Invited Seminar Talk

The Possible Case for Web Scraping and Natural Language Processing in Family Demography

Ross Barker, Vienna Institute of Demography
Laboratory of Digital and Computational Demography, June 07, 2022

Abstract

In this talk, Ross Barker will discuss how family demographers can use web scraping and text analysis to further our understanding of fertility experiences - such experiences that can be difficult to capture with traditional data. Web scraping provides demographers with new data sources that can help us understand micro-level processes, such as complex emotions and experiences across all areas of life, from employment and partnerships, to fertility decisions and issues in conception. Demography has increasingly used macro-level internet data, but there are few examples of the use of micro-level text analysis. He will share the steps of web scraping, followed by topic modelling, and finally discuss how Natural Language Processing could be more broadly used in fertility research. He will link these stages of research to three ongoing projects: a study of web forums discussing issues in conception, a study of fertility-tracking apps and a study of dating apps.

About

Ross Barker is a Research Assistant at the Vienna Institute of Demography. His work focusses on family demography and the uncertainty of fertility decisions while he is increasingly moving towards the application of text-based methods within demography. In pursuing this, he will begin a PhD in September 2022 at the London School of Economics. This PhD is based on the ideas and research he will present in this seminar. Twitter: @rosssbarker