Match.com Collects Data of Love


Dating Website Match.com released its second annual survey on singles last week. Findings raise the question of whether personal data can help predict relationships. In the survey, Match.com sampled 5,000 US singles ages 21 to 65+. The company claims it to be the most comprehensive of its kind ever conducted.

In Analyzing Love: Data Mining on Match.com, I blogged about how the company mines the data created by user interactions on its site. This time, the company teamed with biological anthropologist Dr. Helen Fisher, evolutionary biologist Dr. Justin R. Garcia, and sex and relationship expert Dr. Laura Bermanto to delve into preferences and motivations.

Some may remain skeptical about using data to predict the success of personal relationships. But the data reveals some clues about why singles may choose to enter relationships and how satisfied they are within those relationships based on a variety of factors.

For example, the data suggests men are more likely to fall in love at first sight, more likely to make a commitment either without being in love or without feeling sexually attracted to a partner, and more likely to make public displays of affection.

In addition, it shows that political affiliation may determine the qualities singles seek in partners and how satisfied they are once in a relationship. Liberal Democrats are more likely to seek someone with a sense of humor and similar lifestyle while conservative Republicans are more likely to be searching for someone from the same background and political party and one interested in marriage, the data shows. (As an interesting side note, conservative Republicans apparently outdistance all other groups in being most satisfied with their sex life while married.)

We might ask whether we could use this data, and much more like it, to predict future relationships between singles. Would it be possible to determine happiness and stability of relationships in advance? (Match.com gives no clear indications in the release of what it intends to do with the survey results.) Some additional questions arise.

Can we trust the accuracy of the information users share? Obviously, some will raise doubt over data collected directly from, say, Match.com's user profiles, since users may enhance that information to impress others. Survey results seem more reliable since Match.com keeps respondent names confidential.

Also, did those giving the survey ask the right questions? We can wonder how they defined political affiliations like liberal and conservative, for example, or the term "commitment." We also can ask how Match.com defined "relationship" and the characteristics it considered as marking a successful relationship.

What do you think? Can data help predict love? We'll wait to hear from you on the comment boards.

Shawn Hessinger, Community Editor

Shawn Hessinger is a community manager, blogger, social media and tech enthusiast, journalist, and entrepreneur based in Northeastern Pennsylvania. He serves as community manager and blogger for BizSugar.com, a business news and information Website, and contributes regularly to the online business news source, Small Business Trends. He is the founder of PostRanger.com, an online content and media community, and has provided blogging and social media services and consulting for companies all over the world. He researches and writes on a variety of business, Internet-related, and other tech topics including business intelligence and analytics. He is also keenly interested in computer-aided data management as it relates to his various online ventures. A newspaper journalist with more than 11 years experience as a reporter and then managing editor, Shawn began blogging in 2006 and now provides a variety of consulting and outsourcing services in Search Engine Optimization, Web development, and online marketing to companies large and small. He is a strong advocate for the use of BI and related computer data management in business decision making, whether using software as a service (SaaS), cloud, or other applications, and in the opportunity these technologies provide to transform small startups and larger established businesses alike.

BCBSNC, SAS Team on Advanced Analytics

The key to improving heathcare outcomes is to look at individual needs, the companies say.

Spoofing, Privacy Greatest Barriers for Biometrics

In Wednesday's e-chat, we discussed the analytics of identification and whether the technology might find a bigger role one day in marketing intelligence.


Re: Modeling for True Love
  • 2/16/2012 8:41:35 AM
NO RATINGS

Exactly, how can any sort of a technology predict emotions of humans ? I would like to have a look at their algorythem and methodology which they use to predict or forecast.

Re: Modeling for True Love
  • 2/14/2012 9:35:46 PM
NO RATINGS

@Louis, I agree that at least that dating site was saving her her money, but I can tell you that she did NOT take it so kindly! And no, she hasn't had any luck since then. If anybody in Boston is looking for a blind date, I could maybe work it ... and I won't use any algorithms to determine if you are a good match.

Re: Modeling for True Love
  • 2/14/2012 10:30:25 AM
NO RATINGS

It seems to me that Match.com and similar "matchmaking" or dating services attempt to provide a kind of sophisticated blind dating process for people that, for one reason or another, are willing to engage in blind dating. In one respect, it's sort of like using a complex roulette wheel, but people want to think that there's something more than pure chance involved, so the matchmaking services have tried to adopt increasingly more "science" based, sophisticated trappings, eventually evolving into what is today a kind of de facto analytics.

A kind of parallel phenomenon that comes to mind is the application of analytics to economics and human economic behavior. Treacherous ground — to some extent, that apparently underlay much of the big financial and housing market collapse in 2007-2008.

That doesn't mean that professionals shouldn't keep trying to develop and test more data-based and algorithmic approaches to understanding and predicting human behavior — but in this area (especially with hormones involved), it seems to me to be dealing with some really darn qirky, murky, complicated dynamics.

Nevertheless, we keep trying — in public transportation, we've been endeavoring to improve human behavior prediction for decades through very sophisticated mathematical modeling. It's called ridership forecasting.

Re: Modeling for True Love
  • 2/14/2012 2:44:54 AM
NO RATINGS

Since the authors do not have access to the algorithms that purportedly bring people together, the only aspect that they have to question is the input. The bottom line is that regardless of how perfectly the algorithm may actually perform, there is always the human factor. Because people input information and may or may not be honest with that information, that data is never going to be entirely accurate. Match.com may report that they have the most successes in marriage and in relationships, but do they know if the relationships actually last? To go one step further, what is the divorce rate among couples that have been married as a result of meeting on Match.com? I think that may speak more for the dating site as opposed to the claim of more marriages. They romanticize what may not even be a genuine reality, but only a truth at the hands of the data that gets analyzed. 

Re: Modeling for True Love
  • 2/14/2012 12:20:28 AM
NO RATINGS

Very interesting that this discussion is raising the issue of Match.com and other dating services that supposedly try to use something approaching analytics.

The New York Times published an op-ed just a couple of days ago on basically this exact topic:

The Dubious Science of Online Dating

http://www.nytimes.com/2012/02/12/opinion/sunday/online-dating-sites-dont-match-hype.html

Here are some excerpts relevant to this discussion:

===============

But can a mathematical formula really identify pairs of singles who are especially likely to have a successful romantic relationship?

We believe the answer is no. It's hard to be certain, since the sites have not disclosed their algorithms. But — as we and our co-authors argue in an article to be published this month in the journal Psychological Science in the Public Interest — the past 80 years of scientific research about what makes people romantically compatible suggests that such sites are unlikely to do what they claim to do.

...

One major problem is that these sites fail to collect a lot of crucial information. Because they gather data from singles who have never met, the sites have no way of knowing how two people will interact once they have been matched.

...

Another major problem with the algorithms of dating sites is that the information that they do collect — about individual characteristics — accounts for only a tiny slice of what makes two people suited for a long-term relationship.

=================

Anyway, you can get the authors' basic theme here. Whether dating can ever be adequately served by analytics remains debatable. However, it seems to me the authors are highlighting data input issues rather than a failure in the algorithms (which they're not privy to).

Re: Modeling for True Love
  • 2/12/2012 8:58:42 PM
NO RATINGS

@Broadway   Wow !  Well at least they were "honest" and didn't lead her on just to collect fees.  I hope she found someone  !  : ) 

Re: Modeling for True Love
  • 2/12/2012 8:23:55 PM
NO RATINGS

I know of one case where a dating site admitted it could not analytical find love for a user. A friend of a friend once tried out one of Match.com's biggest competitors. She did the whole initial assessment and the dating company denied her application ... saying they could not guarantee they could find her a match because of her depressing disposition (or something along those lines).

 

Re: Modeling for True Love
  • 2/12/2012 12:46:09 PM
NO RATINGS

@ahdand  I agree with you.  I have little faith in Match.com   There are just too many intangibles that cannot be accounted for in any meaningful way.  As you say, whether most want to admit or not, feelings and opinions change from moment to moment.

Re: Modeling for True Love
  • 2/12/2012 11:26:14 AM
NO RATINGS

I still cant belive how they predict love via analytics. Its something which you cannot predict. Changes every minute based on the mood.

Re: Modeling for True Love
  • 2/8/2012 4:03:17 PM
NO RATINGS

@Ariella, some of what I say might be posturing. I am man married and out of the "scene," so to entertain myself I make up stories. But I also am entertained by friends' stories, both female and male friends ... and usually the story centers around men and their games. Perhaps I need to hang out with different friends ; )

Page 1 / 5   >   >>
INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +