Teck Kiang Tan is a statistician and research fellow at ALSET. In November 2017, he published his first book, “Doubly Classified Model with R.”
Tell us about your background and what brought you to ALSET.
I’m a statistician who spent a large part of my career as an education researcher. Prior to joining ALSET a few months ago, I worked for 4 years as a research fellow at the Institute of Adult Learning and the National Institute of Education in Singapore. I’m passionate about applying the latest statistical techniques to research on learning science and education technology, so I was pleased when the opportunity came up to join a new institute dedicated to these topics at the National University of Singapore.
What is the subject of your new book?
The title of the book is “Doubly Classified Model with R.” It was published by Springer in November 2017. The book explores an important statistical concept called doubly classified models. These are useful in a range of fields. For example, social science researchers might use of doubly classified models to examine the patterns of intergeneration social mobility. Clinicians might use them to analyze associations between the right and left eyes.
What inspired you to write this book?
Publishing a book on statistics and contributing meaningfully to the field was always a dream of mine. I chose to write about doubly classified models because I felt that they were under-discussed in textbooks. My book elaborates on a wide range of relevant doubly classified models, going beyond the commonly discussed models like independence and quasi symmetry models. It also aims to make the subject matter more accessible, including for people who are not comfortable with mathematical symbols and for those who don’t have access to the latest and most common statistical software.
How did you go about making these concepts more accessible?
Since not everyone is comfortable with mathematical symbols, I created a representation with lesser mathematical burden to supplement the mathematical notations that are more complex algebraically into a table form with mathematical symbols to depict the characteristics of a doubly classified model. This “symbolic table” is a graphical tool to help readers understand the various models discussed. The usefulness of symbolic table is not restricted to a single model description—association among models could also be observed through the symbolic tables.
I also made effort to develop syntax for generating models on Package R, a free statistical software that is gaining widespread popularity. Not everyone has access to research funds for more common statistical software like SAS, STATA, MINTAB and SPSS, which can be expensive and not easily accessible.
Since the book includes tips and guidelines on how to develop double classified models, it may also be useful for readers who are interested in learning about Package R and use the guided steps as a lesson of R’s functionalities. Whatever the purpose, I am sure that many readers will appreciate the examples included to illustrate concepts and its applications.
How is the book structured and what else does it contribute to the field?
The doubly classified models are grouped into four main categories, namely asymmetry models, point symmetry models, non-independent models, and asymmetry + non-independent models. It also covers testing of symmetry, modelling strategies, graphical approaches in modelling, creating new doubly classified models in accordance to researchers’ need, and introduces three new doubly classified models.
Founded in 2016, ALSET’s mission is improve education through the application of learning science and education technology. The Institute conducts original research on learning science, technology, and pedagogy; promotes novel and entrepreneurial projects that improve learning outcomes; and works to ensure that the latest research and learning technologies have broad impact, both at NUS and also in the broader education community.