Book Review-1
Book Review-1
Muhammad Tahir
MS Scholar
Department of English, University of Sialkot
Email: [email protected]
The book “Data Analytics for Discourse Analysis with Python: The Case of Therapy
Talk” by Dennis Tay is a unique and helpful resource that connects discourse
analysis with data analytics, especially in psychotherapy. Tay addresses the lack of
quantitative methods in discourse research and provides a straightforward guide to
using Python for analyzing therapy conversations, making it suitable for those with
little programming experience. The book blends theory and practice well,
demonstrating how simulations, clustering, and time series analysis can enhance
qualitative research. Each chapter includes practical case studies and annotated
Python code, allowing readers to implement the techniques in their own projects.
This practical Introduction is particularly beneficial for linguists, data scientists,
psychologists, and mental health professionals looking to use computational
methods in discourse analysis. This book is highly recommended for those in these
fields who want to integrate computational techniques into their work.
Chapter 3 covers cluster analysis, a technique for grouping data points based on
their similarities. Tay applies this method to study therapist-client interactions,
focusing on the linguistic synchrony between them. The chapter is split into
theoretical and practical parts, starting with an overview of clustering methods like
agglomerative hierarchical clustering (AHC) and k-means clustering. Tay clearly
explains how these methods function and highlights their usefulness in organizing
discourse data into significant categories. A case study demonstrates the use of k-
means clustering on therapist-client interactions, illustrating how language patterns
can indicate synchronization or asynchrony in communication styles. This method
provides a fresh way to measure and analyze the dynamics of therapeutic
conversations. The chapter also includes Python code for performing cluster
analysis, adding practical value. Readers are walked through each step, from data
preparation to model validation, making it easy for both linguists and data scientists
to follow. Overall, this chapter shows how cluster analysis can be used with
discourse data, offering insights into communication structure and flow in therapy.
Chapter 5 explores time series analysis, a useful method for studying how language
changes during therapy sessions. Tay explains the basic ideas of time series
analysis and how it helps monitor shifts in communication styles and emotional
expressions between therapists and clients over time. This chapter is important
because it highlights that discourse data is not fixed; it changes throughout therapy.
Tay shows how to model and predict these changes using Python, covering essential
steps like analyzing the time series, calculating autocorrelation, identifying models,
and fitting them to the data. A case study on the development of therapy
conversations across sessions illustrates how time series models can predict
language changes and assess the success of therapy. The chapter is notable for its
careful methodology and practical use. Tay’s clear guide allows readers to use time
series analysis for therapy discussions and other long-term communication data. By
combining predictive modeling with discourse analysis, this chapter provides
valuable insights into tracking and understanding language changes quantitatively.