Typewritten sheets in cover. Thesis (M.A.)--Boston University This item was digitized by the Internet Archive. Bibliography: 10 p. at end.

Ensinar é sem dúvida um dos grandes desafios do século XXI. A busca por alternativas ao tradicional sistema de ensino mostra a necessidade de mudanças, que é inerente à sociedade do conhecimento em contínua evolução. Desta forma, esta pesquisa pretende investigar se as ações desenvolvidas na Escola E. E. Arnolfo Azevedo pelo projeto Laboratório de Ciências e sua interação com a Universidade por meio do programa de Préiniciação Científica contribuem para a construção e manutenção de um modelo que busca a melhoria da qualidade do ensino e aprendizagem em ciências exatas e biológicas nas Escolas Públicas. Foram utilizados questionários qualitativos e quantitativos para construção dos gráficos em forma de barras e linhas de tendência, associando estes dados ao referencial proposto por meio da triangulação. Assim, os resultados indicam que a participação dos alunos nos respectivos projetos: Laboratório e a Pré-iniciação científica, promovem significativa mudança de comportamento associado à dedicação e responsabilidade com os estudos representada pela melhora significativa em seu rendimento escolar e conseqüente melhora na qualidade do ensino na área de ciências naturais. / Teaching is undoubtedly one of the greatest challenges of the 21st century. The search for alternatives to the traditional teaching system shows the need for changing, which is inherent to knowledge society in continuous development. Therefore, this research deals with the investigation whether the actions carried out in the Project of the Laboratory of Sciences at E.E. Arnolfo Azevedo and its interaction with the University through Scientific Pre-Initiation Program have contributed to the construction and maintenance of a model which seeks to improve the teaching and learning quality in Exact and Biological Sciences in Public Schools. Qualitative and quantitative surveys were used to build graphs of trend lines and error bars, associating these data with the proposed framework through triangulation. The results indicate that the students\' participation in the respective projects (Laboratory and Scientific Pre-Initiation) has promoted meaningful behavior change associated with studies dedication and responsibility represented by the significant improvement in their school performance and, consequently, in the teaching quality in Natural Sciences area.

acase@tulane.edu

acase@tulane.edu

acase@tulane.edu

acase@tulane.edu

acase@tulane.edu

This study examined the geographic variation of infant mortality using zip codes as the unit of analysis in order to facilitate site location and site specific health services for community health centers and/or mobile health units. The relationship between infant mortality and the place of residence by zip code was analyzed using factors related to residential deprivation and selected maternal-infant factors. The design of this study was non-experimental retrospective correlational using a mix of ecological and individual variables. The sample chosen included all births and associated infant deaths from 1988 through 1992 in 22 of the 64 parishes in Louisiana. The 22 sample parishes contained 192 zip codes. Variable data was extracted from linked birth and death files and 1990 Census tapes. Chi-square was used to test variation of infant mortality among and within parishes by zip codes. Correlation and multivariate regression analysis were used to examine the association of six maternal/infant factors and five residential deprivation factors, separately and then combined, with infant mortality. A geographic information system was also used to formulate thematic maps for presentation of data. Results indicated that the mother's marital status was the most significant predictor of infant mortality; being married was associated with decreased infant mortality. Identification of significant predictors varied on an individual basis. The results also indicate that infant mortality does vary significantly within each parish. Overall use of this model in order to locate healthcare services or in predicting infant mortality by zip code is questionable, but may be useful in those parishes that have large populations and exhibit greater internal variability and segregation in general / acase@tulane.edu

acase@tulane.edu

acase@tulane.edu