Transcription
Hello! Welcome to the Research and Disinformation Series. These videos were made possible by a partnership between the Library and the Rhetoric department at the University of Iowa. My name is Chris Wei, and all my work on the series has been in collaboration with Tim Arnold and Katie Hassman.
In this episode, "Episode 2: Bias Is a Red Herring," we're going to do a deep dive into bias. By the end, you'll be able to talk about bias, identify its various characteristics, and describe strategies for dealing with it in your own research. [Professor Plum: "So it was all nothing to do with a disappearing nuclear physicist and Colonel Mustard's work on the new fusion bomb!" Wadsworth: "Communism was just a red herring."] The reference to the idiom "red herring" in the title simply means it's a distraction.
You've probably been taught to avoid bias entirely. Yes, bias can be troubling and frustrating—it can even be dangerous—but when people tell you that a source is useless because it's biased, they may not understand how bias works and how ordinary it is. From the perspective of evolutionary biology, for instance, bias is an essential part of being alive; it's a survival instinct. So forget about considering whether bias is "good" or "bad"—it just is. It's expected. It's inevitable.
To clarify, I'll remind us of another term: "disinformation." You'll remember from Episode 1 that disinformation is the bad guy, the supervillain, the thing to really watch out for. We defined disinformation as information that's factually inaccurate or misrepresented, or simply, a lie. [John Mulaney: "I was lying. It was a lie."] Bias, on the other hand, is quite different. A biased source simply exhibits some kind of preference. But, as noted in Episode 1, this preference is sometimes invisible, so we have to read between or around the lines to see it.
There are some systems in place on an institutional level that help mitigate the dangers of bias (things like peer review systems), but we'll talk more about peer review in a future episode about authority. Here, our job is to discuss strategies for identifying and analyzing bias as individual researchers. One strategy is to look for evaluative language. "Evaluative" language assigns value to a phenomenon, state of affairs, or person. This is different than descriptive or analytical language.
Descriptive claims describe: "this banana is yellow," "this banana is unpeeled," "I'm holding it in my hand." These are statements of fact. An analytical claim takes facts to build an analysis, point out a pattern, or make a prediction. If you told me that based on its color this banana is ready to be eaten (culinary analysis), or that I often buy bananas and will likely buy them again (economic or behavioral analysis), these are analytical claims—statements that analyze. An evaluative claim evaluates or casts judgment: "that banana looks delicious," "Chris is holding the banana in a weird way," "the banana metaphor is not helping!" These are statements of opinion. Often, a lot of evaluative language—many opinions—signals a preference or bias. In the worst cases, this language is used to shift your opinion and persuade you to accept conclusions regardless of the facts.
A second strategy is to look for "selective focus." Is the source focused only on a small selection of relevant facts and opinions? Are important contexts or perspectives conspicuously left out? This might signal bias. This is all getting very abstract, so let's do something hands-on. Find a book, article, or even a tweet you've recently read and remember. It could be fiction or non-fiction. I'll find some myself and be back in a few seconds, but feel free to pause the video if you need more time. Okay. Think about how this text positions itself with a certain point of view. [Obi-Wan: "A certain point of view." Luke: "A certain point of view?"]
Here are some examples from my shelves. This book, "We Should All Be Feminists," has a prescriptive sentence in its title, carving out an ideological space. It foregrounds its position before you even open it. It clearly has a preference in favor of feminism and against anti-feminism. Of course, the author adds subtlety and nuance, anticipating and responding to reader presuppositions, but the point is that this book clearly has a point of view—a preference—made explicit on its cover. This is also true of this book… or this book… or this book. The argument is laid out in the beginning. You often see this with online articles and short-form media, but sometimes with books. Other texts are more opaque. They don't tell you their position, so finding biases might require more digging. In either case (whether the text warns you or not), some digging is always needed.
Which brings us to our exercise: ask yourself three questions about your chosen text. You can pause at any time. In a classroom or shared space, discuss these questions out loud. Here they are: Question One: "What are the assumptions that this text makes about the world?" For example, "The Very Bad Bunny" takes for granted a worldview with "good" and "bad" people (or bunnies), and "badness" as a trait. I find this troubling, but my interpretation might be biased! If you feel differently, email me. Question Two: "What kind of perspectives does this text amplify, and what kind of perspectives does it leave out?" This is the "selective focus" phenomenon. Question Three: "Does this text use evaluative language? And if so, what kind, how often, and in what context?" You can do this with fiction as well as non-fiction. With "A Tale of Two Cities," I might ask what characters Dickens emphasizes and what evaluative language he uses.
Thank you for indulging that exercise; I hope it was helpful. Finally, let's disabuse you of a common notion. When people talk about "bias," they often think only of POLITICAL bias ("left" versus "right," etc.), but not all biases are political. Here's a list of common cognitive biases—biases we buy into subconsciously. Bias is defined not just by what we're paying attention to and assuming, but also by what we're ignoring. Our job as researchers is not only identifying bias but also looking deeply and widely enough to discern when important facts are left out. This doesn't mean biased sources are useless. It means they might have questionable reliability and offer an incomplete view. We combat this by drawing from a wide variety of sources and gathering many facts to offset any attempts to brush facts under the rug. As long as the facts are accurate, this strategy will likely work. We only get into real trouble when facts are manipulated and distorted—when we're fed disinformation—when we're lied to. We'll talk more about that in Episode 3. Until then, stay safe, stay informed, and we'll see you next time!