A área de detecção de outliers (ou detecção de anomalias) possui um papel fundamental na descoberta de padrões em dados que podem ser considerados excepcionais sob alguma perspectiva. Uma importante distinção se dá entre as técnicas supervisionadas e não supervisionadas. O presente trabalho enfoca as técnicas de detecção não supervisionadas. Existem dezenas de algoritmos desta categoria na literatura, porém cada um deles utiliza uma intuição própria do que deve ser considerado um outlier ou não, que é naturalmente um conceito subjetivo. Isso dificulta sensivelmente a escolha de um algoritmo em particular e também a escolha de uma configuração adequada para o algoritmo escolhido em uma dada aplicação prática. Isso também torna altamente complexo avaliar a qualidade da solução obtida por um algoritmo/configuração em particular adotados pelo analista, especialmente em função da problemática de se definir uma medida de qualidade que não seja vinculada ao próprio critério utilizado pelo algoritmo. Tais questões estão inter-relacionadas e se referem respectivamente aos problemas de seleção de modelos e avaliação (ou validação) de resultados em aprendizado de máquina não supervisionado. Neste trabalho foi desenvolvido um índice pioneiro para avaliação não supervisionada de detecção de outliers. O índice, chamado IREOS (Internal, Relative Evaluation of Outlier Solutions), avalia e compara diferentes soluções (top-n, i.e., rotulações binárias) candidatas baseando-se apenas nas informações dos dados e nas próprias soluções a serem avaliadas. O índice também é ajustado estatisticamente para aleatoriedade e extensivamente avaliado em vários experimentos envolvendo diferentes coleções de bases de dados sintéticas e reais. / Outlier detection (or anomaly detection) plays an important role in the pattern discovery from data that can be considered exceptional in some sense. An important distinction is that between the supervised and unsupervised techniques. In this work we focus on unsupervised outlier detection techniques. There are dozens of algorithms of this category in literature, however, each of these algorithms uses its own intuition to judge what should be considered an outlier or not, which naturally is a subjective concept. This substantially complicates the selection of a particular algorithm and also the choice of an appropriate configuration of parameters for a given algorithm in a practical application. This also makes it highly complex to evaluate the quality of the solution obtained by an algorithm or configuration adopted by the analyst, especially in light of the problem of defining a measure of quality that is not hooked on the criterion used by the algorithm itself. These issues are interrelated and refer respectively to the problems of model selection and evaluation (or validation) of results in unsupervised learning. Here we developed a pioneer index for unsupervised evaluation of outlier detection results. The index, called IREOS (Internal, Relative Evaluation of Outlier Solutions), can evaluate and compare different candidate (top-n, i.e., binary labelings) solutions based only upon the data information and the solution to be evaluated. The index is also statistically adjusted for chance and extensively evaluated in several experiments involving different collections of synthetic and real data sets.

Evaluation capacity development (ECD) has been acknowledged as a system of processes to help organizations achieve sustainable evaluation practice. Examining the existing evaluation capacity of an organization before starting an ECD process is necessary and will increase the possibilities of success, determined by the establishment or strengthening of an evaluation system into the organization. In response to this need, this study involved the designing of the Organizational Readiness for Evaluation Capacity Development (ORECD) checklist and its initial validation, using a mixed method research design. The study was conducted in four phases, including: (a) the design of the ORECD checklist based on a review of the literature; (b) a review of the ORECD checklist by five experts to obtain face and content validity evidences, with emphasis on relevance and clarity of the items and how well the items fit the corresponding component; (c) a pretesting about the appropriateness of the wording of the items and format of the ORECD checklist by a sample of doctoral graduate students with formal training in evaluation and professional evaluators; and (d) a field study with 32 nonprofit organizations to determine the utility and benefits of using the ORECD checklist and potential improvements to the instrument. This phase generated information about the psychometric properties as well as consequential validity evidence. Findings indicated that the ORECD checklist has great potential to determine the readiness of an organization to develop evaluation capacity, as demonstrated by the feedback received from various groups of participants, establishing face, content, and consequential validity. Results from the psychometric analysis showed correlations that, for the most part, suggested that the components are measuring aspects of the same construct. In addition, the alpha for most of the components supported the reliability of the ORECD checklist. The two components with alphas close to but below .70 required modifications in order to improve their reliability. Also, it was necessary to modify or reword some of the items. Ongoing efforts should provide information about how the changes made to the ORECD checklist are working and additional validity evidences as the one that can be obtained through factor analysis. This will allow the exploration of the underlying structure of the ORECD checklist and its components. It is expected that the ORECD checklist can be a contribution to the body of literature about ECD helping to address organizational readiness in order to support and sustain the development of evaluation capacity within organizations.

Land use regulations are typically established as a response to development activity. For effective growth management and habitat preservation, the opposite should occur. This study considers tree preservation ordinances of the Dallas-Fort Worth metropolitan area as a means of evaluating development regulation in a metropolitan context. It documents the impact urban cores have on regulations and policies throughout their region, demonstrating that the same urban-rural gradient used to describe physical components of our metropolitan areas also holds true in terms of policy formation. Although sophistication of land use regulation generally dissipates as one moves away from an urban core, native habitat is more pristine at the outer edges. To more effectively protect native habitat, regional preservation measures are recommended.

A área de detecção de outliers (ou detecção de anomalias) possui um papel fundamental na descoberta de padrões em dados que podem ser considerados excepcionais sob alguma perspectiva. Uma importante distinção se dá entre as técnicas supervisionadas e não supervisionadas. O presente trabalho enfoca as técnicas de detecção não supervisionadas. Existem dezenas de algoritmos desta categoria na literatura, porém cada um deles utiliza uma intuição própria do que deve ser considerado um outlier ou não, que é naturalmente um conceito subjetivo. Isso dificulta sensivelmente a escolha de um algoritmo em particular e também a escolha de uma configuração adequada para o algoritmo escolhido em uma dada aplicação prática. Isso também torna altamente complexo avaliar a qualidade da solução obtida por um algoritmo/configuração em particular adotados pelo analista, especialmente em função da problemática de se definir uma medida de qualidade que não seja vinculada ao próprio critério utilizado pelo algoritmo. Tais questões estão inter-relacionadas e se referem respectivamente aos problemas de seleção de modelos e avaliação (ou validação) de resultados em aprendizado de máquina não supervisionado. Neste trabalho foi desenvolvido um índice pioneiro para avaliação não supervisionada de detecção de outliers. O índice, chamado IREOS (Internal, Relative Evaluation of Outlier Solutions), avalia e compara diferentes soluções (top-n, i.e., rotulações binárias) candidatas baseando-se apenas nas informações dos dados e nas próprias soluções a serem avaliadas. O índice também é ajustado estatisticamente para aleatoriedade e extensivamente avaliado em vários experimentos envolvendo diferentes coleções de bases de dados sintéticas e reais. / Outlier detection (or anomaly detection) plays an important role in the pattern discovery from data that can be considered exceptional in some sense. An important distinction is that between the supervised and unsupervised techniques. In this work we focus on unsupervised outlier detection techniques. There are dozens of algorithms of this category in literature, however, each of these algorithms uses its own intuition to judge what should be considered an outlier or not, which naturally is a subjective concept. This substantially complicates the selection of a particular algorithm and also the choice of an appropriate configuration of parameters for a given algorithm in a practical application. This also makes it highly complex to evaluate the quality of the solution obtained by an algorithm or configuration adopted by the analyst, especially in light of the problem of defining a measure of quality that is not hooked on the criterion used by the algorithm itself. These issues are interrelated and refer respectively to the problems of model selection and evaluation (or validation) of results in unsupervised learning. Here we developed a pioneer index for unsupervised evaluation of outlier detection results. The index, called IREOS (Internal, Relative Evaluation of Outlier Solutions), can evaluate and compare different candidate (top-n, i.e., binary labelings) solutions based only upon the data information and the solution to be evaluated. The index is also statistically adjusted for chance and extensively evaluated in several experiments involving different collections of synthetic and real data sets.

SILVA, Esequias Rodrigues da. Professor-avaliador: identidade deontológica e formação no contexto do sistema escolar - o caso do programa aprender pra valer. 2014. 224f. – Dissertação (Mestrado) – Universidade Federal do Ceará, Programa de Pós-graduação em Educação Brasileira, Fortaleza (CE), 2014. / Submitted by Márcia Araújo (marcia_m_bezerra@yahoo.com.br) on 2015-06-22T15:33:55Z No. of bitstreams: 1 2014_dis_ersilva.pdf: 19641342 bytes, checksum: 9de657afc898ceaf23eb06475f15570a (MD5) / Approved for entry into archive by Márcia Araújo(marcia_m_bezerra@yahoo.com.br) on 2015-06-23T12:53:29Z (GMT) No. of bitstreams: 1 2014_dis_ersilva.pdf: 19641342 bytes, checksum: 9de657afc898ceaf23eb06475f15570a (MD5) / Made available in DSpace on 2015-06-23T12:53:29Z (GMT). No. of bitstreams: 1 2014_dis_ersilva.pdf: 19641342 bytes, checksum: 9de657afc898ceaf23eb06475f15570a (MD5) Previous issue date: 2014 / Various initiatives to external large-scale evaluation arise in Brazil during the 1990s, especially: National System of Basic Assessment (SAEB) and National Secondary Education Examination (ENEM). After more than twenty years since the birth of those initiatives, discussions about the schools own internal assessment by teachers have been intensified. Bringing near the outer large-scale evaluation to the school’s internal evaluation arises as a problem today in the Brazilian educational scenario. In this context, we have delimited the object of this research: the evaluator-teacher. We have defined as general objective of this research the discussion on the ethical identity of evaluator-teacher. The specific objectives that guided this study were planned as follows: (1) to discuss the ethical identity of evaluator-teacher in the context of the paradigm docimológico school evaluation, (2) to discuss the ethical identity of evaluator-teacher in the context of the pedagogical paradigm of school evaluation, (3) to identify the reference of powers attributed to the evaluator-teacher on Aprender pra valer (Learning for sure) program, by SEDUC-CE. This research falls under the qualitative approach by stating – among other features – the descriptive and interpretative matter, the interest on the use of the results, use various data sources. The methodological procedures substantiated in documentary analysis and literature review. The data were interpreted by content analysis. For purposes on this study we defined as theoretical references Bloom, Hastings and Madaus (1983), Perrenoud (1999), Afonso (2000), Hadji (1994.2001), Tyler (1949), Bonniol and Vial (2001). The survey results indicated that external evaluation on a large scale will influence on ethical identity of evaluator-teacher by proposing a framework for the construction of assessment instruments: the Reference Matrix. The Learning For Sure Program (Aprender Pra Valer) indicates the external evaluation on a large scale as the main agent of the management system, school management and the management of the classroom. However, this program does not have a specific project for the evaluator-teacher’s education evaluator and to confirm the importance of internal school evaluation. The docimologic paradigm allows us to identify the duties of the evaluator-teacher from the perspective of evaluation instruments. While the formation of the evaluator-teacher, based on educational paradigm, lies in their education in a complex frame of the systemic approach initially announced by Tyler (1977) and confirmed by Bloom, Hastings and Madaus (1983), Perrenoud (1999, 2000, 2002) and Hadji (1994,2001). / Diversas iniciativas de avaliação externa em larga escala surgiram no Brasil durante a década de 1990, a destacar: Sistema Nacional da Avaliação Básica (SAEB) e Exame Nacional do Ensino Médio (ENEM). Passados mais de vinte anos, desde o nascimento dessas iniciativas, intensificaram-se as discussões em torno da própria avaliação escolar interna realizada pelos professores. Aproximar a avaliação externa em larga escala da avaliação interna escolar coloca-se como um problema da atualidade no cenário educacional brasileiro. Nesse contexto, delimitamos o objeto desta pesquisa: o professor-avaliador. Definimos como objetivo geral deste trabalho discutir a identidade deontológica do professor-avaliador. Os objetivos específicos que orientaram este trabalho foram assim planificados: (1) discutir a identidade deontológica do professor-avaliador no contexto do paradigma docimológico da avaliação escolar, (2) discutir a identidade deontológica do professor-avaliador no contexto do paradigma pedagógico da avaliação escolar, (3) identificar o referencial de competências imputado ao professor-avaliador do programa Aprender pra Valer SEDUC-CE. Esta pesquisa se inscreve na abordagem qualitativa por manifestar, entre outras características, o caráter descritivo e interpretativo, o interesse pela utilização dos resultados, a busca por diversas fontes de dados. Os procedimentos metodológicos consubstanciaram-se em análise documental e pesquisa bibliográfica. Os dados foram interpretados pela análise de conteúdo. Para efeito deste estudo, como referenciais teóricos, demarcamos Bloom, Hastings e Madaus (1983), Perrenoud (1999), Afonso (2000), Hadji (1994, 2001), Tyler (1949), Bonniol e Vial (2001). Os resultados da pesquisa indicaram que a avaliação externa em larga escala exerce influência sobre identidade deontológica do professor-avaliador ao propor um referencial para a construção de instrumentos de avaliação: a Matriz de Referência. O Programa Aprender Pra Valer indica a avaliação externa em larga escala como principal agente da gestão do sistema, da gestão da escola e da gestão da sala de aula. No entanto, esse programa não apresenta um projeto específico para a formação do professor-avaliador nem confirma a importância da avaliação escolar interna. O paradigma docimológico nos permite identificar os deveres do professor-avaliador na perspectiva dos instrumentos de avaliação, enquanto a formação do professor-avaliador, baseada no paradigma pedagógico, situa a formação dele em um quadro complexo da abordagem sistêmica anunciada inicialmente por Tyler (1977) e confirmada por Bloom, Hastings e Madaus (1983), Perrenoud (1999, 2000, 2002) e Hadji (1994, 2001).

nÃo hà / iversas iniciativas de avaliaÃÃo externa em larga escala surgiram no Brasil durante a dÃcada de 1990, a destacar: Sistema Nacional da AvaliaÃÃo BÃsica (SAEB) e Exame Nacional do Ensino MÃdio (ENEM). Passados mais de vinte anos, desde o nascimento dessas iniciativas, intensificaram-se as discussÃes em torno da prÃpria avaliaÃÃo escolar interna realizada pelos professores. Aproximar a avaliaÃÃo externa em larga escala da avaliaÃÃo interna escolar coloca-se como um problema da atualidade no cenÃrio educacional brasileiro. Nesse contexto, delimitamos o objeto desta pesquisa: o professor-avaliador. Definimos como objetivo geral deste trabalho discutir a identidade deontolÃgica do professor-avaliador. Os objetivos especÃficos que orientaram este trabalho foram assim planificados: (1) discutir a identidade deontolÃgica do professor-avaliador no contexto do paradigma docimolÃgico da avaliaÃÃo escolar, (2) discutir a identidade deontolÃgica do professor-avaliador no contexto do paradigma pedagÃgico da avaliaÃÃo escolar, (3) identificar o referencial de competÃncias imputado ao professor-avaliador do programa Aprender pra Valer SEDUC-CE. Esta pesquisa se inscreve na abordagem qualitativa por manifestar, entre outras caracterÃsticas, o carÃter descritivo e interpretativo, o interesse pela utilizaÃÃo dos resultados, a busca por diversas fontes de dados. Os procedimentos metodolÃgicos consubstanciaram-se em anÃlise documental e pesquisa bibliogrÃfica. Os dados foram interpretados pela anÃlise de conteÃdo. Para efeito deste estudo, como referenciais teÃricos, demarcamos Bloom, Hastings e Madaus (1983), Perrenoud (1999), Afonso (2000), Hadji (1994, 2001), Tyler (1949), Bonniol e Vial (2001). Os resultados da pesquisa indicaram que a avaliaÃÃo externa em larga escala exerce influÃncia sobre identidade deontolÃgica do professor-avaliador ao propor um referencial para a construÃÃo de instrumentos de avaliaÃÃo: a Matriz de ReferÃncia. O Programa Aprender Pra Valer indica a avaliaÃÃo externa em larga escala como principal agente da gestÃo do sistema, da gestÃo da escola e da gestÃo da sala de aula. No entanto, esse programa nÃo apresenta um projeto especÃfico para a formaÃÃo do professor-avaliador nem confirma a importÃncia da avaliaÃÃo escolar interna. O paradigma docimolÃgico nos permite identificar os deveres do professor-avaliador na perspectiva dos instrumentos de avaliaÃÃo, enquanto a formaÃÃo do professor-avaliador, baseada no paradigma pedagÃgico, situa a formaÃÃo dele em um quadro complexo da abordagem sistÃmica anunciada inicialmente por Tyler (1977) e confirmada por Bloom, Hastings e Madaus (1983), Perrenoud (1999, 2000, 2002) e Hadji (1994, 2001). / Various initiatives to external large-scale evaluation arise in Brazil during the 1990s, especially: National System of Basic Assessment (SAEB) and National Secondary Education Examination (ENEM). After more than twenty years since the birth of those initiatives, discussions about the schools own internal assessment by teachers have been intensified. Bringing near the outer large-scale evaluation to the schoolâs internal evaluation arises as a problem today in the Brazilian educational scenario. In this context, we have delimited the object of this research: the evaluator-teacher. We have defined as general objective of this research the discussion on the ethical identity of evaluator-teacher. The specific objectives that guided this study were planned as follows: (1) to discuss the ethical identity of evaluator-teacher in the context of the paradigm docimolÃgico school evaluation, (2) to discuss the ethical identity of evaluator-teacher in the context of the pedagogical paradigm of school evaluation, (3) to identify the reference of powers attributed to the evaluator-teacher on Aprender pra Valer (Learning for Sure) program, by SEDUC-CE. This research falls under the qualitative approach by stating â among other features â the descriptive and interpretative matter, the interest on the use of the results, use various data sources. The methodological procedures substantiated in documentary analysis and literature review. The data were interpreted by content analysis. For purposes on this study we defined as theoretical references Bloom, Hastings and Madaus (1983), Perrenoud (1999), Afonso (2000), Hadji (1994.2001), Tyler (1949), Bonniol and Vial (2001). The survey results indicated that external evaluation on a large scale will influence on ethical identity of evaluator-teacher by proposing a framework for the construction of assessment instruments: the Reference Matrix. The Learning For Sure Program (Aprender Pra Valer) indicates the external evaluation on a large scale as the main agent of the management system, school management and the management of the classroom. However, this program does not have a specific project for the evaluator-teacherâs education evaluator and to confirm the importance of internal school evaluation. The docimologic paradigm allows us to identify the duties of the evaluator-teacher from the perspective of evaluation instruments. While the formation of the evaluator-teacher, based on educational paradigm, lies in their education in a complex frame of the systemic approach initially announced by Tyler (1977) and confirmed by Bloom, Hastings and Madaus (1983), Perrenoud (1999, 2000, 2002) and Hadji (1994.2001).

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-09-22T15:37:17Z No. of bitstreams: 1 carlamadureiravalentelopes.pdf: 4998902 bytes, checksum: 1ec5f575c14402542158b6454a900a2b (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-09-27T13:38:43Z (GMT) No. of bitstreams: 1 carlamadureiravalentelopes.pdf: 4998902 bytes, checksum: 1ec5f575c14402542158b6454a900a2b (MD5) / Made available in DSpace on 2017-09-27T13:38:44Z (GMT). No. of bitstreams: 1 carlamadureiravalentelopes.pdf: 4998902 bytes, checksum: 1ec5f575c14402542158b6454a900a2b (MD5) Previous issue date: 2016-12-15 / A presente dissertação é desenvolvida no âmbito do Mestrado Profissional em Gestão e Avaliação da Educação (PPGP) do Centro de Políticas Públicas e Avaliação da Educação da Universidade Federal de Juiz de Fora (CAEd/UFJF). O caso de gestão estudado focaliza um colégio de ensino médio da rede estadual do Rio de Janeiro, referência para a comunidade local, com bons resultados de aprovação, com monitoramento da frequência escolar, com planejamento de projetos pedagógicos e apoio da Universidade local, mas com desempenho abaixo da meta projetada pela Secretaria de Estado de Educação do Rio de Janeiro (SEEDUC-RJ) no Sistema de Avaliação do Estado do Rio de Janeiro (SAERJ) em Língua Portuguesa e Matemática ao final da Educação Básica, na série histórica acompanhada. Na dissertação são abordados os objetivos, limites e contribuições da avaliação externa frente ao fazer pedagógico e da avaliação interna aliada aos projetos interdisciplinares como instrumentos de favorecimento ao desempenho escolar. Os objetivos definidos para este estudo foram problematizar a avaliação externa como instrumento colaborativo e/ou impositivo ao processo ensino aprendizagem, e o desencontro entre o trabalho pedagógico desenvolvido pelo colégio e o que é esperado pela SEEDUC-RJ através do desempenho do colégio na avaliação externa do SAERJ. O aporte teórico da investigação realizada está apoiado em Soares (2002), Vieira (2007), Lück (2009), Soligo (2010), Bonamino e Sousa (2012), Alves e Soares (2013) e Crahay e Baye (2013) na busca de contribuições que possam melhorar a aprendizagem com base nos resultados da avaliação externa, sendo esse um instrumento de informações que apresentam dados importantes para a tomada de decisão no processo ensino aprendizagem. Para tanto, utilizamos como metodologia uma abordagem qualitativa e como instrumentos: entrevista semiestruturada com a equipe gestora e professores e, análise documental. Esta investigação confirmou a necessidade de reflexão sobre a utilização dos dados produzidos pela avaliação externa no âmbito escolar, sobre a apropriação de resultados como estímulo ao replanejamento pedagógico e da necessidade do investimento em formação contínua dos docente e da equipe gestora para a melhoria da qualidade educacional, envolvendo os alunos e responsáveis nas decisões pedagógicas. / The present dissertation is developed under the Professional Master in Management and Evaluation of Education (PPGP) of the Center for Public Policies and Education Evaluation of the Federal University of Juiz de Fora (CAEd / UFJF). The case of management has as central problem to identify why a high school of the state network of Rio de Janeiro, reference to the local community, with good results of approval, with monitoring of the school attendance, with planning of pedagogical projects and support of the Local University performs below the goal set by the State Department of Education of Rio de Janeiro (SEEDUC-RJ) in the Rio de Janeiro State Assessment System (SAERJ) in Portuguese Language and Mathematics at the end of Basic Education, in the historical series Accompanied. The dissertation addresses the objectives, limits and contributions of external evaluation in relation to pedagogical and as a counterpoint, the internal evaluation allied to interdisciplinary projects as instruments to favor school performance, highlighting the difference between the evaluation objectives of the mentioned instruments. The objectives defined for this study were to problematize the external evaluation as a collaborative and / or imposing instrument to the learning teaching process, and the mismatch between the pedagogical work developed by the college and what is expected by SEEDUC-RJ through the performance of the college in the external evaluation Of SAERJ. The theoretical contribution of the research is supported by Soares (2002), Vieira (2007), Lück (2009), Soligo (2010), Bonamino and Sousa (2012), Alves e Soares (2013) and Crahay and Baye Search for contributions that can improve learning based on the results of the external evaluation, which is an information tool that presents important data for decision making in the teaching learning process. For that, we used as a methodology a qualitative approach and as instruments: semi-structured interview with the management team and teachers, and documentary analysis. This research confirmed the need for reflection on the use of the data produced by the external evaluation in the school context, on the appropriation of results as a stimulus to the pedagogical replanning and of the necessity of the investment in continuous formation of the teacher and the management team for the improvement of the educational quality, involving students and decision makers.

Esta Tese resulta de pesquisa, cujo objeto foram as concepções de professores sobre avaliação educacional, com o objetivo geral de verificar possíveis influências das avaliações externas nessas concepções. Foi conduzida numa escola de ensino fundamental da Rede Municipal de Ensino de São Paulo (RME-SP) e teve como justificativa, entre outros motivos, a importância da avaliação no trabalho docente, sobretudo para os resultados escolares. Há na pesquisa, ainda,um traço marcante que é o fato de que a RME-SP, há vários anos, tem sido envolvida em muitas avaliações externas, e tal situação permitiu considerar, por hipótese, que elas estariam influenciando as concepções docentes sobre avaliação educacional que, por sua vez, estariam repercutindo no trabalho docente, considerando, ainda, as características do contexto escolar como fator adicional. Um grande desafio metodológico concentrou-se em como apreender tais concepções dadas as características de suas manifestações, mediante investigação desenvolvida por meio de pesquisa de campo, com entrevistas, aplicação de questionário, análise documental e observações de reuniões coletivas, envolvendo professores dos anos iniciais e finais do ensino fundamental. Nesse quadro, a questão de pesquisa foi: Quais seriam os elementos das concepções de avaliação educacional dos professores que estariam indicando a influência do conjunto de avaliações externas que perpassam a RMESP? Após termos desenvolvido trabalho de campo e analisado os dados levantados, efetuou-se síntese com os principais achados. Esperando contribuir para a ampliação do debate das influências das avaliações externas nas concepções de avaliação educacional dos professores, concluímos que existem marcas importantes dessas avaliações no ambiente escolar, mas sem sustentar que as avaliações externas estariam controlando o trabalho e as concepções docente no que tange à avaliação educacional. / This Thesis results from research, whose object were the conceptions of teachers about educational evaluation, with the general objective of verifying possible influences of external evaluations in these conceptions.It was conducted in a primary school of São Paulo Municipal Education Network (RME-SP) and had as justification, among other reasons, the importance of evaluation in teaching work, especially for school results.In the research there is still a landmark trace which is the fact that RME-SP, for several years, has been involved in many external evaluations, and such situation allowed to consider, by hypothesis, that the external evaluations would be influencing the educational conceptions about educational evaluation that, in turn, would be impacting on the educational work, considering also the characteristics of the school context as an additional factor.A great methodological challenge was focused on how to apprehend such conceptions given the characteristics of its manifestations, through research developed through field survey, with interviews, questionnaire application, documentary analysis and collective meetings observations, involving teachers from early and ending years of the elementary school. In this context, the research question was: What would be the elements of the teachers\' educational evaluation conceptions that would indicate the influence external evaluations that permeate RME-SP? After having the fieldwork developed and analyzing the collected data, we synthesized the main findings and concluded, hoping to contribute to the expansion of discussion of the influences of external evaluations on the conceptions of teachers educational evaluation, noting important marks of these evaluations in the school environment, but without sustaining that the external evaluations would be controlling the work and teacher conceptions with regards to educational evaluation.

Esta Tese resulta de pesquisa, cujo objeto foram as concepções de professores sobre avaliação educacional, com o objetivo geral de verificar possíveis influências das avaliações externas nessas concepções. Foi conduzida numa escola de ensino fundamental da Rede Municipal de Ensino de São Paulo (RME-SP) e teve como justificativa, entre outros motivos, a importância da avaliação no trabalho docente, sobretudo para os resultados escolares. Há na pesquisa, ainda,um traço marcante que é o fato de que a RME-SP, há vários anos, tem sido envolvida em muitas avaliações externas, e tal situação permitiu considerar, por hipótese, que elas estariam influenciando as concepções docentes sobre avaliação educacional que, por sua vez, estariam repercutindo no trabalho docente, considerando, ainda, as características do contexto escolar como fator adicional. Um grande desafio metodológico concentrou-se em como apreender tais concepções dadas as características de suas manifestações, mediante investigação desenvolvida por meio de pesquisa de campo, com entrevistas, aplicação de questionário, análise documental e observações de reuniões coletivas, envolvendo professores dos anos iniciais e finais do ensino fundamental. Nesse quadro, a questão de pesquisa foi: Quais seriam os elementos das concepções de avaliação educacional dos professores que estariam indicando a influência do conjunto de avaliações externas que perpassam a RMESP? Após termos desenvolvido trabalho de campo e analisado os dados levantados, efetuou-se síntese com os principais achados. Esperando contribuir para a ampliação do debate das influências das avaliações externas nas concepções de avaliação educacional dos professores, concluímos que existem marcas importantes dessas avaliações no ambiente escolar, mas sem sustentar que as avaliações externas estariam controlando o trabalho e as concepções docente no que tange à avaliação educacional. / This Thesis results from research, whose object were the conceptions of teachers about educational evaluation, with the general objective of verifying possible influences of external evaluations in these conceptions.It was conducted in a primary school of São Paulo Municipal Education Network (RME-SP) and had as justification, among other reasons, the importance of evaluation in teaching work, especially for school results.In the research there is still a landmark trace which is the fact that RME-SP, for several years, has been involved in many external evaluations, and such situation allowed to consider, by hypothesis, that the external evaluations would be influencing the educational conceptions about educational evaluation that, in turn, would be impacting on the educational work, considering also the characteristics of the school context as an additional factor.A great methodological challenge was focused on how to apprehend such conceptions given the characteristics of its manifestations, through research developed through field survey, with interviews, questionnaire application, documentary analysis and collective meetings observations, involving teachers from early and ending years of the elementary school. In this context, the research question was: What would be the elements of the teachers\' educational evaluation conceptions that would indicate the influence external evaluations that permeate RME-SP? After having the fieldwork developed and analyzing the collected data, we synthesized the main findings and concluded, hoping to contribute to the expansion of discussion of the influences of external evaluations on the conceptions of teachers educational evaluation, noting important marks of these evaluations in the school environment, but without sustaining that the external evaluations would be controlling the work and teacher conceptions with regards to educational evaluation.

The policy and practice of RPL (Recognition of Prior Learning) remains a contested area in the higher education sector. While a growing body of research on RPL has become available, little is known about the quality assurance dimensions of this policy and its current expression in higher education practice. Accordingly, this study seeks to provide a comprehensive and detailed portrait of the manner in which RPL is implemented in the Faculty of Education at the University of Pretoria. The central question is does the RPL system that is in place at this institution meet national and international requirements for quality and quality assurance? If not, what are the reasons and how can the faculty improve its RPL practice? The research sub-questions addressed are the following: <ul> <li> What is the quality of the inputs used to design the RPL that is in place in the Faculty of Education at the University of Pretoria?</li> <li> How does the Faculty of Education at the University of Pretoria assess RPL candidates for their prior learning? </li> <li> What is the effect of the output of the RPL system on client satisfaction?</li></ul> A mixed methods research design was used for this study. A single Faculty (Education) was selected as the data collection site, to reveal the deeper and nuanced impact of the process of implementation of the RPL programme. A semi-structured interview schedule administered to the senior managers of the faculty was to elicit information on how the RPL system was conceptualised and designed. This process included the Dean (Faculty of Education); Head of Department (Curriculum Studies); Head of Department (Educational Management, Law and Policy Studies); Director (Centre for Evaluation and Assessment) and the Director (Centre for Joint Science, Mathematics and Technology Education). To determine whether there is a link between what the Quality Assurance Unit of the university promotes and application of such principles and procedures at service delivery level, an interview with the Director of the QA Unit was done. Other interviews involved students (undergraduates and postgraduates); the non-academic staff and lecturers within all the departments of the faculty, to determine whether they knew or were aware of RPL related activities in the faculty. An observation tool was constructed to examine the quality of the assessment process, which involved RPL learners, assessors, evidence facilitators, verifiers, moderators and RPL administrators. A questionnaire was administered to RPL learners involved in the assessment process to determine their satisfaction with the output of the RPL programme. Lecturers who participated in the RPL assessment process were interviewed to determine their experiences. Finally, an observational checklist was used to determine quality indicators at macro (administrative) and micro (academic) levels. The data was analysed using pattern matching, discrepancy, content and interpretational analyses methods. The research findings presented are in the form of a “thick” narrative on the quality of RPL implementation, that is, what the faculty should do to improve or strengthen the current system, and a portrayal of how the RPL programme truly operates. The findings indicate that a relatively good system of RPL provisioning is in place in the Faculty of Education, with a few areas of concern (weaknesses). The major problem is that this system is not benefiting the majority of people it was intended for. The system is “selective” and “exclusionary” in nature. There are clear procedures and processes for RPL assessment, which are adhered to strictly by faculty assessors. The RPL system that is currently in place is satisfactory to those who were assessed for prior learning during the period 2003-2006 and unsatisfactory at the level of the lecturers who participated in the assessment process. Most of them indicated that RPL is an add-on activity to their workloads, with very little incentives from management. To those who were not part of the assessment process, but were assumed to have received information from the faculty, the findings indicated that they knew very little about RPL and how it is being assessed in the faculty. From the client’s perspective, most (eighty four percent) said if they knew how this system operates in the faculty, they would want to be assessed for their prior learning. An extensive examination of the RPL practice in the Faculty of Education gave useful insights on the quality of RPL provisioning. Future research needs to concentrate on evaluations on how RPL is implemented in the other faculties of the university. Second to this, is to begin to provide answers as to what causes full-scale implementation of RPL problematic in the higher education sector, to provide empirical data to policy makers for decision-making purposes. Thirdly, to provide solutions towards the sustainability of the RPL system in the higher education sector, there is a need to do studies on the cost-effectiveness of RPL implementation. / Thesis (PhD (Assessment and Quality Assurance in Education and Training))--University of Pretoria, 2007. / Curriculum Studies / PhD / unrestricted