Recent Advances in Data Anonymization and Data Modelling for Education and Poverty Research
Category: International Association of Survey Statisticians (IASS)
In this invited paper session, recent advances in data anonymization methods will be discussed, as well as the challenges faced by analysts in estimating statistical models focused on research applied to education evaluation, the estimation multidimensional poverty, and social policy impact evaluation, using administrative data and/or complex survey data, collected either cross sectionally or longitudinally, that have undergone anonymization procedures. Classes of models considered shall include graded response models, item response models, fixed and random effects models, and impact evaluation models such as difference in differences models and propensity scores matching methods, among other. We expect the methodologies, applications and results presented in this session to be very useful for many of the 64th ISI World Statistics Congress participants, analysts, researchers and decision makers of education and anti-poverty public policies, which are among the UN's sustainable development goals. More specifically, UN goals 1 and 2 are "end poverty in all its forms everywhere" and "end hunger, achieve food security and improved nutrition and promote sustainable agriculture", respectively, while goal 4 is "ensure inclusive and equitable quality education and promote lifelong learning opportunities for all". This session will have the participation of researchers from three continents - Latin America and Caribbean, Europe and Africa, being four presenters and one discussant, four women and one man, professors in higher education and research institutions, with experience, relevant contributions and publications in journals in the areas of social statistics, psychometry, education, economics, applied social sciences and epidemiology.