Why Do We Love Data?

Source: Geralt/Pixabay

We are living in an age in which data is revered, especially when it's reported in the media. Data is information that has allegedly been scientifically gathered, making the findings seem like irrefutable proof of whatever is being studied. “Big Data,” as it is often called, is considered even more valid and reliable as it aggregates very large amounts of information. Attaching numbers, especially lots of them, to any particular claim lends a good degree of perceived factualness, much more so than the once venerated but now often antiquated cognitive-based aptitudes of intuition, opinion, and judgment.

In Bad Data: Why We Measure the Wrong Things and Often Miss the Metrics That Matter, Peter Schryvers showed how what he called “information hubris” can often lead to poor analytical outcomes. “Big data is often touted as the key to understanding almost every aspect of contemporary life,” his book read, not a good thing given all the variables and assumptions that are made in any study.

Schryvers didn’t go far enough, however. Because it is quantitative, data is assumed to be not just truthful but somehow meaningful—even if different studies report very different findings. Percentiles are considered hard evidence, even if the conclusions drawn from the data don’t mean a whole lot. For example, “40 percent of adults say they use Instagram and about three-in-ten report using Pinterest or LinkedIn,” a Pew Research Center survey reported in April 2021. Is that a lot or a little? I don’t really know, but the numbers are presented as being significant in some way.

Unless an argument is “evidence-based,” however, making any case is deemed flimsy and weak. One might repeatedly see something with one’s own eyes, but, in this era of data, it matters little. Observations and any conclusions drawn, even by experts in a relevant field, have been demoted to “personal perspectives,” this despite the fact that they are based on real, documented behavior. Most data-driven research, on the other hand, is based on people’s beliefs and attitudes, which I consider to be far flimsier than what people actually do.

In short, if something is published in a journal with quantitative support, it carries much weight. If a cultural anthropologist with a Ph.D. (like myself, it so happens) reports a particular trend from field research, however, it is considered light fluff or anecdotal fodder that may complement the “real,” i.e., data-derived studies. People want to know the “sample size” of cultural research, not understanding that valuable insights into human behavior can be derived without metrics and analytics.

The many flaws associated with quantitative research also should be acknowledged. There are infinite ways to design a study and gather findings, each one likely to produce different results. The move to online research has made findings that much more questionable, as nothing gleaned from the Internet should be taken too seriously. Beyond that, how questions are both asked and answered is highly subjective and dependent on many variables. I may answer a question entirely differently based on whether I’m having a good or bad day (or had my coffee yet), for example, but our attraction to all-things-data persists.

Why do we love data so much? Words are vague and have multiple meanings, while numbers are precise and definitive, a big reason why we put so much faith and trust in data regardless of its source.

To legitimize this post, note that 88.6 percent of what I’ve said is true.