Fernando de Jesus Moreira Junior

COMPARAÇÃO ENTRE O MÉTODO WARD E O MÉTODO K-MÉDIAS NO AGRUPAMENTO DE PRODUTORES DE LEITE

In this paper method Ward and method K-Means have been usedto compare the result of the clusters to the producers of milk in the regionof Santa Maria, RS. Were used 231 sample of raw milk from the 63 producerscolleted from 6 to 30 of September of 2004. Were analyzed the flowingvariables: percentage of water; percentage of fat, acidity in degree Dornic;density in g/cm3. Both two methods show three clusters for the producerswhich 52 producers stay in the same clusters in both two methods.

Avaliação da satisfação de alunos por meio do Modelo de Resposta Gradual da Teoria da Resposta ao Item

The aim of this study was to use the Item Response Theory (IRT) as a tool to assess the satisfaction of students of the Basic and General Statistics course at the Federal University of Santa Maria (UFSM). Through the application of the Graded Response Model (GRM), a measurement scale for satisfaction was created, where the items of satisfaction and the respondents were positioned in the same unit of measure. The results showed that the GRM of the IRT was suitable for the analysis of data, allowing the identification of problematic items, the construction of a scale for student satisfaction, and the identification of anchor levels. The findings will allow the staff responsible for the teaching of Statistics at the institution to identify how procedures can be improved so that teaching becomes increasingly satisfactory.

SATISFAÇÃO DOS USUÁRIOS DO RESTAURANTE UNIVERSITÁRIO DA UNIVERSIDADE FEDERAL DE SANTA MARIA: UMA ANÁLISE DESCRITIVA

Essa pesquisa tem por objetivo avaliar o nivel de satisfacao dos usuarios do Restaurante Universitario da Universidade Federal de Santa Maria - RS, por meio de metodos estatisticos descritivos. Foi elaborado um questionario com questoes de perfil, avaliacao da satisfacao e da importância em relacao a 41 itens relacionados ao Restaurante Universitario. Os itens de satisfacao apresentaram uma consistencia interna de 0,93. Participaram da pesquisa, 1.855 usuarios, entre alunos, docentes e tecnicos administrativos. Observou-se que a maioria dos usuarios e do sexo feminino, solteiros, situados na faixa etaria de 17 a 30 anos, alunos sem o beneficio socioeconomico, que cursam a graduacao e almocam em grupo. A maioria dos respondentes frequenta o restaurante entre o meio dia e uma hora da tarde, e ate pelo menos tres vezes por semana, onde o preco atrativo e a principal motivacao para frequenta-lo. A maioria dos atributos foram avaliados de forma positiva, com destaque para a iluminacao interna, o preco cobrado, a localizacao, a acessibilidade, as formas de pagamento, o atendimento, o tamanho do restaurante, a organizacao do buffet, o conforto e o layout. Com relacao aos itens avaliados insatisfatoriamente, merecem destaque a ventilacao, o abrigo contra o sol e a chuva e os itens relacionados com a demanda de usuarios.

Evaluación de satisfacción de alumnos a través del Modelo de Respuesta Gradual de la Teoría de Respuesta al Ítem

The aim of this study was to use the Item Response Theory (IRT) as a tool to assess the satisfaction of students of the Basic and General Statistics course at the Federal University of Santa Maria (UFSM). Through the application of the Graded Response Model (GRM), a measurement scale for satisfaction was created, where the items of satisfaction and the respondents were positioned in the same unit of measure. The results showed that the GRM of the IRT was suitable for the analysis of data, allowing the identification of problematic items, the construction of a scale for student satisfaction, and the identification of anchor levels. The findings will allow the staff responsible for the teaching of Statistics at the institution to identify how procedures can be improved so that teaching becomes increasingly satisfactory.

Student satisfaction evaluation through the Gradual Response Model of Item Response Theory

The aim of this study was to use the Item Response Theory (IRT) as a tool to assess the satisfaction of students of the Basic and General Statistics course at the Federal University of Santa Maria (UFSM). Through the application of the Graded Response Model (GRM), a measurement scale for satisfaction was created, where the items of satisfaction and the respondents were positioned in the same unit of measure. The results showed that the GRM of the IRT was suitable for the analysis of data, allowing the identification of problematic items, the construction of a scale for student satisfaction, and the identification of anchor levels. The findings will allow the staff responsible for the teaching of Statistics at the institution to identify how procedures can be improved so that teaching becomes increasingly satisfactory.

CONTRIBUTIONS OF ITEM RESPONSE THEORY IN EDUCATIONAL ASSESSMENTS

The design, development and implementation of evaluation methods to represent the learning outcomes have been long been a concern among pedagogues, psychologists and educators. Traditionally, educational assessments are based on scores from the so-called Classical Test Theory (CTT) and Classical Theory of Measure (CTM). In this context arises the Item Response Theory (IRT), a powerful statistical tool that can meet the needs arising from the limitations of TCT. The purpose of this paper is to present the main contributions of IRT in relation to CTT, in the context of educational evaluations, and present an illustration of these contributions IRT, through simulations.

Algoritmo de um teste adaptativo informatizado com base na teoria da resposta ao item para a estimação da usabilidade de sites de e-commerce

Abstract This paper proposes an algorithm of a computerized adaptive testing based on Item Response Theory, designed to estimate the degree of usability of e-commerce sites. Five algorithms were tested by simulation. The algorithm with the best performance was applied to real data from 361 e-commerce sites. The results showed that the algorithm could obtain good estimates for the degree of usability of e-commerce sites with the application of 13 items. Keywords Computerized adaptive testing. Item response theory. Usability. e-commerce sites. SCHENKMAN, B. N.; JONSSON, F. U. Aesthetics and preferences of web pages. Behaviour & Information Technology, v. 19, n. 5, p. 367-377, 2000. http://dx.doi.org/10.1080/014492900750000063SUKAMOLSON, S. Computerized Test/Item Banking and Computerized Adaptive Testing for Teachers and Lecturers. Information Technology and Universities in Asia TOIT, M. – ITUA, 2002.STRAUB, D. W. Validating instruments in MIS research. MIS Quarterly, v. 13, n. 2, p. 147-169, 1989. http://dx.doi.org/10.2307/248922TAVARES, H. R.; ANDRADE, D. F.; PEREIRA, C. A. Detection of determinant genes and diagnostic via item response theory. Genetics and Molecular Biology, v. 27, n. 4, p. 679-685,http://dx.doi.org/10.1590/ 2004. S1415-47572004000400033TEJADA, A. J. R. Pasado, presente y futuro de los Tests Adaptativos Informatizados: entrevista con Isaac I. Bejar. Psicothema, v. 13, n. 4, p. 685-690, 2001.TEZZA, R.; BORNIA, A. C.; ANDRADE, D. F. Measuring web usability using item response theory: Principles, features and opportunities. Interacting with Computers, v. 23, n. 2, p. 167-175,http://dx.doi.org/10.1016/j. 2011. intcom.2011.02.004IRT from SSI: BILOG-MG, MULTILOG, PARSCALE, TESTFACT. Scientific Software International, 2003.VAN DER LINDEN, W. J.; GLAS, C. A. W. Computerized Adaptive Testing: Theory and Practice. Dordrecht: Kluwer Academic, 2000.WAINER, H. CATs: Whither and whence. Psicologica: revista de metodologia y psicologia experimental, v. 21, n. 1, p. 121-133, 2000a.WAINER, H. Computerized Adaptive Testing: A Primer. New Jersey: Lawrence Erlbaum Associates, 2000b.

ESTIMAÇÃO DA USABILIDADE DE SITES E-COMMERCE PELO MÉTODO DA MÁXIMA VEROSSIMILHANÇA

This article investigates the performance of Maximum Likelihood (ML)method to estimate the usability of e-commerce sites and compares theperformance between the software Bilog-MG ® and Excel ® in theestimation of these usability. For this, we used real data from a study onthe degree of usability of 361 e-commerce sites, which were applied 32items calibrated by the unidimensional logistic model of two parameters(MLU2) of Item Response Theory (IRT). The estimation process by MLwas developed in BILOG® and Excel® softwares. The results showedthat the ML method is flawed when there is a constant pattern of response,which may occur during application of the first items on the questionnaire.However, the method performs well when the pattern of responses is notconstant. Moreover, process performance prepared in Excel® was betterthan in conventional software BILOG-MG®. The parameters of the itemsalso influence the estimation of ML.