@inbook{24a4cec63e9546fe8a3de5517c68a136,
title = "Causal models and big data learning analytics: Evolution of causal relation between learning efficiency and instructional effectiveness",
abstract = "New statistical methods allow discovery of causal models purely from observational data in some circumstances. Educational research that does not easily lend itself to experimental investigation can benefit from such discovery, particularly when the process of inquiry potentially affects measurement. Whether controlled or authentic, educational inquiry is sprinkled with hidden variables that only change over the long term, making them challenging and expensive to investigate experimentally. Big data learning analytics offers methods and techniques to observe such changes over longer terms at various levels of granularity. Learning analytics also allows construction of candidate models that expound hidden variables as well as their relationships with other variables of interest in the research. This article discusses the core ideas of causality and modeling of causality in the context of educational research with big data analytics as the underlying data supply mechanism. It provides results from studies that illustrate the need for causal modeling and how learning analytics could enhance the accuracy of causal models.",
keywords = "Artificial intelligence in education, Bayesian networks, Big data learning analytics, Causal models",
author = "Kumar, {Vivekanandan Suresh} and Kinshuk and Clayton Clemens and Steven Harris",
note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2015.",
year = "2015",
doi = "10.1007/978-3-662-44659-1_3",
language = "English",
series = "Lecture Notes in Educational Technology",
number = "9783662446584",
pages = "31--53",
booktitle = "Lecture Notes in Educational Technology",
edition = "9783662446584",
}