I had intended to take a break from topics with obvious political overtones, but as a periodic browser of online bookstores and their linked purchase data, I couldn’t pass up this study of related book purchases and what they reveal about how our politics and our science mingle. The researchers started with two political books with clear party affinity and followed the “Customers who bought this title also bought” links from those two books to all associated titles, and from those titles to their associated titles and so on. They found strong interest in the sciences across the political spectrum; the commentary expresses at least mild surprise at this finding, but given that their study population is customers who purchase multiple books I’d guess they are dealing with a relatively well-educated and well-read cohort. At the same time, there were notable differences in both the specific disciplines of interest to consumers with different political leaning and the types of books chosen within certain disciplines of common interest.
The researchers summarized the findings as a tendency among liberals to prefer books on basic or abstract science while conservatives tend to prefer more applied sciences. Of course, it should be noted there is opportunity to skew interpretation simply by choosing these categories. For example, in this article one of the authors describes the fields preferred by liberals as driven by “curiosity,” a description I wouldn’t exactly call value-neutral or even necessarily accurate as it is possible to be curious about how to achieve a practical result in medicine or engineering. I was also not encouraged by the use of ‘red’ and ‘blue’ as political category labels; I am concerned that our overuse of these arbitrary terms creates or reinforces the impression that we are simply two teams within inscrutable reasoning and motivations. ‘Liberal’ and ‘conservative’ are reductive as it is, but they at least suggest something of an ideology that can be understood. ‘Red’ or ‘blue’ voters are simply and unknowably Other. Thus in talking about this work, which I am willing to accept did reveal genuine clusters of books that differed along party lines, we may be inadvertently revealing additional facets of the communication challenges posed by our present political landscape.
Incidentally, I’ve actually explored these online book connections on a much smaller scale for my own personal curiosity; I start with a book I’m interested in and try to find related books. In my case, I’m looking for further reading material and also trying to understand the breadth of material on a given topic. In the future I may need to be more aware of the limitations of this approach in exploring the full range of a topic. I would also add that those related purchases results can differ depending on whether the retailer knows it is you browsing their site or not. So there may be further filtering with potential ideological slants introduced by the algorithms attempting to personalize your shopping experience.
While my personal curiosity about online bookstores initially drew me to the story, there are some practical considerations as well. Going forward with my blog posts, I may not be able to avoid political associations to the degree I might desire. I’m still going to try to cover a range of topics, and I’m still going to have the same personal interests, but I might start to wonder if covering quantum entanglement or biotechnology will be construed politically. Have you ever encountered unexpected political associations or other kinds of unexpected connections in your field?