie, nous exposons les trois aspects suivants : 1) les hypothèses relatives au profil actuel et historique d'Haïti ; 2) les hypothèses relatives à notre approche plurielle d'analyse des politiques publiques ; 3) les deux groupes d'impératifs contradictoires dans les arrangements politico-administratifs des politiques éducatives en Haïti. Dans la seconde partie, nous essayons de réaliser une analyse interprétative des résultats de nos deux enquêtes qualitative et quantitative. Nous tentons de mettre en évidence la "démocratisation uniforme" de l'enseignement en Haïti, au cours des trois dernières décennies. En guise d'ouverture pour des recherches ultérieures, nous maintenons une approche constructiviste des problèmes publics. A notre sens, le système éducatif haïtien est un "construit social" qui résiste aux politiques de la réforme éducative, de 1979 à 2013. Il nous semble donc pertinent de chercher à approfondir les grands enjeux de la "déconstruction" et de la "reconstruction" pour favoriser la "démocratisation égalisatrice" de l'enseignement en Haïti. / This research dealt with the public policy analysis of educational reform in Haiti from 1979 to 2013. I regarded public action in education as an illustrative example of public policy. Then I adopted a flexible analytical approach that contains four cyclic stages, namely: the agenda setting, programming, implementation and evaluation. In this vein, I developed the argumentation in two parts. In the first part, I presented the following three aspects: 1) hypotheses relating to current and historical profile of Haiti; 2) hypotheses relating to the pluralistic approach to public policy analysis; 3) the two groups of conflicting requirements in the political-administrative arrangements of educational policies in Haiti. In the second part, I tried to do an interpretative analysis of the results of both qualitative and quantitative research. I tried to emphasize the "uniform democratization" of education in Haiti during the past three decades. As an opening for further research, I maintained a constructivist approach to social problems. In my view, the Haitian educational system was a "social construct" that resisted the policies of educational reform from 1979 to 2013. So it seems appropriate to go deeper into the major issues of the "deconstruction" and "reconstruction" in order to promote the "equalizing democratization" of education in Haiti.

In the light of growing concerns regarding the quality of higher education after a period of rapid expansion, in 2003, the Chinese government launched the Undergraduate Teaching Evaluation (UTE), the first nation-wide evaluation of universities. In 2008, the UTE was replaced by the Quality Project, which signalled a change in its quality assurance approach, with a move from evaluations to the issuing of awards. In order to investigate the impact of the two national quality assurance policies, along with the impact of two long-standing internal quality assurance mechanisms employed by universities - class observation and student evaluation of teachers - on teaching, the researcher adopted policy analysis and a case study approach. Three different universities in the same region were selected as cases. Semi-structured interviews with 56 heads of department and teachers across three departments at each university were conducted. National and university policy documents, as well as interview data, were analysed thematically in the light of concepts derived from the political sciences, namely Knoepfel et al.'s (2007; 2011) framework for policy analysis and Schneider and Ingram's (1990) classification of policy tools. University policies and interview data revealed the patterns of impact of these two quality assurance policies. Faced with the UTE inspection, universities shifted their focus from assuring the quality of teaching to achieving good results in the exercise and therefore engaging in 'game-playing'. In order to ensure good UTE results, universities even resorted to the manipulation of data. University policies resulting from the UTE required teachers to produce standardised documents and to follow particular procedures. The majority of interviewed teachers regarded the impact of the UTE with cynicism, seeing it as a waste of time and effort, an interference with academic freedom, and believing it had the effect of undermining ethics. However, some teachers reported as positive impact of the fact that the UTE helped to keep teachers disciplined, and that it provided an impetus for ensuring teaching quality. Interview data showed that the Quality Project awards only had an impact on the award winners, who perceived the incentives on offer to be substantial. These award winners considered the acknowledgement given by the awards and the information obtained through reflection and good examples as valuable impacts. However, the Quality Project awards did not have impact on non-award winners, i.e. the majority of teachers. This study revealed that in order to have impact the institutional practices of class observation and the student evaluation of teachers relied on a number of conditions being in place. These practices had impact when the stakes were high or when university teachers received valid and reliable feedback. Whilst high stakes were found to lead to mixed impact, valid and reliable feedback contributed to the improvement of teaching. This study contributes to the understanding of the impact of quality assurance policies and mechanisms on teaching in Chinese higher education, an area that has not yet been the subject of significant empirical research. Covering the two major quality assurance policies in recent years, and also the institutional mechanisms teachers face, this research was able to capture the interdependence between these policies. No such research has previously been conducted in the context of Chinese higher education. In regard to the theoretical aspect of the research, the empirical evidence that was collected and a comprehensive review of other empirical research enabled the development of an Impact Framework. The Impact Framework identified patterns in the impact of various quality assurance policy tools and revealed their inherent strengths and weaknesses. Hence, the Impact Framework can serve in the future as an important reference for policy-makers who are seeking to design and implement effective quality assurance policy tools.

La réalisation des politiques de protection sociale en Afrique soulève des questions complexes, notamment leur raison d’être et le choix des contenus (Merrien, 2013). L’originalité de cette thèse tient au fait qu’elle soit la première du genre à analyser l’élaboration des politiques de protection sociale au Burkina Faso autre que celles dans le domaine de la santé. À travers une analyse de la politique nationale de protection sociale et programme pilote de protection sociale en santé des personnes démunies, la thèse répond aux questions suivantes : pourquoi et comment la protection sociale est devenue une priorité gouvernementale et quels sont les facteurs qui ont influencé la mise en forme de ces deux politiques publiques et la mise en œuvre du programme de solidarité ? À partir de cadres d’analyse issue de diverses perspectives disciplinaires (science politique, sociologie, anthropologie), la réalisation des politiques publiques de protection sociale est considérée comme le fruit de l’interaction d’une multitude d’acteurs et de leurs intérêts divers : organisations internationales, coopération bilatérale, structure étatique, associations et ONG locales. La méthodologie est essentiellement qualitative. L’entrevue semi-dirigée approfondie, l’analyse documentaire et l’entrevue informelle ont été utilisées pour collecter les données. Les résultats sont structurés en cinq articles scientifiques. La décision de formuler ces deux politiques était principalement motivée par la recherche d’une cohésion sociale pour maintenir la paix et l’ordre social (article 1 et 4). Le processus de la formulation n’a pas permis de conduire une réflexion sur le problème à résoudre, d’identifier les besoins spécifiques des bénéficiaires, d’évaluer la capacité des solutions pour choisir celles appropriées. Les logiques des acteurs nationaux étaient de maintenir leurs acquis financiers et de mobiliser des ressources supplémentaires. Les acteurs internationaux ont été guidés par la logique de promotion d’instrument de protection sociale en adéquation avec leurs valeurs, leurs idées, leurs missions, et ce pour la mise en œuvre de leur mandat institutionnel. Cela a donné lieu au constat empirique de « non-design » ou de non-formulation (article 2 et 4), a révélé l’influence des idées et des intérêts des acteurs dans le choix des solutions, mais a aussi montré qu’une volonté politique est importante pour que les choix puissent répondre aux besoins des populations pauvres et vulnérables. Par ailleurs, les connaissances issues de la recherche scientifique ont été très peu utilisées pour éclairer les choix comparativement aux connaissances issues de la littérature grise (article 3). Les gains politiques, c’est-à-dire la perception des conséquences sociales et politiques associées à une recommandation fondée sur des connaissances scientifiques, influencent la prise en compte de cette dernière. L’analyse de la mise en œuvre montre que des lacunes de formulation influencent l’efficacité d’une politique. Aussi, la mise en œuvre offre des circonstances opportunes d’adaptation et de reformulation pour intégrer les spécificités du contexte et les besoins des bénéficiaires en vue d’améliorer l’efficacité (article5). / The development of social protection policies in Africa raises complex issues, including its rationale and choice of content (Merrien, 2013). This thesis is the first of its kind to analyze the development of social protection policies in Burkina Faso other than those in the field of health. Through an analysis of the National Social Protection Policy and pilot program of Social Protection in the Health of the Poor, this thesis answers the following questions: why and how social protection has become a government priority; and what factors have influenced the shaping of these two public policies and the implementation of the solidarity program? Based on analytical frameworks from various disciplinary perspectives (political science, sociology, anthropology), the process of developing public social protection policies is the result of the interaction of a multitude of actors and their diverse interests: international organizations, bilateral cooperation, state structure, associations, and local NGOs. The methodology is essentially qualitative. This thesis collects data thorough semi-structured interviews, literature review, and informal interviews. The results are structured in five scientific articles. The desire for social cohesion to maintain peace and social order motivated the decision to formulate these two policies (Articles 1 and 4). The formulation process did not lead to a reflection on the problem to be solved, to identify the specific needs of the beneficiaries, to evaluate the capacity of the solutions to choose the appropriate ones. The motive of the national actors was to maintain their financial assets and to mobilize additional resources. International actors have been guided by the logic of promoting a social protection instrument in line with their values, their ideas, and their missions, for the implementation of their institutional mandate. This implementation gave rise to the empirical observation of "non-design" or non-formulation (Article 2 and 4), revealed the influence of the ideas and interests of the actors in the choice of solutions, but also showed that it is important for choices to meet the needs of the poor and vulnerable. In addition, knowledge derived from research data was used very little compared to knowledge from gray literature (Article 3). Political gains, that is, the perception of social and political consequences associated with a recommendation based on scientific knowledge, influence the latter's consideration. The iv analysis of the implementation of the solidarity program shows that formulation gaps influence the effectiveness of a policy. Thus, the implementation offers opportune circumstances for adaptations and reformulations to integrate the specificities of the context and the needs of the beneficiaries with an aim of improving efficiency (Article 5). However, these opportunities have not been utilized by program implementers, helping to keep those who are impoverished excluded from health services.