Numerical information is ubiquitous and people are continuously engaged in evaluating it by means of intuitive statistical judgments. Much research has evaluated if people’s judgments live up to the norms of statistical theory but directed far less attention to the cognitive processes that underlie the judgments. The present thesis outlines, compares, and tests two cognitive models for intuitive statistical judgments, summarized in the metaphors of the lazy and eager intuitive statistician. In short, the lazy statistician postpones judgments to the time of a query when the properties of a small sample of values retrieved from memory serve as proxies for population properties. In contrast, the eager statistician abstracts summary representations of population properties online from incoming data. Four empirical studies were conducted. Study I outlined the two models and investigated whether an eager or a lazy statistician best describes how people make intuitive statistical judgments. In general the results supported the notion that people spontaneously engage in a lazy process. Under certain specific conditions, however, participants were able to induce abstract representations of the experienced data. Study II and Study III extended the models to describe naive point estimates (Study II) and inference about a generating distribution (Study III). The results indicated that both the former and the latter type of judgment was better described by a lazy than an eager model. Finally, Study IV, building on the support in Studies I-III, investigated boundary conditions for a lazy model by exploring if statistical judgments are influenced by common memory effects (primacy and recency). The results indicated no such effects, suggesting that the sampling from long-term memory in a lazy process is not conditional on when the data is encountered. The present thesis makes two major contributions. First, the lazy and eager models are first attempts at outlining a process model that could possibly be applied for a large variety of statistical judgments. Second, because a lazy process imposes boundary conditions on the accuracy of statistical judgments, the results suggest that the limitations of a lazy intuitive statistician would need to be taken into consideration in a variety of situations.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The economy globalization forced the companies to invest in other forms of management and in technologies that intensify its competitiveness on the market and provide an increasing in the quality and in the cost/benefit of their products and services. In this context, the purpose of this research consists of creating an interactive site about Statistical Process Control (SPC), that could help companies to have theoretical-practical notions of the operation of this tool and of its application. In order to get that, the tools of Statistical Process Control will be used, helping companies to forecast the output trials and, at the same time, increase the quality of its products and services, avoiding scraps and redo. In this way, the concepts and the software used in the analysis, construction and implementation of the interactive site are described. Subsequently, the created tool has been tested and evaluated, by mean a cases. Results obtained revealed that it is efficient, because the graphics generated shown the user a real and concrete vision based on the supplied information for the CEP on-line program. Finally, based on the results, a deeper study regarding some limitations presented by the tool must continue, like the size of the sample and review calculus of limits, which are not performed by the program. / Com a crescente globalização da economia e abertura de novos mercados, as empresas estão sendo impulsionadas a investir em outras formas de gestão e em tecnologias que intensifiquem a sua competitividade no mercado e proporcionem um aumento na qualidade e no custo/benefício de seus produtos e serviços. Neste contexto, o objetivo deste trabalho consiste em criar um site1 interativo sobre o Controle Estatístico do Processo (CEP), o qual possa contribuir para que as empresas tenham noções teórico-práticas do funcionamento dessa ferramenta e de sua aplicação. Para tanto, serão utilizadas as ferramentas do Controle Estatístico do Processo (CEP), visando auxiliar as empresas a tornar previsíveis as saídas de seus processos de produção e, ao mesmo tempo, aumentar a qualidade de seus produtos e serviços, evitando o refugo e o retrabalho. Dessa forma, descrevem-se os conceitos e os softwares2 utilizados, na análise, construção e implementação do site interativo. Posteriormente, o aplicativo criado foi testado e avaliado, utilizando-se para isso amostras de estudos de casos. Os resultados obtidos revelaram que é eficiente, uma vez que os gráficos gerados permitem ao usuário uma visão real e concreta baseada nas informações fornecidas ao aplicativo CEP on-line . Finalmente, com base nos resultados, deve-se proceder a um estudo mais aprofundado com relação a algumas limitações que o aplicativo apresenta, tais como o tamanho da amostra, pois se restringe a amostras somente de tamanho constante, e os cálculos dos limites revisados, os quais não são efetuados pelo aplicativo.

The research presented in this dissertation touches on all aspects of survey methodology, from questionnaire design to final estimation. We first approach the questionnaire development stage by proposing a method of developing matrix sampling designs, a design where a subset of questions are administered to a respondent in such a way that the administered questions are predictive of the omitted questions. The proposed methodology compares favorably to previous methods when applied to data collected from a household survey conducted in the Nampula province of Mozambique. We approach the data collection stage by proposing a structured procedure of implementing small-scale surveys in such a way that non-sampling error attributed to data collection is minimized. This proposed methodology requires the inclusion of the statistician in the data editing process during data collection. We implemented the structured procedure during the collection of household survey data in the city of Maputo, the capital of Mozambique. We found indications that the data resulting from the structured procedure is of higher quality than the data with no editing. Finally, we approach the estimation phase of sample surveys by proposing a model-based approach to the estimation of the mean squared error associated with synthetic (indirect) estimates. Previous methodology aggregates estimates for stability, while our proposed methodology allows area-specific estimates. We applied the proposed mean squared error estimation methodology and methods found during literature review to simulated data and estimates from 2010 Census Coverage Measurement (CCM). We found that our proposed mean squared error estimation methodology compares favorably to the previous methods, while allowing for area-specific estimates. / Ph. D.

在統計學上我們經常會遭遇到如下的問題： minimze subject to 其中 和Ｃ都是已知。 上述非線性規劃（NONLINEAR PROGRAMING）問題的最佳解，是相當複雜的，以致於我 們無法用簡單的式子，將其解明確的表示出來。 RAO-GHANGURDE （１９７２）在“從有限母體抽樣的貝氏最佳解”這一篇文章中，對 這種非線性規劃問題，提出一個反覆演算的解法，來解決這類問題，由於，我們無法 看出其演算法的立論根據何在，收斂結果的精確性有多高，於是，本文在ｋ＝２及ｋ ＝３的情形下，由直覺的幾何觀點，提出了另一個求最佳解的方法，來驗證RAO-GHAN GURED 反覆演算法的類確性。 最後，本論文將上述非線性規劃問題的解法，應用到下面兩個例子上： （ａ）在 COCHRAN的“抽樣技巧”（ SAMPLING TECHNIQUES）這一本書裡，有關雙重 抽樣（DOUBLE SAMPLING ）的理論中，也遭遇到要解決這一類問題，但由他的公式， 所計算出來的解，並不一定會萬足所需要的限制條件。 （ｂ）在SMITH-SEDRASK （１９８２）的“推估魚群年齡成份的貝氏最佳解“和JINN -SMITH-SEDRASK（１９８７）的“推估魚群年齡成份的貝氏最佳雙重抽樣”這兩篇的 文章中，同樣的也遭遇到這一類的問題。

Interest in effectively publishing academic articles stems from involvement in statistical research support provided to academic researchers conducting their research. In the context of this study research output (RO) is defined as the publication of research findings (articles) in academic journals accredited with the South African Department of Higher Education and Training’s (DHET). The vantage point of this research is that of research support statisticians. New knowledge is continually required to drive decision making, policy formulation, industry, economies, regulation, development, innovation and progress (SESCES 2015:9; Pullinger 2014). Quality published research serves as a reliable source of new information. Therefore measures are globally and nationally implemented to stimulate article publication. Such measures and incentives include measurement of publication rate; journal impact ratings; government funding of research based on research output; acknowledgement as research-intensive institutions, promotion opportunities linked to publication rate and more. Although the literature reports on aspects of the production and publication of research findings, limited research is reported on research output productivity (ROP) viewed from the perspective of the statistical community that support research within the research process. Therefore a theoretical framework for ROP had to be developed. Classic grounded theory (GT) proved to be an appropriate methodology for this research based on its theory-develop properties. The literature, responses to an open- and closed-ended questionnaire, observational field notes of this researcher and informal discussion notes were inter alia used as data bases in the cycles of data-collection-analysis-and-comparison that characterise GT implementation. Theoretical components (‘categories’) that emerged in the research include the research process as central concept (the ‘core category’), a research practice component; role players in the research process; the attitude of researchers; knowledge of researchers; skills and attributes of researchers; research resources and research resource centres; and the research climate of the researcher environment. These components constitute the factors that impact ROP. Relational links - which forms the second leg of a developing theory - between these components are explained quantitatively in terms of multivariate linear regression equations; a profile of researcher-type (discriminant analysis) and qualitatively by means of the literature and field notes of this researcher. The emerged theoretical model indicates that knowledge and skills of academic researchers, as well as researcher-type directly impact on the research process and therefore on ROP. Furthermore attitude forms a discriminatory attribute of academic researchers. The objective with the development of the model of ROP was to identify important components of RO delivery and propose grassroots recommendations to promote ROP. / Curriculum and Instructional Studies / D. Ed. (Didactics)

Interest in effectively publishing academic articles stems from involvement in statistical research support provided to academic researchers conducting their research. In the context of this study research output (RO) is defined as the publication of research findings (articles) in academic journals accredited with the South African Department of Higher Education and Training’s (DHET). The vantage point of this research is that of research support statisticians. New knowledge is continually required to drive decision making, policy formulation, industry, economies, regulation, development, innovation and progress (SESCES 2015:9; Pullinger 2014). Quality published research serves as a reliable source of new information. Therefore measures are globally and nationally implemented to stimulate article publication. Such measures and incentives include measurement of publication rate; journal impact ratings; government funding of research based on research output; acknowledgement as research-intensive institutions, promotion opportunities linked to publication rate and more. Although the literature reports on aspects of the production and publication of research findings, limited research is reported on research output productivity (ROP) viewed from the perspective of the statistical community that support research within the research process. Therefore a theoretical framework for ROP had to be developed. Classic grounded theory (GT) proved to be an appropriate methodology for this research based on its theory-develop properties. The literature, responses to an open- and closed-ended questionnaire, observational field notes of this researcher and informal discussion notes were inter alia used as data bases in the cycles of data-collection-analysis-and-comparison that characterise GT implementation. Theoretical components (‘categories’) that emerged in the research include the research process as central concept (the ‘core category’), a research practice component; role players in the research process; the attitude of researchers; knowledge of researchers; skills and attributes of researchers; research resources and research resource centres; and the research climate of the researcher environment. These components constitute the factors that impact ROP. Relational links - which forms the second leg of a developing theory - between these components are explained quantitatively in terms of multivariate linear regression equations; a profile of researcher-type (discriminant analysis) and qualitatively by means of the literature and field notes of this researcher. The emerged theoretical model indicates that knowledge and skills of academic researchers, as well as researcher-type directly impact on the research process and therefore on ROP. Furthermore attitude forms a discriminatory attribute of academic researchers. The objective with the development of the model of ROP was to identify important components of RO delivery and propose grassroots recommendations to promote ROP. / Curriculum and Instructional Studies / D. Ed. (Didactics)