Happy New Year! As I reflected over the last few days on the previous year, one of the things I thought about was how I tend to enjoy academic work that crosses disciplines – I am interested in the history of science and mathematics; the economics underlying communications; the art of statistics. Is it the intersection of categories that I enjoy? Or is it just that my research interests do not align with many established communities of practice? I’m inclined to think that it is the intersection, or if not the intersection, the breadth and novelty afforded by interdisciplinary work.
I also had another stray thought related — at least thematically — to this line of thinking. How much do academic interdisciplinarians (such as digital humanities scholars) interact with applied interdisciplinarians (evaluation consultants and the like)? What skills are specific to interdisciplinary work; do these skills translate between settings?
A still from Tron, because I used the word “users” a lot in my talk.
Back in October, I spoke as part of a panel at the Society for Utopian Studies conference, discussing artificial intelligence and dystopian/utopian visions of the future of higher education. I posted some rough notes from my presentation here: LINK
On a recent episode of The Joe Rogan Experience, Mr. Rogan interviews Sean Carroll, a professor of physics at the California Institute of Technology with a new book to promote. Dr. Carroll is a proponent of the Many Worlds interpretation of quantum physics, one of several frameworks proposed for understanding various counter-intuitive subatomic phenomena like uncertainty or wave-particle duality. Where some interpretations make no statements about the measurement of things like electron spin beyond the probability of a particular measurement occurring, Many Worlds proposes that all possible observations are contained within multiple timelines.
Why does Dr. Carroll find this interpretation appealing compared to other interpretations? He contrasts Many Worlds with “textbook” interpretations of quantum mechanics, specifically concerning thethe “measurement problem” of quantum mechanics. Briefly, the measurement problem is that systems cannot be observed from the outside without physically affecting the system under observation. Only apparent at extremely small scales, it’s related to and often confused with the famous Uncertainty Principle. Rogan accurately describes this as the point where people, “want to go woo-woo” when discussing quantum mechanics and follows up with a good question: “Is it the act of measuring that changes things?” Carroll responds that, “this is The Puzzle… of quantum mechanics” and describes that what is taught in quantum physics courses is merely that, “when a system is observed, when it is measured… its wave function changes suddenly, dramatically, and unpredictably.” This does not answer Rogan’s question, and leaves the viewer with the impression that “observation” here might have something to do with consciousness rather than subatomic interactions.
How would you [examine a single electron]? Well, one way to do it would be to look at it with a very fine instrument, let’s say an electron microscope… the electrons [from the microscope] would be so energetic that they would destroy the quantum state.
Observation might be a confusing term, but in context all it means is that measuring quantum objects requires the use of quantum objects (since everything is made of quantum objects). Not a groundbreaking discovery, but an important realization during the process of understanding how things work at the quantum scale. Interpreting the “unpredictability” of observations that Carrol describes, on the other hand, leads to the different interpretations of quantum mechanics including Many Worlds. Late in the podcast, Rogan brings up Laplace’s Demon and the idea of a deterministic universe, which leads Carroll to say that Many Worlds allows for a clockwork multiverse, while other interpretations of quantum mechanics posit observable reality as random and fundamentally unpredictable. In either case, the math works out the same. I have a feeling that determinism is the main reason that Carroll prefers the Many Worlds interpretation – it allows physics to be discussed not just as an interpretation of observations, but as a potentially complete model of reality.
Is this the primary goal of the thing we call science? To discover absolute truth or fully model the universe? Based on what I heard in the Joe Rogan interview, I think that Carroll would agree with this; popular depictions of “science” opposing “religion” as a worldview also reinforce this idea. However prevalent this view, I do not think that it is productive in the long run. If science is imagined to be the purview of very smart people who can provide big affirmative answers about the universe the focus of discussion becomes the answers rather than the process. Although researchers might include caveats about uncertainty in their findings, when public discussion focuses on the results rather than the process, the evidence observed in the process will get lost in the shuffle. Particularly if those results have direct policy implications.
I think that the discussion over climate change has been set back due to emphasis on the results of science rather at the expense of the methods. The standard environmentalist appeal to lawmakers and the public is to “listen to the scientists” when considering whether to regulate greenhouse gas emissions. Evidence of anthropogenic climate change is also provided, but it usually takes the form of imperfect correlations (between global temperature and carbon emissions, for example). The details of how evidence for the insulating properties of atmospheric CO2 has been accumulated over the last 150 years through experiments and natural observations are left murky, dismissed as too complicated for mere non-scientists to understand. Thus, the discussion becomes not a matter of forming an opinion based on evidence but on selecting the evidence that supports your opinion.
Of course, it’s not just the practice of focusing on scientific results over the research process that leads to this situation; materialism and the promotion of bad-faith arguments in the media also play a role. However, I do not think that the answer-focused messaging has helped sway public opinion toward accepting the reality of climate change. I expect that as time goes on, public opinion will continue to shift toward the environmentalist position, but this will be more due to observable catastrophes like rising sea levels, heat waves and forest fires than sudden belief in scientific pronouncements.
Am I saying that there is no value in scientific discovery? No, but I do think that it is overvalued in science communications at the expense of educating the public about the scientific process. In an era of increasing mistrust, there is something to be said for explaining not just what we think we know, but how we think we know it.
Several months ago, I went to Roxborough State Park in Littleton, CO to practice observational drawing. At the time, I was in the midst of writing my dissertation; I thought that this might be a good break from running analysis, writing up results, thinking about research methodology. The break in routine left me feeling refreshed, or at least less fatigued. However, I did not get the break that I expected. Instead, the action of sketching prompted me to think about my research from a different perspective.
When I was making these sketches I noticed that the act of trying to match the marks I was making on the page to what I was seeing was leading me to think harder about the structure of my subjects. Walking along the trail, I might notice some interesting flowers and leaves, but trying to draw the plants made it more likely that I would try to understand the way that the buds fit to the stem or if the leaves belonged to one plant or several different species. When I was sketching the trail landscape, I had to pause to consider how the trail had been made – was it dug out with a shovel or worn down through foot traffic? By sketching the trails, the trees, the leaves, I was not perfectly representing the scenery, but the process led me to actively consider the relationships between objects and the resulting work emphasized key parts of the whole.
My dissertation focuses on using a structural model incorporating confirmatory factor analysis (CFA) to describe relationships between financial inputs (such as tax revenue) and several measures of public library service access. This modeling approach is meant to test theories about measurement and causal relationships. However, I was having a hard time sustaining my interest in the theory examined in my dissertation. Is it really that surprising that higher revenues lead to more services and higher levels of use? I was having second thoughts about my proposed tests and was considering revising my proposal.
Visiting the park and sketching helped me to think about my dissertation from a different perspective. What if I thought of the analysis not as a test of whether my theory was “true,” but as a process of identifying the patterns in the data emphasized by my chosen medium/method? Could this approach provide more insight into the reality represented by the numbers? I’m not sure if observational drawing is a perfect metaphor for this sort of quantitative analysis, but it helped me find a more productive way to approach my dissertation. It’s important to remember that research should be a cycle of building theory by interpreting observations; not a process of confirming theories that already exist. Despite the name of the method using the word “confirmatory,” I think that confirmatory factor analysis models are useful descriptive frameworks that can be used to improve our understanding of reality; thinking of them as definitive tests of a theory’s validity misses the bigger picture.
Published by an organization called Ad Fontes, it purports to show how different media outlets score in terms of political bias (“left” versus “right” on the horizontal axis) and content (“fact reporting” versus “fabricated information” with several other categories such as “analysis” and “opinion” forced into the vertical axis). I have several issues with this approach, the largest being that the vertical axis of the chart merges the idea of trustworthiness with reporting style, leading to statistical reliability issues not reflected on the production version of the chart (Ad Fontes discusses their methodology here and here). As someone who occasionally teaches information literacy, I also object to using using the chart as a reference because it de-emphasizes critical thinking. When evaluating news stories, is deferring to expert opinion is a useful habit to practice?
Whatever my issues with the chart, I can see the appeal of using a quick reference when trying to teach information literacy/media literacy in a short period of time. Rather than simply presenting this (or a similar chart) as a reference, is there a way to use it to introduce key ideas about news source evaluation?
Ad Fontes has published the methodology underlying their visualization; looking at the thought process underlying the chart could be a useful way to introduce some key ideas in evaluating news sources.
To create their chart, Ad Fontes asked a group of 20 analysts to rate a large set of news articles and broadcasts on two scales: “bias” and “reliability”. The bias scale (the horizontal axis of the chart) was intended to measure the political orientation of the articles, while the reliability scale was intended to measure some combination of accuracy and the balance between statements of fact and opinion statements. There are serious interpretive issues involved in combining the idea of accuracy and voice in a quantitative study. However, the analysts providing raw data were also asked to score items on a separate scales for accuracy and a fact-versus-opinion scale. The scores for these subscales did not make it into the final chart product.
Political bias, accuracy, and opinion are difficult-to-define concepts. However, I think that they are sufficiently distinct to suggest a worthwhile exercise – when looking at a news story – or your regular news sources – place yourself in the role of an analyst. How would you rate the article in terms of political bias, reliability, and opinion content? Why? While you might not come up with specific numerical values or nifty visualizations, I think that the act of considering these questions might help in taking a step back from the act of consuming news media and allow some space to practice the act of evaluating sources.
Thanks for joining me! Apparently, WordPress autopublishes a “this is my first post” post when you start a blog with them. I only noticed this eight months after the fact. Anyway, hi! My name is Ian Burke. I am (in no particular order) an information researcher, statistician, occasional art-doer, and hiker. The name of this blog is Emordnilap, which is “palindrome” spelled backwards. I’m hoping it hits the “witty at first, but less funny each time” sweet spot.
Emordnilap 