CURSO:MODELOS DE ERROR DE MEDICION TRADUCCION:MEASUREMENT ERROR MODELS SIGLA:SOL4038 CREDITOS:05 MODULOS:02 CARACTER:OPTATIVO TIPO:CATEDRA CALIFICACION:ESTANDAR DISCIPLINA:SOCIOLOGIA PALABRAS CLAVE:METODOLOGIA DE ENCUESTAS, CIENCIA DE DATOS NIVEL FORMATIVO:MAGISTER I.DESCRIPCIÓN DEL CURSO Surveys reflect the opinions or facts researchers are after only partly ? the other part will be measurement error, which can seriously bias analyses of interest. To remove such biases it is essential to estimate the extent of measurement error in survey variables, which is precisely the goal of statistical measurement error modeling. In this course, we will discuss how measurement error can be defined, how its presence can be detected using specialized data collection designs and models, and how to perform error-corrected statistical analyses of substantive interest. II.OBJETIVOS DE APRENDIZAJE 1.Define measurement error conceptually, including the concepts of reliability and validity; 2.Explain the different approaches to estimating measurement error and their respective advantages and drawbacks; 3.Interpret the results of statistical models used to estimate measurement error in the absence of a gold standard; 4.Perform regression analyses from which the influence of measurement error has been removed and interpret the results. III.CONTENIDOS 1.Concepts of measurement error: reliability, validity, common method variance, misclassification, true score, 2.Measurement error in continuous survey variables: test-retest, consistency, multitrait-multimethod, quasi-simplex, structural equation modelling 3.Measurement error in categorical survey variables: sensitivity, specificity, hidden markov models (HMM), latent class analysis (LCA) 4.Correcting regression analyses for the effects of measurement error: correction for attenuation, errors-in-variables regression, covariance reduction, three-step analysis Prerequisites 1.Knowledge of basic statistics including regression analysis; 2.Ability to run an R script, for example from RStudio; a cursory understanding of R; 3.In-depth knowledge of R or latent variable models is NOT required. IV.METODOLOGIA PARA EL APRENDIZAJE -Flipped-classroom -Lectures delivered through pre-recorded online video sessions -Live meetings, via a web platform, with discussions -Quizzes and Homeworks -Readings V.EVALUACION DE APRENDIZAJES -online quizzes: 20% -online homeworks: 20% -A final open-book exam: 50% -Class participation in online meetings and forum: 10% VI.BIBLIOGRAFIA Required readings (articles)* Alwin, D. (2007). Margins of Error. New York: Wiley. Chapters 1-3. Saris, W.E. & Gallhofer, I.N. Design, evaluation, and analysis of questionnaires for survey research. New York: Wiley. Chapter 9. Alwin, D. (2007). Margins of Error. New York: Wiley. Chapters 4-6. Saris, W.E. & Gallhofer, I.N. Design, evaluation, and analysis of questionnaires for survey research. New York: Wiley. Chapter 10. Saris et al. (2011). Final report about the project JRA3 as part of ESS Infrastructure. Chapters 6 & 7. Survey Quality Predictor (SQP 2.1). Tutorial. URL: http://sqp.upf.edu/ Alwin, D. (2007). Margins of Error. New York: Wiley. Chapter 11. Biemer, P. (2011). Latent class analysis of survey error. New York: Wiley. Chapters 4; 6-7. Kreuter, F., Yan , T. & Tourangeau, R. (2008). Good item or bad ? can latent class analysis tell?: the utility of latent class analysis for the evaluation of survey questions. J. R. Statist. Soc. A, 171 (3) Bakk, Tekle & Vermunt (2013). Estimating the Association between Latent Class Membership and External Variables Using Bias-adjusted Three-step Approaches. Sociological Methodology, 43, p. 272. Castellarnau, A. & Saris, W.E. (2015). A simple procedure to correct for measurement errors in survey research. URL: http://essedunet.nsd.uib.no/cms/topics/measurement/ Saris, W.E. & Gallhofer, I.N. Design, evaluation, and analysis of questionnaires for survey research. New York: Wiley. Chapter 15. Muthen & Asparouhov. Three step webnote. Additional required and recommended readings will be made available on the course website. None of the information in the recommended readings will be included on a homework assignment or the final exam. PONTIFICIA UNIVERSIDAD CATOLICA DE CHILE INSTITUTO DE SOCIOLOGIA / NOVIEMBRE 2018