eHarmony Blog.com

Do you like this article? Post a comment on Facebook

More on eHarmony Blog

What others are reading right now

Trackbacks & Pingbacks 1

1. From Class action lawsuit should signal a call to action for eHarmony · eHarmony Blog on 05 Jul 2009 at 2:52 am ¶

   [...] in eHarmony? Read our eHarmony Troubleshooter. Got something to say? Post a new discussion.The class action lawsuit initiated last April by Californians Lynda Kelly and Miranda Soegi alleging eHarmony’s lack of a [...]

Comments 5

1. SincerelyEthical wrote:

   It needs to be understood that a “scientific finding” only requires a Very low percentage of Reliable predictiveness. It’s the ‘reliability’ which makes it ‘scientifically proven’. For example, a “scientific finding” may involve a matching system which “reliably” predicts an outcome only 10% of the time. Would you pay hundreds of dollars if they said their system’s predictive reliability is, for example, under 10%? Obviously not! There is no way of knowing whether or not eHarmony’s matching system’s offers reasonable reliability without divulging their research data and having it reviewed objectively by a qualified and unbiased party. So, until eHarmony releases their research data, we shouldn’t assume any reliable predictiveness. The seemingly very low ‘success rate’ based upon the huge number of claimed membership is irrelevant to whether or not it is scientifically viable.
   If eharmony truly had data supporting a high percentage of reliability, they would proudly announce it to the world. Their secretiveness is likely proof that the system’s predictive reliability is embarrasingly low, so why tell the world if you’re making crazy money selling it?
   It is important to know that mutual filtering has been proven to be superior to algorithms when used to predict success of romantic love. This is because of the irrational triggers of love, which cannot be generalized through statistical assumptions, and vary from individual to individual. Bottom line: What works for matching Tim and Jen may not work for matching Joe and Sarah. That’s why mutual filtering, with subtle questioning to get closer to a person’s true unconscious desires and truer self description, is a much better predictor of a good match for a particular person than algorithms. As a side note, there is no such thing as a definite number of “dimensions”… There could be less than 29, or many more than 29, depending on how you slice the pie… We need to know the percentage of reliable predictiveness of their system, and not be hooked by the many marketing ploys they use. That’s the only way we’ll ever know if it is worth trying or not.

   Posted 19 Jun 2009 at 10:42 am ¶
2. Fernando Ardenghi wrote:

   eHarmony Labs could not prove couples married by eHarmony have a more stable and satisfying marriage, and very low divorce rates than other couples married by other methods as the control group.

   “scientifically proven” should mean the online dating site offering a compatibility matching method can prove its matching algorithm can match prospective partners who will have more stable and satisfying relationships than couples matched by chance, astrological destiny, personal preferences, searching on one’s own, or other technique as the control group in a peer_reviewed Scientific Paper.

   All actual online dating sites offering compatibility matching methods are like placebo, because
   * they are only fueled by big marketing budgets.
   * they have less or at least the same precision as searching on one’s own, in the range of 3 or 4 persons compatible per 1,000 persons screened, when calculating compatibility between prospective mates.
   * That problem arises because they use:
   a) simplified versions of personality traits, instead of the 16PF5 or similar -complying with IRT (Item Response Theory)- with the complete inventory, 16 factors/variables, as established by Dr. Raymond Cattell in 1949.
   b) inadequate quantitative methods to calculate compatibility between prospective mates, like eHarmony which uses Dyadic Adjustment Scale or other sites which use multivariate linear / logistic regression equations o other equations.

   SincerelyEthical, you wrote:
   “It is important to know that mutual filtering has been proven to be superior to algorithms when used to predict success of romantic love. ”
   Please cite source, because my own experience since 2003 is: with mutual filtering, you can achieve a range of 3 or 4 persons compatible per 1,000 persons screened.

   You do not need to prove a piston engine when you need a jet engine to break sound barrier.

   piston engine: in the range of 3 or 4 persons compatible per 1,000 persons screened, like mutual filtering.

   jet engine: achieve a whole precision of the compatibility system in the range of 3 persons compatible per 100,000 persons screened, 100 times better than any actual online dating site offering compatibility methods.

   The next generation of serious online dating sites offering compatibility matching methods
   - should express/indicate compatibility between prospective mates like 92.55033557% +/- 0.00000001%
   - will have to reach at least:
   3 most compatible persons in a 100,000 persons database,
   12 most compatible persons in a 1,000,000 persons database,
   48 most compatible persons in a 10,000,000 persons database,

   The only way to achieve the eighth decimal is using analysis and correlation with quantized patterns (a high precision quantitative method to assess similarity between quantized personality patterns, a part of pattern recognition by correlation)

   Regards,

   Fernando Ardenghi.
   Buenos Aires.
   Argentina.
   ardenghifer@gmail.com

   Posted 19 Jun 2009 at 4:38 pm ¶
3. SingleGuyInNC wrote:

   I do agree that to make the claims they are making, they should be publishing data. As for subjecting the masses to statistics/confidence intervals, it isn’t going to make them any more confident in the service. They need that data interpreted for them by letting the software determine thresholds. The average person signing up for these services doesn’t care about what goes on in the black box and the statistics. They just want the outcome.

   The other issue with determining success is that you don’t have complete participation. I don’t think that even they don’t know how well it truly works because they end up dealing with only a small sample of people that bother to communicate with each other.

   The longer I’m on eH, the more I continue to think it is just another way to meet people and not to take it so seriously. Matching two people is not an easy task. You can get some things close but in the end, it is up to the many intricacies of their personality and how they mesh together that I think would be very hard to acquire, short of a complete brain-dump and analysis.

   Posted 19 Jun 2009 at 9:54 pm ¶
4. SincerelyEthical wrote:

   On most sites, you react to attractive photos of a member, and you think to yourself, Wow, this could really work! Then you learn about their personality and you say No Way…

   On eHarmony, you might find out a little more about the person, and you think to yourself, Wow, this could really work! Then you see their picture, and say No Way…

   It’s 6 of one, half dozen of the other…

   Posted 20 Jun 2009 at 6:22 pm ¶
5. SingleGuyInNC wrote:

   Very true. During FCWs, when I wasn’t a member, there were a number of matches that I thought would be pretty good when I read their profiles. Upon subscribing again, I found that I wasn’t attracted to their appearance, which is, if we believe everything NCW says, one of the key dimensions. I tend to believe that you have to be close to each other on the “numbers” or it doesn’t work out.

   eH used to not do photos, way back before I was a subscriber. I don’t know how folks had much success back then. Maybe it was who was on the service (the very few) that made that approach work.

   The flip-side to this is that I’ve met folks who had what I thought was a great profile and pictures that were nice. I met them in person and the pictures were not a reflection of what they looked like “on an average day” and the personality was off the charts in the wrong direction.

   On the other hand, I met someone who I thought was “so-so” in their photos but was very attractive in person.

   The whole thing is hit or miss. Photos can be very misleading in either direction. This is why I aim to move to meeting in person ASAP before too many expectations are created and not met. I find that on the other sites, what folks write about themselves tends to be very shallow, so you HAVE to meet them to dig deeper and figure out what they are about.

   Posted 21 Jun 2009 at 1:48 pm ¶

Post a Comment