Thea was once an online dating success story. The petite blonde in her early 40s was one of the first people I knew in New York to join an online dating site — Swoon.com, which she joined in 1998. Everyone else at the time found the idea of online dating and mating alienating, but the adventurous theater producer felt she had no other choice.
“I worked in off-Broadway at the time — the only men I met in real life were gay,” she jokes.
Outgoing, vivacious and curious, Thea loved online dating. She loved being able to fine-tune her searches, and learned how to zero in on the red flags and green lights in a man’s profile. She got good at writing a good online profile that emphasized her wit, intelligence and fun-loving nature. Men flocked to her — her flattering photos also helped.
The result? A fun, angst-free dating odyssey that lasted for about a year. It ended when Thea met and married Peter, a creative, smart, educated arts professor who loved to cook and travel, on an online date. Online dating brought Thea to the man of her dreams.
Or so she thought. Fifteen years later, Thea and Peter split in a messy divorce. The couple looked good on paper — but in real life, they were missing a certain essential spark.
Now, Thea re-enters the dating game online nearly a decade and a half after she first tried it. But the new romantic landscape bewildering and almost indecipherable. Tinder is “too young,” and she dislikes the hookup vibe of OkCupid and Plenty of Fish. Match yields too many bland results, and eHarmony hasn’t worked out as well.
She’s also questioning her own judgment on what she should be looking for and how to find it, especially online. “Clearly my approach before was effective to some degree,” she says. “But I feel like I’m missing something I’m not seeing or getting — and that something will make the difference.”
Thea’s missing X-factor is about to be filled in, thanks to the reams of data already gathered through dating and social networking sites. Data scientists are sifting through this trove of so-called “big data” to figure out the gulf between what we think we say and want — and what we actually do that works. The results are surprising — and should help you find love, no matter where you go to find it.
Thea got back into the game by joining Match. Looking to repeat her success, she filled out a solid profile and got professional-quality photos taken. According most conventional wisdom, the best profiles have good pictures, primarily one where you smile and make eye contact with the camera, as well as some of you participating in your interests.
Beyond sharp visuals, Thea also followed advice to keep things zippy, specific and brief, and keep attention focused on positive sociable interests. (According to Wired, the most popular interests mentioned in online dating profiles were yoga and surfing.) Most people approach reading online dating profiles with the question of “What will this person bring to my life?” To guarantee some response, you want to seem dynamic, happy and interesting. Thea made sure to include photos of her on her travels, as well as pursuing her cultural interests in art and music.
Thea also made sure to avoid negativity in her profile and her overall communication on dating sites. Bashing your exes is a no-no in a dating profile. You want to avoid cheaters, liars, no-good wastrels and gold-diggers, but stating it in your profiles comes across as overly negative or demanding, even if you soften it with a joke or a winking emoji. And remember the basics: look over your profile for spelling and other mistakes, since bad grammar is a big turnoff for smart people.
For Thea, her new profile yielded a promising beginning crop of dates, mostly with other distinguished, older artsy divorced types. She seemed on the right track to finding love again during her Dating 2.0 foray, positioning her to join the millions of people in the U.S. whose relationships began online. According to the University of Chicago, one-in-three marriages between 2005 and 2012 began online. Partners in marriages that began online reported higher levels of satisfaction as well.
Some, however, questioned the data — the University of Chicago study was commissioned by dating site eHarmony, and one of the researchers is a scientific advisor to the company. Some think online couples aren’t as successful as their offline counterparts, in fact — a joint Michigan State and Stanford study found that couples who met online had significantly higher separation and divorce rates than offline marriages.
Thea worries about being one of those online couples with a higher divorce rate again. While she enjoyed her initial series of dates, she felt “something was missing.” Online dating, with its ability to select, screen out criteria, brought her what she thought she wanted: a professional, career-minded man with a passion for the arts she seemed compatible with. But she couldn’t shake the feeling that she was dating “another version of my ex-husband, who, as wonderful as he was, had certain tendencies that made me feel lonely, even when we were together.”
Thea did some soul-searching: she realized she wasn’t gun-shy, had hidden commitment issues or was being too picky. A romantic optimist, she refused to believe being older was an obstacle to finding love.
But she did start to believe that she wasn’t using online dating to its best capacity — she needed new criteria. “You can lavish the most care on a profile, have the most beautiful photos, do everything right that you’re supposed to — but maybe there’s something about the online model that doesn’t work,” Thea speculates. “Like, are we even asking the right questions to begin with, besides what are your interests and what are you looking for? There’s this huge database at your fingertips, but maybe we’re just not sifting through it right.”
Questions about mutual interests and the like aren’t actually the right ones to ask when it comes to online dating. Often in relationships, we have a gap between what we think we want versus what actually works in practice and in real life.
This insight isn’t coming top-down from psychologists, relationship therapists or other experts — it’s being generated from us. Thanks to the surge of online dating as an acceptable, widespread social option, one-in-ten U.S. adults have tried OkCupid, Match and eHarmony, among others — and happily answered quizzes, tests and detailed questionnaires in the search for true love. According to the BBC, Match, for example, has more than 70 terabytes — or 70,000 gigabytes — of data about its customers. Sites can gather not just user-generated data, but also track clicks, messaging rates and other interactivity, gathering evidence of what we’re attracted to and what really inspires responses.
In the process, we’ve created a pool of random, anonymous data for scientists to analyze. Thea wasn’t just looking for love when she jumped on the online dating bandwagon early on — she was participating in one of the largest social-data experiments of her time.
Scientists can slice and dice this pool of raw data, using sophisticated statistical models and algorithms to parse out trends and patterns that elude human understanding. Computers may not understand emotions, exceptions or the simple quirkiness of being a human, but they can take in a wider, broader swath of information and find patterns that may otherwise evade human detection and biases.
What can this big data tell us about relationships? Mainly, we value the wrong things when it comes to romance. Take the issue of men, women and age. On dating sites, women tend to rate slightly older men as attractive until about age 30, when they begin to favor men closer to their own age. But men, no matter their age, rate women in the mid-20s as the most attractive, according to NPR, making women in the 20s age range by and large the most popular, competitive demographic.
Yet if you take a closer look at more of the data — and what actually works — we see that higher success rates in terms of responses and dates actually go to men who seek out women closer to their age. If you’re a man in his late 30s or early 40s, then, you may find a 25-year-old female attractive on a site, but you’re much more likely to get a response from a prospect if you message someone in the early-to-mid-30s range.
This dovetails with the most recent research looking at relationships and age. Focusing on a specific age or income level, as many people do with online dating, isn’t a fruitful strategy for finding a life partner, according to Notre Dame.
Instead, finding a match in a number of categories — such values, looks, age, and level of accomplishment — works better. When looking for a lifelong mate, it’s best not to look for an ideal trophy that makes us look good or boosts our status — stereotypically, a tall, rich, handsome man or a young, attractive woman — but someone who is already at a similar level as you.
Yet when it comes to the criteria we use on dating sites, we often rely on looks, age, occupation and status first to filter people out first. It also doesn’t help that people often lie online on these seemingly crucial factors — men, for example, traditionally fudge their heights and incomes, while women often finesse their ages, weight and builds. If significant amounts of people lie in their online dating profiles, what use are traditional search criteria, anyway?
What does work when it comes to finding love online, at least according to the computers analyzing our big data? It’s more important to look at how we actually behave online, OkCupid founder Christian Rudder wrote in his book, “Dataclysm: Who We Are (When We Think No One’s Looking).”
His idea is that big data shows us who we really are when we think no one is looking, measuring or judging — and also cuts through our own tendencies to both lie online and misjudge ourselves. (You might say you’re a classical music fan, for instance, but an analysis of your iTunes, Spotify and Pandora activity could reveal you’re actually into another genre entirely.) You can say one thing, but your actual clicks reveal something else.
Big data can be scarily accurate in guessing who you really are, judging by your clicks and likes. For instance, thanks to the data users happily give up about themselves on the network, Facebook can predict with 88 percent accuracy whether or not you’re gay, based on your likes, according to The New Yorker. The company can also tell whether your parents were divorced before you turned twenty-one within 60 percent accuracy.
Sometimes the picture gleaned from the data isn’t pretty. Take the sticky issue of race, for example: by and large, when asked, many people say race doesn’t matter when it comes to their romantic preferences.
Yet when it comes to the actual data, according to Rudder, race is very much a dividing line, particularly against black people, who suffer from a high bias against them on dating sites. When it comes to how people rate them as prospects, how often their messages are replied to and how many messages they get, black online daters get the short end of the stick. We give lip service to the idea of a color-blind world, but judging by their online actions of clicking, responding, winking and the like, most Internet daters don’t want to date black people.
Other results are surprising, counterintuitive and often funny — helping us find a match in more unexpected, yet reliable, ways. According to Rudder, for example, if you’re looking for a one-night fling on OkCupid, the best predictor is whether or not a user answers yes to the question “Do you like the taste of beer?” Beer lovers were most likely to say yes to sex on the first date.
If you’re looking for something a little more enduring, however, you might want to pay attention to how someone answers the questions “Do you like scary movies?” and “Have you ever traveled alone to another country?” According to Rudder, three-out-of-four couples that came together on OkCupid answered both questions the same way.
This piece of information seems merely fun and quirky at first, but when investigated further in light of the latest findings on research in relationships, it makes sense. According to Ty Tashiro, a University of Maryland professor who wrote “The Science of Happily Ever After: What Really Matters in the Quest for Enduring Love,” evolutionally speaking, looks (for women) and income (for men), once worked as good relationship filters during an age when marriage was about economic survival and meeting basic needs like food and shelter.
But now that we partner up for happiness — and must make marriages last during our longer lives — we need to judge people on compatibility. Compatibility, according to Tashiro, isn’t just sharing interests or even having similar beliefs and lifestyles, though these are important. It’s how you match up based on three personality traits or temperaments: agreeableness, lack of neuroticism and lack of seeking novelty. We need to be aware of deeper appetites and personality issues: our attitudes, tolerances and appetites for solitude, stability, novelty, change, adaptability and intensity.
Whether or not you like scary movies, then, might be a concrete, real-life measure of our approaches to novelty and excitement: if we love adrenaline and seek out constant excitement, we might be better matched with partners that share or understand the same need. However, if we dislike horror movies — and prefer stability, peace and reliability — we’re better off with less high-octane partners.
And questions about solo travel indicate our taste for adventure as well as our need and desire for solitude: someone who needs a lot of independence and alone time, for instance, might find a cozier partner clingy and claustrophobic.
By focusing on traits and temperament, we place less emphasis on things we might initially judge as deal-breakers. For example, disparate political views were not as much of a relationship red flag as many think, according to OkCupid — what matters in a couple is how passionate and important politics are to each partner in general. If your partner shares your same passion for politics, regardless of what side of the fence he or she is on, he or she is a better match in general than someone who could care less about Democrats, Republicans or social justice in general. Same goes for religion: sharing the same religion might matter less than agreeing on the importance of it in general in your life.
“People tend to overemphasize the big splashy things: faith, politics, and certainly looks,” Rudder told Elle. “They don’t matter nearly as much as everyone thinks.”
In the end, long-term romantic success is much less about finding a perfect partner than creating a good relationship with the partner you choose. According to the Michigan State-Stanford study, longer-lasting relationships tended to endure because of what social science researchers call “higher relationship quality.” It’s important that we feel accepted, understood and safe in our intimate relationships. Do you feel appreciated or taken care of in the way you like? Do you laugh easily or argue without fear of being judged too harshly? Do you feel comfortable, like you can be yourself, on a date?
OkCupid also echoes the importance of authenticity and connection when it comes to finding love is the simplest: the best advice, in the end, is to be you. According to the site’s data, people who rated conventionally attractive to a wide set of users based on traditional factors like looks actually didn’t have the highest rate of success on the site. Instead, the people who tended to polarize opinion — people who could incite both strong like and dislike, thanks to their genuine, expressive profiles — fared best in terms of responses and overall relationship success.
It’s no use to twist yourself in a pretzel trying to appeal to the largest swath of users as possible — what’s important to be authentic and express yourself fully. Only then can both people in a couple judge whether or not they’re a genuine match.
As Thea re-enters the online game, she says she’s decided to take less of a “checklist” approach to online dating. She made the effort to learn more about compatibility and relationships and is retooling her searches in light of this new information. She realizes she and her ex were compatible in interests, but not in terms of Tashiro’s criteria: she needs someone with a stronger sense of adventure and novelty, but more easy-going and less neurotic in general.
The new approach is trickling down to her actual conversations and dates. Instead of asking shallow questions about work and jobs during messaging and initial communicating, she says she wants to hear about her dates’ latest trips, or what they’ve learned from their last relationship.
And even her first dates are different. “I’m trying to get beyond dinner and a movie,” she says. “Trying to find something new and adventurous for both of us to do, and to see how we work together as a team.”
Thea recently went ice skating on a first date with a woodworker and carpenter who lives in upstate New York — not the artsy, cosmopolitan type she initially would have picked for herself earlier. But when he messaged her on Match, she was intrigued by how much he liked to read — and how he built his own bookcases from wood he sourced himself in Bali. “He’s a working-class type, but well-read and likes to explore and travel,” she says.
More importantly, he seems to share her taste for novel experiences and openness to experience, as well as a sense of gaiety. “Both of us hadn’t been ice-skating in years, so we fell down a lot,” she reports. “But we laughed it off together and had fun. Now we’ve got plans to go skiing, and we’re kind of talking about taking a trip somewhere in a few months.”
Thea doesn’t know if her new beau is a perfect match, but as long as she feels comfortable and is having fun, she’ll stick it out. Despite all the big data out there — and the insights it can gather about who we are and how we love — Thea is beginning to think the idea of a perfect match is a fallacy, anyway.
“I see love as having a mutual sense of mission now,” she says. “Maybe it’s adventure, maybe it’s family. What feels more important is feeling like you’re in it together, have each other’s back and can talk and laugh about it every now and then. Everything else is kind of just noise.” ♦