Constellate has now added 44 of their most-demanded instructional sessions to their YouTube channel: https://www.youtube.com/@Constellate_org. Check them out if you are looking for a way to start applying text analysis techniques to your research problem!
Category: research
Categories
SQSP 2025
Marcus Charlesworth et moi sommes ravis d’animer la Table ronde 3 au Congrès annuel de la Société québécoise de science politique: https://sqsp.uqam.ca/congres/programme-preliminaire-congres-2025-uqam/
Categories
CPSA 2025
I will be presenting and chairing a panel at CPSA this coming June: Classical Political Thought, Democratic Theory and Practice. Look me up if you are planning to attend!
For those interested in thoughtful commentary on current issues in Canadian political economy and international affairs, I recommend having a look at my friend Andrew Strebkov’s blog: Opinions International.
If you are interested in case studies of democratization, Polity has just released an article where I analyze the role of religious practices and institutions in normalizing a democratic transition: Democratic Civil Religion and the Kleisthenic Reforms.
I am pleased to announce the launch of The Parisian Parliamentarians Projet as a collection on the Internet Archive!
This society may be of interest to those pursuing digital humanities projects on a topic in French history: https://h-france.net/. The index of (digital) primary sources it maintains is particularly comprehensive: https://h-france.net/professional-resources/research-tools/.
Categories
Text Analysis Guides
For those who might be interested in experimenting with text analysis I highly recommend Melanie Walsh’s guide. It walks researchers through the basics of installing and using Python, data collection, and applying text modelling tools. Researchers can even experiment with running the code directly on the website. Those interested in smaller guides to specific portions of the research pipeline can find a wealth of notebooks on the Constellate GitHub site.
Here is a handy introduction to the basic principles of machine learning: Supervised vs Unsupervised Learning – What’s the Difference?. While reading it I paused to wonder whether there is any way to use machine learning for analyzing causal relationships. Machine learning automates the creation and evaluation of models but it doesn’t identify confounding variables, to run robustness checks, etc. I imagine that the predictions based on ML results tend to be rather conservative since they extrapolate the present to the future, rather than use a series of past events to identify root causes. Please feel free to correct me if there are already ML tools that have causal research functionality!
Categories
Reading as Visual Processing
I ran across this very interesting article explaining that data visualization works because it draws our brain’s attention to outstanding features within our field of vision. This helps us get on with the business of thinking about what we are seeing rather than simply looking for something to focus on. I got to thinking about the way that I read text sources and decided that something similar must be going on. If I ready a difficult passage several times, I always read the exact same words. This defeats the purpose of filling in the context of the phrases I did understand, but it suggests that our brains are actively searching for salient features in a text without us knowing about it.
I don’t know how these insights could help me read better, but it does help me trust that my version of speed-reading picks up whatever I am able to process in the moment. Has anyone else had an experience similar to the one I am describing? Do you find the presence of marginal notes or keywords helpful in drawing your eye to particular passages? Does font or paragraph length make a difference to how much you pick out?
Lucas Jerusalimiec 