The increased use of polytomous item formats has led assessment developers to pay greater attention to the detection of differential item functioning (DIF) in these items. DIF occurs when an item performs differently for two contrasting groups of respondents (e.g., males versus females) after controlling for differences in the abilities of the groups. Determining whether the difference in performance on an item between two demographic groups is due to between group differences in ability or some form of unfairness in the item is a more complex task for a polytomous item, because of its many score categories, than for a dichotomous item. Effective DIF detection methods must be able to locate DIF within each of these various score categories. The Mantel, Generalized Mantel Haenszel (GMH), and Logistic Regression (LR) are three of several DIF detection methods that are able to test for DIF in polytomous items. There have been relatively few studies on the effectiveness of polytomous procedures to detect DIF; and of those studies, only a very small percentage have examined the efficiency of the Mantel, GMH, and LR procedures when item discrimination magnitudes and category intersection parameters vary and when there are different patterns of DIF (e.g., balanced versus constant) within score categories. This Monte Carlo simulation study compared the Type I error and power of the Mantel, GMH, and OLR (LR method for ordinal data) procedures when variation occurred in 1) the item discrimination parameters, 2) category intersection parameters, 3) DIF patterns within score categories, and 4) the average latent traits between the reference and focal groups. Results of this investigation showed that high item discrimination levels were directly related to increased DIF detection rates. The location of the difficulty parameters was also found to have a direct effect on DIF detection rates. Additionally, depending on item difficulty, DIF magnitudes and patterns within score categories were found to impact DIF detection rates and finally, DIF detection power increased as DIF magnitudes became larger. The GMH outperformed the Mantel and OLR and is recommended for use with polytomous data when the item discrimination varies across items.

No description available.

No description available.

Educational measurement research faces an unresolved dilemma: competently meeting the longstanding demand for improved vocabulary strength (depth) aspect assessments. My original contribution to knowledge in the written receptive vocabulary knowledge construct research domain is twofold. My first contribution is proposing an a priori metasynonymy awareness hypothesis based on a vocabulary strength aspect extension of O’Connor’s (1940) written receptive vocabulary acquisition developmental stage theory. My second contribution is designing and constructing a vocabulary extent (the nexus between vocabulary size (breadth) and strength aspects) test. The test, called the Polysemous Academic Vocabulary Extent Test, utilizes ordered triple rank (OTR) responses and a complementary six-tier incremental scoring guide rubric. An example test item includes a sentence stem with a bold keyword and three options, such as: All the reviews of the movie were positive. positive: a) sure b) good c) enviro / Language Arts

De nos jours, la quantité de données génétiques séquencées augmente de manière exponentielle sous l'impulsion d'outils de séquençage de plus en plus performants, tels que les outils de séquençage haut débit en particulier. De plus, ces données sont de plus en plus facilement accessibles grâce aux bases de données en ligne. Cette plus grande disponibilité des données ouvre de nouveaux sujets d'étude qui nécessitent de la part des statisticiens et bio-informaticiens de développer des outils adaptés. Par ailleurs, les progrès constants de la statistique, dans des domaines tels que le clustering, la réduction de dimension, ou les régressions entre autres, nécessitent d'être régulièrement adaptés au contexte de la bio-informatique. L’objectif de cette thèse est l’application de techniques avancées de statistiques à des problématiques de bio-informatique. Dans ce manuscrit, nous présentons les résultats de nos travaux concernant le clustering de séquences génétiques via Laplacian eigenmaps et modèle de mélange gaussien, l'étude de la propagation des éléments transposables dans le génome via un processus de branchement, l'analyse de données métagénomiques en écologie via des courbes ROC ou encore la régression polytomique ordonnée pénalisée par la norme l1. / Nowadays, the quantity of sequenced genetic data is increasing exponentially under the impetus of increasingly powerful sequencing tools, such as high-throughput sequencing tools in particular. In addition, these data are increasingly accessible through online databases. This greater availability of data opens up new areas of study that require statisticians and bioinformaticians to develop appropriate tools. In addition, constant statistical progress in areas such as clustering, dimensionality reduction, regressions and others needs to be regularly adapted to the context of bioinformatics. The objective of this thesis is the application of advanced statistical techniques to bioinformatics issues. In this manuscript we present the results of our works concerning the clustering of genetic sequences via Laplacian eigenmaps and Gaussian mixture model, the study of the propagation of transposable elements in the genome via a branching process, the analysis of metagenomic data in ecology via ROC curves or the ordinal polytomous regression penalized by the l1-norm.

return and payback period. All these above techniques will be analysed in three different scenarios, namely: 1. Mine X will stay with its current operations without any new projects. 2. The development project will begin immediately. 3. A six–month delay in development of the project. The study found that the net present value was positive, the internal rate of return was more than the discount rate and the payback period was shorter than the project’s life–time regarding to all three above–mentioned scenarios. The highest net present value is calculated in case the project starts immediately. Both the internal rate of return and the payback period indicated that a six month delay in the project is the most viable. After considering all the facts, the study concluded due to the highest net present value the best feasible recommendation would be to start the project immediately. The value of this study is that it is the first study to investigate the relationship between the viability to delay or to start the investment project immediately in the South African mining industry. This study is also unique, since it takes into account how mining industries world–wide can achieve long–term success through development projects without losing key players, due to impulsive short–term downsizing decisions. To be able to use personality tests in the most reliable and valid manner there are many considerations to be taken into account. Variables such as the population used, the culture of the test–takers, the mode of administration, whether pencil–and–paper or computer–based testing procedures, familiarity with computers when using computer–based tests and the response format to be used when administering the personality questionnaire are but some of the considerations. Within South Africa it is that much more important to consider the mode of administration, whether pencil–and–paper tests or computer–based tests, as there are many individual groups who have been historically disadvantaged when it comes to the use of computers as a testing method. It is just as important to consider the response scale to be utilised when administering personality testing as this may influence the results obtained and can influence the reliability and validity of these results. The objective of this study was to determine which response scale, dichotomous or polytomous, was the best to use when conducting computer–based personality testing. The questionnaire that was utilised was the South African Personality Inventory (SAPI) questionnaire; however, only items from the Soft–Heartedness cluster were employed as the objective was not to test the questionnaire but to test the most reliable and valid response scale to be used in conjunction with the questionnaire. A convenience sampling approach was utilised and the questionnaire was administered to students who were available and able to take the test (N = 724). Descriptive statistics, factor analysis and Cronbach Alpha coefficients were used to analyse the data obtained. / Thesis (M.Com. (Industrial Psychology))--North-West University, Potchefstroom Campus, 2012.

return and payback period. All these above techniques will be analysed in three different scenarios, namely: 1. Mine X will stay with its current operations without any new projects. 2. The development project will begin immediately. 3. A six–month delay in development of the project. The study found that the net present value was positive, the internal rate of return was more than the discount rate and the payback period was shorter than the project’s life–time regarding to all three above–mentioned scenarios. The highest net present value is calculated in case the project starts immediately. Both the internal rate of return and the payback period indicated that a six month delay in the project is the most viable. After considering all the facts, the study concluded due to the highest net present value the best feasible recommendation would be to start the project immediately. The value of this study is that it is the first study to investigate the relationship between the viability to delay or to start the investment project immediately in the South African mining industry. This study is also unique, since it takes into account how mining industries world–wide can achieve long–term success through development projects without losing key players, due to impulsive short–term downsizing decisions. To be able to use personality tests in the most reliable and valid manner there are many considerations to be taken into account. Variables such as the population used, the culture of the test–takers, the mode of administration, whether pencil–and–paper or computer–based testing procedures, familiarity with computers when using computer–based tests and the response format to be used when administering the personality questionnaire are but some of the considerations. Within South Africa it is that much more important to consider the mode of administration, whether pencil–and–paper tests or computer–based tests, as there are many individual groups who have been historically disadvantaged when it comes to the use of computers as a testing method. It is just as important to consider the response scale to be utilised when administering personality testing as this may influence the results obtained and can influence the reliability and validity of these results. The objective of this study was to determine which response scale, dichotomous or polytomous, was the best to use when conducting computer–based personality testing. The questionnaire that was utilised was the South African Personality Inventory (SAPI) questionnaire; however, only items from the Soft–Heartedness cluster were employed as the objective was not to test the questionnaire but to test the most reliable and valid response scale to be used in conjunction with the questionnaire. A convenience sampling approach was utilised and the questionnaire was administered to students who were available and able to take the test (N = 724). Descriptive statistics, factor analysis and Cronbach Alpha coefficients were used to analyse the data obtained. / Thesis (M.Com. (Industrial Psychology))--North-West University, Potchefstroom Campus, 2012.

Made available in DSpace on 2015-07-22T20:50:38Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-03-25 / Por ser uma prática criminosa, o aborto provocado acaba sendo realizado clandestinamente tornando-se um grave problema de Saúde Pública. O objetivo deste estudo foi o de estimar a prevalência de mulheres em idade fértil com histórico de aborto. O estudo é transversal, resultante de uma amostra aleatória de mulheres 15 a 49 anos-, residentes no subdistrito da Vila Mariana, 2006. Os dados foram coletados mediante aplicação de questionários. Foi considerada como variável dependente classificação da mulher quanto ao aborto: sem aborto, aborto espontâneo e aborto provocado; e independentes: idade, defasagem do número ideal de filhos, atividade remunerada, escolaridade, estado civil, uso de contraceptivos e opinião sobre o aborto provocado. Para análises foram utilizados testes de qui-quadrado e modelos de regressão logística multinomial policotômica. Dentre o total de mulheres entrevistadas (n=1121), 84,4% (n=946) são de mulheres sem histórico de aborto; 11,2% (n=126) são de mulheres com histórico de aborto espontâneo e, 4,4% (n=49) são de mulheres com histórico de aborto provocado. A razão de chances de ter realizado aborto provocado sobre a sem aborto é 6,33 vezes maior (p0,001) entre mulheres que aceitam esta prática; 4,58 vezes maior (p=0,002) entre as mulheres que possuem menos de 4 anos de estudo e ainda, as chances da mulher declarar um aborto provocado comparado às sem aborto é 7% maior a cada ano em que a mulher envelhece. Dentre as 1121 mulheres, 49,5% (n=555) declararam ter tido alguma gravidez. Para que engravidaram a prevalência de mulheres com aborto espontâneo foi de 22,7% (n=126) de aborto provocado 8,85 (n=49). A razão de chances de ter realizado aborto provocado sobre a sem aborto é 28,34 vezes maior (p0,001) entre as que não possuem nenhum filho nascido vivo; 6,42 vezes maior (p0,001) entre as que aceitam esta prática; 4,96 vezes maior (p=0,002) entre as que possuem menos de 4 anos de estudo; e as chances de declarar um aborto provocado comparado as sem aborto é 8% maior a cada ano a mais de vida. Por outra parte, este estudo revela ainda que entre o total de mulheres a razão de chances de ter tido aborto espontâneo sobre a sem aborto é 0,34 (p0,001) entre as mulheres que não possuem nenhum filho nascido vivo; e, as chances da mulher declarar um aborto espontâneo comparado às sem aborto é 4% maior a cada ano de idade da mulher. O comportamento reprodutivo das mulheres deste estudo é equiparável ao das residentes em países desenvolvidos. Ao ter acesso a métodos contraceptivos considerados eficazes o aborto provocado legalizado, não seria utilizado de forma irresponsável. / Induced abortions are illegal in Brazil, leading many women to seek out clandestine clinics and practitioners, resulting in a serious public health problem. The purpose of this study was to estimate the number of women in the general population of fertile age with a history of abortion. This is a retrospective transversal study, based on a random sample of women – 15 to 49 years old –, residing at the Vila Mariana neighborhood of São Paulo in 2006. Data was collected through questionnaires. As the dependent variable we used different abortion categories, reflecting different types of experiences with abortion, which included: no abortion, spontaneous abortion and induced abortion. As independent variables we used: age, the difference between number of children and ideal number of children, employment and marital status, level of education, use of contraceptives, and personal opinion about induced abortion. Analyses were carried out using chi-square tests and polytomous multinomial logistic regressions. Furthermore, 84,4% (n=946) had no history of abortion; 11,2% (n=126) indicated having had a spontaneous abortion; and 4.4% (n=49) indicated having had an induced abortion. We found that it is 6,33 times more likely (p0,001) to have had an induced abortion versus no abortion among women who are pro-choice; 4,58 times more likely (p=0,002) among women who have less than 4 years of formal education; and the chances of a woman admitting an induced abortion compared to no abortion are 7% higher for each additional year of age. We surveyed a total of 1121 women, among which 49.5% (n=555) indicated that they had been pregnant at least once. Among the latter, 22,7% (n=126) indicated having at least one spontaneous abortion and 8,85% (n=49) indicated having at least one induced abortion. Our results show that among women with no live birth pregnancies it is 28,34 times more likely that they have undergone induced abortion versus no abortion (p0,005); among those that are pro-choice it is 6,42 times more likely (p0,001); among those who have less than 4 years of formal education it is 4,96 times more likely (p=0,002); and the chances of admitting to an induced abortion versus no abortion increases by 8% higher for each additional year of age. Finally, this study reveals that women with no live births are 0,34 more likely (p0,001) to have had an spontaneous abortion versus no abortion; and the chances of a woman admitting spontaneous abortion compared to no abortion is 4% higher for each additional year of age. In conclusion, the reproductive behavior of women in this study is comparable to the behavior of women who live in developed countries. With broad access to effective contraceptive methods, legalized induced abortion would not be carried out irresponsibly. / TEDE

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Social demands on soil information have grown dramatically, meanwhile the soil surveys are seldom carried out in the country. Digital soil mapping techniques can be applied to infer the spatial distribution of soil from existing soil maps or from reference areas, extrapolating this information to areas not mapped. The purpose of this study was to apply in a Geographic Information System the Multiple Logistic Regressions (MLR) using Principal Components (PC) as explanatory variables to predict soil classes spatial distribution. The study area was the region of municipality São Pedro do Sul / RS. For the development of predictive models a set of nine terrain attributes were used. Model training was executed on an existing soil map and with a survey carried out in a reference area, both in a 1:50.000 scale. The first three retained PC explained 65.57% of the data variability. The predictive models which used PC had lower values of kappa index. The most accurate predicted map reached a kappa value of 63.20% and was generated by using the nine attributes of land as predictive covariates. The mapping accuracy is sensitive to similarities between the mapped classes, and mapping in a more homogeneous categorical level reduces the accuracy of the predicted maps. Soil classes relatively not representative in the training maps are not properly spatialized. The use of MLR allows spatializing of soil classes to areas not mapped, although the use of PC needs to be tested with a larger number of covariates. / As demandas da sociedade pela informação solo têm crescido, porém levantamentos pedológicos praticamente não ocorrem mais no país. Técnicas de Mapeamento Digital do Solo podem ser empregadas para inferir a distribuição espacial de classes de solos a partir de mapas existentes e áreas de referência, extrapolando esta informação para áreas não mapeadas. O objetivo deste estudo foi empregar em um Sistema de Informações Geográficas as Regressões Logísticas Múltiplas (RLM) utilizando-se de Componentes Principais (CP) como variáveis explicativas para a predição espacial de classes de solos. A área de estudo foi na região do município de São Pedro do Sul / RS. Para o desenvolvimento dos modelos preditivos foram utilizados um conjunto de nove atributos do terreno. O treinamento dos modelos foi executado em um mapa de solos existente, e em um levantamento realizado em áreas de referência, ambos na escala 1:50.000. As três primeiras CP retidas explicaram 65,57% da variabilidade dos dados. Os modelos preditivos que empregaram CP obtiveram menores valores do índice kappa. O mapa predito mais acurado empregou os nove atributos do terreno e alcançou um valor de kappa de 63,20%. A acurácia do mapeamento é sensível a semelhança entre as classes mapeadas, e o mapeamento em níveis categóricos mais homogêneos reduz a precisão dos mapas preditos. Classes de solos relativamente pouco representativas não são corretamente espacializadas. O emprego de RLM permite espacializar classes de solos para áreas não mapeadas, embora o emprego de CP necessite ser testado com um maior número de covariáveis.

Orientador: Caio Lucidius Naberezny Azevedo / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-24T12:51:18Z (GMT). No. of bitstreams: 1 Ferreira_EduardoVargas_M.pdf: 8131052 bytes, checksum: f344cd1f11e8d818f3aac90f48396cbc (MD5) Previous issue date: 2014 / Resumo: No presente trabalho propõem-se novos modelos da Teoria de Resposta ao Item para respostas politômicas nominais e ordinais (graduais), via dados aumentados, para grupos múltiplos. Para a modelagem das distribuições dos traços latentes de cada grupo, considera-se normais assimétricas centradas. Tal abordagem, além de acomodar a característica de assimetria aos dados, ajuda a garantir a identificabilidade dos modelos estudados, a qual é tratada tanto sob a ótica frequentista quanto bayesiana. Com relação aos métodos de estimação, desenvolveu-se procedimentos bayesianos através de algoritmos de Monte Carlo via cadeias de Markov (MCMC), utilizando o algoritmo de Gibbs (DAGS), com a verossimilhança aumentada (dados aumentados) e Metropolis-Hastings, considerando a verossimilhança original. As implementações computacionais foram escritas em linguagem C++, integradas ao ambiente computacional, gráfico e estatístico R, viabilizando rotinas gratuitas, de código aberto e alta velocidade no processamento, essenciais à difusão de tais metodologias. Para a seleção de modelos, utilizou-se o critério de informação deviance (DIC), os valores esperados do critério de informação de Akaike (EAIC) e o critério de informação bayesiano (EBIC). Em relação à verificação da qualidade do ajuste de modelos, explorou-se a checagem preditiva a posteriori, que fornece meios concretos de se avaliar a qualidade do instrumento de medida (prova, questionário etc), qualidade do ajuste do modelo de um modo global, além de indícios de violações de suposições específicas. Estudos de simulação, considerando diversas situações de interesse prático, indicam que os modelos e métodos de estimação produzem resultados bastante satisfatórios, com superioridade dos modelos assimétricos com relação ao simétrico (o qual assume simetria das distribuições das variáveis latentes). A análise de um conjunto de dados reais, referente à primeira fase do vestibular da UNICAMP de 2013, ilustra o potencial da tríade: modelagem, métodos de estimação e ferramentas de diagnósticos, desenvolvida neste trabalho / Abstract: In this work, we propose new Item Response Theory models for nominal and ordinal (gradual) polytomous responses through augmented data schemes considering multiple groups. For the distribution of the latent traits of each group, we consider a skew-normal distribution under the centered parametrization. This approach will allow for accommodating a possible skewness of the latent trait distribution, but is also helpful to ensure the identifiability of the models, which is studied under frequentist and Bayesian paradigms. Concerning estimation methods, we developed Bayesian methods through Markov chain Monte Carlo (MCMC) algorithms by using the Gibbs algorithm (DAGS), with augmented likelihood (augmented data) and Metropolis-Hastings algorithms, considering the original likelihood. The computational environment was written in the C++ language and integrated with the R program (a statistical computational and graphical environment), allowing for free, open source and high-speed routines which, in turn, are essential to the dissemination of the developed methodologies. In terms of model selection, we considered the deviance information criterion (DIC), the expected Akaike information criterion (EAIC) and expected Bayesian information criterion (EBIC). Regarding model-fit assessment tools, we explore the posterior predictive model- checking which allows for assessing the quality of measurement, instruments (tests, questionnaires, and others), the model fit in a global sense, besides providing directions toward violations of specific assumptions. Simulation studies, considering different situations of practical interest, indicate that the models and estimation methods produced reasonable results, with outperformance of skew models when compared to symmetric ones (which assumes symmetry of the latent trait distribution). Analysis of a data set which corresponds to the first phase of the 2013 written examination of UNICAMP (State University of Campinas), illustrates the potential of the following triad: modelling; estimation methods; and diagnostic tools developed in this work. / Mestrado / Estatistica / Mestre em Estatística