OKCupid founder Maxwell Krohn speaks before the Harvard Humanist Community. OkCupid is one of many sites that claims to have cracked the code on online dating. (RNS photo by Dominick Reuter)

Does online dating actually work?

It’s a simple question and a common one — one whose answer could determine the fates of both a multi-billion dollar industry and millions of lonely hearts. It’s a question that seems distinctly answerable: we have user data, surveys, clear metrics for success or failure, entire books full of colorful charts.

And yet, just this week, a new analysis from Michigan State University found that online dating leads to fewer committed relationships than offline dating does — that it doesn’t work, in other words. That, in the words of its own author, contradicts a pile of studies that have come before it. In fact, this latest proclamation on the state of modern love joins a 2010 study that found more couples meet online than at schools, bars or parties. And a 2012 study that found dating site algorithms aren’t effective. And a 2013 paper that suggested Internet access is boosting marriage rates. Plus a whole host of dubious statistics, surveys and case studies from dating giants like eHarmony and Match.com, who claim — insist, even!! — that online dating “works.”

This much should be obvious: We don’t actually know.

Some of the reasons for that ambiguity are clear in this latest study. For starters, there’s this greater cultural issue of how we define relationship success: Is it marriage? Is it monogamy, a la Patti Stanger? Is it what OkCupid’s data team calls a “fourway” — four messages back and forth between two semi-interested parties? That’s a tough one to parse, and different studies have defined it different ways. (This one, for the record, looked at marriages and other long-term relationships; if you’re not looking to tie the knot, its conclusions aren’t for you.)

Then there’s a sort of secondary issue in how we define a site’s actual function, because despite the marketing hype, that isn’t clear. Most paid sites claim, for instance, that it’s their highly scientific matching algorithms that lead people to serious relationships; in his 2013 book on the subject, however, the journalist Dan Slater concludes that most of those claims are bunk. (“Everyone knows that all personality profiling is bull****,” a former Match executive told him. “As a marketing hook, it works great.”)

In reality, dating sites are most effective as a kind of virtual town square — a place where random people whose paths wouldn’t otherwise cross bump into each other and start talking. That’s not much different from your neighborhood bar, except in its scale, ease of use and demographics. But in terms of actual function, the things we think of as uniquely “online” in online dating — the algorithms, the personality profiles, the “29 dimensions of compatibility” — don’t appear to make too much of a difference in how the enterprise “works.”

Meanwhile, all this is happening during a time of enormous revolution in the way we conceive of relationships and commitment. A record number of Americans have never been married, and only a scant majority — 53 percent — want to be. Americans get married later every year, if they choose to get married at all. Women habitually stay single into their 30s and 40s, a tidal shift in how they viewed commitment even one or two generations ago. And while reliable data on sexual partners is hard to come by, there’s some suggestion that modern singles get around more than they used to.

Surely online dating has fed this trend in part, providing the constant buffet of alternative options that sociologists say plays a large part in determining whether a relationship fails; but at the same time, apps like Tinder could never have caught on if people weren’t already approaching sex and dating more casually. It’s a bit of a chicken-or-egg problem: maybe online dating has made us more cavalier, or maybe our growing casualness fed online dating, or maybe these things both exist together in a miasma of hook-ups and right-swipes and shifting social standards.

“On Tinder everything’s disposable, there’s always more, you move on fast,” one Tinder-user told the Guardian Monday, explaining how the app had single-handedly transformed her from a serial monogamist to a hook-up artiste. And yet, by the end of the interview, she’s off Tinder and in a relationship with a guy she met on the app. Who really had the agency there: the dating app, or the dater?

It’s a question that applies equally well to offline dating, too: When a relationship fails, what or who is ultimately responsible? The place where the couple met? The length of time they took getting to know each other? Or something squishier, something less precise — a factor not captured in charts and telephone surveys?

After all, 2.1 million people get married in the U.S. every year, and half of those couples will divorce. If we parsed their fates according to the exact venue in which they met, or any other number of arbitrary factors, we would probably turn up the same kind of confusing, self-contradicting results that research into online dating perennially seems to.

But those contradictions wouldn’t be blamed on the Internet — we’d credit the vagaries of the human heart. For some reason, no one’s content to see online dating that same way.