Widgets Magazine

Tweets by @Stanford_Daily

RT @StanfordSports: Stanford's no stranger to PKs. Another scoreless game came down to sudden death as @StanfordMSoccer takes the title: ht…: 8 hours ago, The Stanford Daily
RT @StanfordSports: Two straight saves by goalkeeper Andrew Epstein make @StanfordMSoccer back-to-back national champions!: 11 hours ago, The Stanford Daily
RT @StanfordSports: After 10 rounds of penalty kicks in the College Cup semifinals, @StanfordMSoccer advances to the final for the second y…: 2 days ago, The Stanford Daily

OPINIONS

Sad stories lead to bad policy: Endemic intellectual laziness and the disdain for data

Most people like stories more than data. Research shows that people will donate more to charities when they’re told about one individual’s suffering than a whole group of people in pain. They remember narratives, not numbers. Stories can change minds where data just invites disbelief or apathy.

When interacting with most audiences, then, the common advice is to make an emotional, narrative appeal rather than dwelling on facts and figures. This cliché is likely to be trotted out in any presentation-teaching class at Stanford. And it’s sound advice; stories are far more fun, easier to remember and more evocative than data. Perhaps most importantly, one can find a story to fit almost any point one’s trying to make. That makes stories far more convenient than data, which generally require at least some facility with statistics to mold to one’s will.

And that’s a problem. If we take a broader perspective, we’ll notice that every plausible large-scale policy will have benefits and drawbacks. If one goes looking for a story about how a policy took away someone’s livelihood or tore a family apart, odds are she will find it. Stories do a lot of emotional work, but the work isn’t very thorough or balanced. They don’t necessarily prove anything that we didn’t already know: Big policies have real tradeoffs.

For example, it is absolutely true that vaccines do occasionally have adverse side effects. Very rarely, those side effects kill people, including innocent children. This is tragic. Other innocent children die because they don’t have access to vaccines, which is also grievously sad, and also proves nothing in itself, without the accompanying data about how children are affected by vaccinations or the lack thereof. And this is an easy example; most people are aware that vaccines save lives, and extreme side effects are rare. There are likewise multiple examples of gun owners using their guns to save their loved ones, and gun owners and their children accidentally killing themselves or others. For more obscure issues like minimum wage legislation and drug policy, the results of scientific research are often even more overshadowed by the many gripping stories on all sides. But without accepting the moral importance of numbers over narrative appeal, none of these stories suggest much about what any of us should do. And they force most of us to avoid or downplay other stories that our opponents employ, which is uncomfortable and intellectually unhealthy.

To be clear, our point is that we should be suspicious of stories rather than embracing them, not that all storytellers are incorrect or intentionally deceptive. Almost everyone with an agenda (or objectives of any sort) uses stories to appeal to others, regardless of their political affiliation and good or bad intentions. Almost any class with a unit on making arguments will teach students to use pathos, not logos. And that’s a sound and intelligent decision, at least from an individual perspective. Stories work, and students that are taught to avoid them may well lose out to students with less epistemically scrupulous instructors, creating an arms race-like situation even for those who would rather use more representative methods.

The plural anecdote is not data. The qualities that make something a good anecdote (being extreme, novel, or unusual, and thus memorable) usually make it an unrepresentative datum, so a simple aggregate of those anecdotes will likely look nothing like actual data free from these biases. For example, some have speculated that especially controversial or complex examples of a phenomenon are more likely to be reported on in the media, because they provide more fodder for those on all side of the debate to keep arguing. This was a proposed explanation for why Michael Brown’s death in Ferguson, Missouri initially received more media coverage than Eric Garner’s, even though Garner’s death occurred before Brown’s.

On a socio-cultural level, we all stand to lose. Policy debates seem to have become story wars, narratives tossed back and forth with new names and dates affixed. Senator Inhofe can toss snowballs and Bostonians can jump out their windows into the snow, and the solution isn’t to add in more stories about an especially sad California farmer suffering from drought. Having a greater number of true stories won’t help, as long as there are enough to switch them out as they become old news. On issues where there are powerful stories to be told and few new technological developments to break a stalemate (abortion, gun control, etc.), this trading can go on indefinitely. Using scientific data seems to be a better way to win a debate than just shouting louder, or better, and we need to reward those who avoid these cheap shots.

But, an approach focused on data rather than individual stories is often considered cold or calculating. Garnishes of hokey narrative examples are demanded by PWR teachers, audiences, admissions committees and a host of other, potentially important figures of authority. However, the calculated use of the intimacies of people’s lives, trimmed and sterilized to fit a simplistic narrative, is actually far more cold and calculating than trying to represent experiences thoroughly and without cheap appeals. Manipulating peoples’ life stories just to make a point seems more degrading and dehumanizing than using reason and information to figure out how to help as many people as possible. And spoon-feeding students and other audiences tale after tale rather than trying to communicate a situation fully seems like a strange way to show them respect, or help our society collectively move towards the truth.

That’s not to say that remembering, acknowledging and representing the emotional and narrative content of people’s lives is always inappropriate, or that data is never used callously or disrespectfully, or any other highly hyperbolic version of a point we’ve made. Rather, we seek to highlight the fact that at the university undergraduate level, at least at Stanford and likely outside of it, there is shamefully little ethical pushback against the use of stories instead of science to try to prove points about the world. Any one individual who avoids storytelling will probably do worse than she could, but it’s unfortunate that our institutions don’t at least have the sense of shame to tell students exactly how much we’re sacrificing just to get people to hear what we have to say.

Contact Claire Zabel at czabel ‘at’ stanford.edu and Joseph (Joey) Zabel at joezabel ‘at’ stanford.edu.