Skip to main content

Educational Data Science


5 weeks

About this course


The application of quantitative analysis techniques in education involves certain specific difficulties. Students study in institutions with their own characteristics and metrics of academic success may contain non-negligible measurement errors. The recent development of data analysis techniques to large databases can enrich data-driven decision making in education. This course is an opportunity for professionals in the educational sector who wish to be able to combine recent data analysis techniques with their knowledge and experience in the educational field where they work or aspire to work.

What I will learn

In this course you will learn to identify the problems that can be solved using these techniques, as well as their correct application and interpretation of results. To this end, we will assess the information provided by learning analytics, institutional databases and international educational assessments. Regarding data preparation, we will work on techniques for the treatment of missing and outliers, as well as discretization and data balancing. We will also review different approaches to combining data from different sources. Among the data analysis techniques, we will start with descriptive analysis, and then address stratified analysis, correlated data, visualization and hypothesis testing. Finally, we will learn how to obtain and interpret predictive models useful in education, such as factor analysis, reliability analysis, cluster analysis and multilevel regression models. Additionally, we will describe how to define and construct composite indicators useful for decision making in educational institutions.


There are no specific requirements. Knowledge of programming and basic statistics is recommended.


  1. Introduction
  2. Data sources and Data preparation in Education
  3. Data analysis and understanding
  4. Predictive models in education I
  5. Predictive models in education II
  6. Final Exam

Course Faculty

Andrea Conchado

Andrea Conchado holds a PhD in Industrial Engineering by the Universidad Politécnica de Valencia. She works as an assistant professor in the Department of Applied Statistics and Operational Research and Quality in this university. She has co-authored research papers in scientific journals, research books and book chapters with prestigious publishers. She has collaborated as a researcher in research projects, most of them funded by the European Commission. Her research interests lie in the areas of research methodology and applied statistics for social sciences, specifically the application of structural equation modelling to the assessment of the validity and reliability of measurement instruments in Education.

Cláudia Antunes

Cláudia Antunes is an Associate Professor at Instituto Superior Técnico – Universidade de Lisboa. Concluded her PhD in Information Systems and Computer Engineering by the same university, on 2005, in the data science domain, proposing new methods and methodologies to deal with temporal data, in particular for mining sequential patterns. Her main research interests are in the area of Data Science, in particular to what concerns to the usage of knowledge domain, the exploration of the temporality to enrich the classification task, and to use it to enhance the educational process. She has coordinated and participated on several national and European research projects and has more than seventy papers published in journals and international conferences. Along with this work, she supervised Master and PhDs students, accounting for about forty students who concluded their work under her supervision. She has been lecturing data science courses for twenty years, both in graduation and post-graduation programs.


La Universitat Politècnica de València utiliza cookies propias y de terceros por motivos de seguridad, y también para mejorar la experiencia del usuario y conocer sus hábitos de navegación. Recuerda que, al utilizar nuestros servicios, admites nuestro aviso legal y nuestra política de cookies. Entendemos que, si continuas navegando, es porque apruebas estos términos.