(Image credit: MercuryNews.com)
The epic Star Wars series is chock full of life lessons. Good versus evil, the struggle to claim one’s true power, nature versus technology. But what can the beloved sci-fi adventures from a galaxy far, far away teach us about data science? So, so much. Here are the Top 5 most influential lessons in data science that BlastPoint has learned from Star Wars!
Warning: May contain spoilers for the Disney+ Series, The Mandalorian!
5. Getting units to agree is one of the greatest challenges in data, especially when use isn’t consistent.
“It’s the ship that made the Kessel run in less than twelve parsecs. I’ve outrun Imperial starships. Not the local bulk cruisers, mind you. I’m talking about the big Corellian ships, now. She’s fast enough for you, old man.”
—Han Solo, Episode IV, A New Hope
Units of measurement only mean something when the people using those units agree on what they stand for. If we think about things that rely on accurate amounts (money, doses of medication, food servings…) in different, incongruent ways, we can end up making costly or even life-threatening mistakes.
Take the Mars Climate Orbiter that got lost in space in 1999. The unit discrepancy there? The spacecraft’s designers used the American measuring system to build the thing, while the team in charge of navigating it to Mars used the metric system. It’s little wonder, then, that its acceleration system went berserk: the navigators were measuring the force of its thrust by newtons per square meter while the designers had measured it in pounds. It was an expensive mistake that could have been avoided.
Just as Han Solo does in our Episode IV example above, where parsecs equal 3.26 light years—not length of time—we can’t go relying on different units of measure and expect accurate results.
4. Probability is hard for humans.
“Never tell me the odds!”
—Han Solo, Episode V, The Empire Strikes Back
We humans are emotional creatures. It’s hard for our delicate ears to hear terrifying statistics such as, “There’s a one in 5.4 million chance you will die in a plane crash!” (source: The Economist). Or, “There’s a 4 percent chance your baby will have a birth defect!” (source: Emory University School of Medicine).
Stark numbers like these put our sensitive feelings on the defensive. Just like Han Solo in The Empire Strikes Back, we’d rather take our chances than know just how many awful things *could* happen to us at any given time.
For data scientists, it’s worth remembering this as we crunch numbers and analyze statistics to present to customers. Explaining why such stats are relevant puts analytics into context, making them easier for humans to process and understand.
3. AI is not the same as human cognition.
“Well, if droids could think, there’d be none of us here, would there?”
— Obi-Wan Kenobi, Episode II, Attack of the Clones
Artificial intelligence is intricately woven into the fabric of our lives. We see it more and more each day, in targeted online advertising, at our doctors’ offices, in local fitness facilities, and beyond. So it’s tempting to assume A.I. can make more accurate predictions and take more effective action than humans can.
But, as Obi-Wan Kenobi explains in Star Wars, Episode II, where droids are lousy in warfare because they are, by design, slow, incompetent, and therefore expendable, the Separatists would not exist if droids had human brains. People (the Separatists) wouldn’t stand a chance if droids had human capabilities. They’d be overpowered and outnumbered.
Thankfully for Obi-Wan (and thankfully for us humans), neither Droids nor A.I. are creative beings. They don’t learn new things or improve their skills—not without human intervention, anyway. So we cannot rely solely on them to carry out our missions, whether those missions are meant for good or for The Dark Side.
2 Self-reported data can be unreliable.
“Han Solo: Uh, everything is under control. Situation normal.
Intercom: What happened?
Han Solo: [flustered] Uh, had a slight weapons malfunction. But, uh, everything’s perfectly all right now. We’re fine. We’re all fine here, now, thank you. How are you?”
—Han Solo, Episode IV, A New Hope
Have you ever fudged the truth while filling out a questionnaire at your doctor’s office (e.g., when asked, “How often do you consume alcohol?”)? Or, have you ever accidentally typed the wrong address or phone number into an online form?
If so, welcome to the majority. People misreport personal information all the time. Whether they simply forget relevant details or report falsehoods intentionally, as our friend Han Solo did in the Episode IV quote above, we need to remember not to rely on every detail people tell us as being gospel.
Self-reporting bias is of particular relevance in medical research. People may provide answers to survey questions based on what they think their doctor wants them to answer, or on what they think is socially acceptable, rather than what’s actually true.
Alternatively, people often mis-remember details. If you’ve ever applied for child abuse clearances, for example, you may have been asked to provide every address you’ve lived at and with whom you’ve shared residences throughout your entire life. It may have been difficult, then, to recall with accuracy what your street number was when you shared a room for one semester in an off-campus apartment with “that one dude” during sophomore year of college.
Or, like Han Solo in this scene from Episode IV, just after he, Luke Skywalker and Chewbacca decimate the Imperial troops in the detention block, we may wish to purposefully withhold delicate information so as not to anger those receiving it.
In order to avoid self-reporting bias, we remind ourselves to ask timely questions that don’t require information from years past. And we focus more heavily on questions of preference and opinion rather than asking sensitive information that people may wish to keep private.
1. Don’t Make Assumptions About Customers Based On Age
“Truly wonderful, the mind of a child is.”
—Yoda, Attack of the Clones
What’s in a number, really? And how relevant is a person’s age, anyway?
After all, not all seven-year-olds love unicorns (some are obsessed with the sinking of the Titanic instead). And not all 65-year-olds are inept at using smartphones (app designers, take note: There’s a viable market for smartphone apps that appeal to folks over 65, so don’t count them out when building yours).
Yet, as marketers, software engineers and researchers, we are often tempted to squish people into a certain stereotype, especially if it means we’ll get our products into the sales pipeline faster.
As Yoda teaches us in Attack of the Clones, despite his vast experience and centuries-long existence, even the young have something worth teaching the old; and even the student can instruct the teacher. It’s important for data scientists to remember this when designing software or writing algorithms to analyze people.
Traveling backward in Star Wars time to the new Disney+ series, The Mandalorian, we learn more about age bias from the irresistibly cute Baby Yoda, a.k.a. “The Child.” Don’t you just want to pick the little guy up and bounce him on your knee? Not so fast! The guy’s 50 years old. He just happens to be born to a species that ages very slowly.
Here on planet Earth, it’s good to remember there are plenty of people who don’t fit into stereotypes. That’s why we treat everyone as unique individuals.
What’s your favorite Star Wars quote? What life lessons have you learned from Luke and the gang? What insights about data science has Star Wars taught you? We’d love to hear! Drop us a line through email to let us know.
*Full credit goes to BlastPoint CEO and Co-Founder Alison Alvarez, whose genius and creativity generated the idea and laid the foundation for this post!