Dr. James Houran has accepted Dr. Joel Block’s challenge. With several caveats.
Update: Dr. Block responds to Dr. Houran in the comments for Personality Testing Smackdown. Wow, just go read it and come back.
Dr. Houran says:
…my team is confident we can build an original customized application that’s shorter and more entertaining test than your questionnaire.
However, we don’t work for free. The estimated cost to combine all of the features described above in a new application, to meet APA testing standards, and being ready within approximately six months will cost approximately USD $350,000.00…. We will require full payment before the project will begin. We can also accept a wire transfer.
I’m sorry, did I miss something? Dr. Houran is taking the shootout and trying to turn a profit. He’s completely ignoring the original shootout idea- put your best test up against each other and see which one is more effective (among other factors.)
Nobody said anything about the length and amusement factor of tests. We’re talking about finding out which matching system is most effective.
It gets better. Dr. Houran is on a role here:
Your dismissive, cavalier attitude about science is neither consistent with the messaging you have on your site nor with your clinical education and credentials. As an academic, I find it appalling and bordering on unethical. I’m even tempted to contact all professional organization to which you belong and notify them of the situation.
So Dr. Houran wants to charge a competitor $350,000 for a personality test and also report him to the authorities? Now I’m really lost. Someone wrote me
Dr. Houran says that his team has developed several proprietary matching systems on some of the biggest dating sites. I wish he would talk about this more. What sites? Why not share the results from these sites? Why not use an existing site for the shootout?
There really is no need for things to get this complicated. That’s what happens when ego’s and professional reputations are at stake. All the academic-speak, personal attacks and self-important banter is tiring. Dr. Houran has effectively ended the conversation on his own terms.
That’s too bad, because Lynne Sandler has a fantastic idea:
Here is my off-the-cuff idea for a “real-life†challenge: Get a sample size of singles. Each single takes all of our personality tests and receives lists of ranked matches from each of us. Put them in a room for a huge speed-dating-like event. Have them meet and rank their dates on score cards. Let them tell us which system introduced them to people they liked and want to see again. We can have periodic follow-up interviews to measure success.
Brilliant. This is much more doable. Easy to set up, inexpensive and will provide some real-world results. I met Lynn over the summer and we had a great talk. I really like her and her perspective on matching people. I like her even more after hearing her idea for a challenge.
As a last resort, I can always get some volunteers to sign up for dating sites and follow a protocol and report results.
It’s too bad consumers don’t care about matching systems, if they did they would vote with their mouse and their wallet and this conversation would be completely different. Dating sites would be touting their efficiency levels in their marketing campaigns, just like drug companies, and constantly evolving their algorithms to adapt to what they learn along the way. AFAIK eHarmony is the only company that actively tweaks their system.
Turns out the dating industry is just as messy and organic and chaotic as dating itself.
No related posts.
{ 18 comments… read them below or add one }
I do not want to be annoying but the Butterflies_Again approach has 2 terribly weaks points
1) TEST
Butterflies_Again_Compatibility_Test: proprietary IPSATIVE instrument based on FIRO B.
Serious online dating sites offering compatibility matching methods should be based on NORMATIVE tests.
2) COMPATIBILITY / How compatibility results are displayed
Butterflies_Again indicates compatibility/affinity between prospective mates with only a 2 integers figure like 68%, 70%, 75%.
It seems some persons do not understand that compatibility between prospective mates expressed as an integer figure like 68%, 70%, 75% indicates “as low as” 3 / 4 persons high compatible per 1,000 persons, so in a 10,000,000 persons database, you have as many as 30,000 / 40,000 persons to contact/date, that means a WHOLE PRECISION LESS than anyone could achieve by searching on one’s own!!!!
30,000 persons is the population of an average small city!!!
You do not need to prove a piston engine when you need a jet engine to break sound barrier.
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,
Kindest Regards,
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
I’ve been struggling to understand the role of personality testing in online dating. I mean, at first look you’d think you can match up people by common traits, but is that really how relationships work? Are people really looking for their long lost twin? Or, is there also a focus on complementary personalities rather than similar ones? I don’t want to dismiss the idea of matching on personality but based on experience and what I’ve seen with people I know, I never thought personality matching could be that important.
Keith, some systems match people based on compatibility, which can include the concept that opposites attract. Matching is all about efficiency. You would know your partners worldview, personality traits and tastes soon enough just by spending time with them. Matching supposedly systems bring that information to light in the beginning.
Personality matching is a tool, plenty of people get married that don’t look good “on paper.”
Hi Keith,
The research is clear — complementarity matching is superior to strict similarity. Overall, romantic partners show strong similarity in age, political, and religious attitudes; moderate similarity in education, general intelligence, and values; and little or no similarity in personality characteristics. Thus, you’re right in that matching people based on similar personality profiles is not effective.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
That is exactly the kind of useful information I’m sure people want to hear more about. I’d like to hear more about the differences between tests on dating sites. Avoiding the science for the layperson, but explaining how PerfectMatch differs from True, Chemistry from eHarmony. Can consumers become educated enough to say “I like sites where I’m matched in a certain way, so I belong to service abc?”
Yes,
A table could be constructed that outlines the different tests and on what variables they match. But variables are only part of the equation — science also has to come into play. Some tests are simply more psychometrically sound than others (i.e., their internal quality). Consumers should ideally be educated on level of psychometics supporting the test and the variables that the test purportedly measures. For example, consumers may like/prefer to be matched on personality, but the science may not support that approach and the personality test used may be also of poor psychometric quality.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
Nice job. There should be some sort of simple color-coded identifiers that makes it easy to differentiate between sites.
Give people the opportunity to drill down into the details as far as they want to go, all the way down to Fernando-land :-)
I’ve always wanted to make a simple questionnaire to help people decide which site to try.
“What are your expectations”
“How much are you willing to spend”
“What is your sexual orientation”
“Where do you live”
Add “What’s most important when it comes to compatibility with someone” or something similar.
Talk about tests… I actually scoped out an online tool that did this automatically. It would ask several questions similar to ones you mentioned and more and then it would make specific dating site recommendations as well as specific tips and guidelines for getting the most from that particular website (knowing that dating agendas can and often do differ across sites).
I think people would get a great of use from something like this.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
Whats the url?
Just to clatify…
Tests and assessments are both gauges or measures of underlying traits or characteristics. Both can use a myriad of different question types to establish trait levels. The only real difference is that with assessments there is evaluative feedback or clinical/practical application. Tests simply give descriptive “grades” or scores, not neccesarily evaluations of those grades per se. That is what an assessment does.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
There’s no URL as the tool doesn’t exist outside of an outline of scope and business requirements.
But if any company or organization wishes to sponsor the costs involved in building it, my team would be happy to bring it to life.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
I have what you are talking about pretty much built out here on my hard drive, have played around with the concept for years. Doesn’t take a team, takes a programmer and I few weeks of after-hours coding to knock it out. There is no science involved, doesn’t need to be.
If someone were to think about hiring your for a project, is there a url people can visit to learn more about your team’s capabilities and experience?
Dave,
I’ve no doubt that simple versions can be made easily. However, I’m talking about a tool that goes beyond simple demographic questions. The tool I envision measures a broad range of other salient attitudes, needs and traits. Thus, it’s akin to a psychological assessment and is grounded in scientific measurement. There are many advantages to the consumers who receive reports based on this approach — it can gauge hidden idiosyncrasies about which even the user is unaware. This allows us to give much more specific and relevant feedback and guidance than would be possible otherwise. Plus, it uses a computerized adapative testing (CAT) approach, so new questions are raised based on a user’s responses to previous questions. This type of stuff is well beyond the capabilities of typical programmers.
Potential clients almost always contact me via email, telephone or through OnlineDatingMagazine.com to discuss customized applications. I then share information that’s pertinent to a client’s needs. My group doesn’t advertise per se.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
Last February 2005, using a couple-centered approach, Drs. Klohnen and Luo wrote in a paper “People may be attracted to those who have similar attitudes, values, and beliefs and even marry them (at least in part) on the basis of this similarity. However, once individuals are in a committed relationship, it may be primarily personality similarity that influences marital happiness. This suggests that attitude and value similarity may play a different role in relationship development than personality similarity does. For example, whereas similarity in attitudes and values appears to be important early on in the relationship and may play an important role in relationship progression, personality similarity becomes more important as the relationship reaches greater commitment.
………..
future research designed to better understand these underlying processes is needed.”
In August 2005, Dr. John A. Johnson told me, “There are probably undiscovered factors beyond similarity of any type that determine relationship quality.”
“Opposites attract but then they attack“, Dr Buckwalter’s dixit
Kindest Regards,
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
“Opposites attract but then they attack” — this is not surprisingly an eHarmony mantra. They use it in media interviews and such to justify their rigid similarity matching methods. The fact is that people with strong similarities also can and do “attack” each other. Long-term couples use both similarity and complementarity variables to maintain a relationship.
And relationship maintenance (stability and satisfaction) is not dependent upon personality — many studies contradict the notion that personality similarity is the key. The key actually resides in cognitive psychology: the tendency to view positive perceptions as more important than negative perceptions in the relationship, as well as the tendency to alter the importance of specific perceptions as needed. Couples simply (or not so simply) use their unique pattern of similarities and complementarities to justify staying in a relationship. Moreover, leading edge research can predict which issues men versus women will seize upon to build up or tear down a relationship.
In simpler terms, relationship success is grounded in accommodation, not “physical chemistry” or “personality similarity.” For more on this, I refer people to:
Houran, J., & Lange, R. (2004). Expectations of finding a ‘soul mate’ with online dating. North American Journal of Psychology, 6, 297-308.
Thanks,
James Houran, Ph.D.
http://www.OnlineDatingMagazine.com
Hi Dr. Houran:
I could argue that the word similarity means different things for different persons, it depends on how you exactly mathematically defines it!
e.g.
http://www.plentyoffish.com/personality_faq.aspx
“…
For example look at the graph below that depicts two hypothetical couples with each factor shown horizontally (X-axis) and the scores vertically (Y-axis). These two couples can be seen similar or different, depending on which criterion is being used. The data of the couple shown in the top section are very similar scores (i.e., the values are roughly the same), but their test results (i.e., the overall shape of the lines) bear no relation to each other. By contrast, the data of the couple in the bottom section are very different (i.e., one person is ‘high’ overall, whereas the other is ‘low’), but their test results are very similar.
GRAPH
http://img100.imageshack.us/my.php?image=pofcpsimdisfh8.jpg
The preceding thus illustrates that depending on which type of ‘similarity ‘is being used, quite different results are obtained. In the top portion of the graph the two people are similar because the two people are both ‘average’ in Confidence, Openness, Easy Goingness, Self-Control and Family Orientation. The pair in the bottom section is ‘similar’ because both people score highest on Family Orientation and lowest on Openness – i.e. the two sets of scores are ranked about the same.
…
Also, we found that using both types of similarity simultaneously better captures the people’s individual characteristics and the dynamics of a couple. Therefore, our similarity algorithm combines the two types of indices in a proprietary fashion. ”
Using the high precision quantitative method I had invented to compare similarity between quantized patterns:
The patterns of the couple shown in the top section have a high degree of similarity between them
and
The patterns of the couple shown in the bottom section have a very low degree of similarity between them
exactly opposite as POFCP calculates similarity!
By the way, Dr. Houran, could you disclose all the tests/assessments/instruments used in actual online dating sites you had constructed/co_authored/peer_reviewed ?
Like:
True Compatibility Test (constructed / co_authored),
True Sexploration Test (constructed / co_authored),
Cybersuitors (improved, re_constructed using IRT),
Chemistry (peer_reviewed)
WeAttract’s Compatibility Test used in Yahoo!Personals (peer_reviewed)
POFCP PlentyOfFish Chemistry Predictor (constructed / co_authored)
PlentyOfFish Needs Assessment (constructed / co_authored)
Kindest Regards,
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
Hi Dr. Houran!!!
I could argue that the word similarity means different things for different persons, it depends on how you exactly mathematically defines it!
e.g.
http://www.plentyoffish.com/personality_faq.aspx
“…
For example look at the graph below that depicts two hypothetical couples with each factor shown horizontally (X-axis) and the scores vertically (Y-axis). These two couples can be seen similar or different, depending on which criterion is being used. The data of the couple shown in the top section are very similar scores (i.e., the values are roughly the same), but their test results (i.e., the overall shape of the lines) bear no relation to each other. By contrast, the data of the couple in the bottom section are very different (i.e., one person is ‘high’ overall, whereas the other is ‘low’), but their test results are very similar.
The preceding thus illustrates that depending on which type of ‘similarity ‘is being used, quite different results are obtained. In the top portion of the graph the two people are similar because the two people are both ‘average’ in Confidence, Openness, Easy Goingness, Self-Control and Family Orientation. The pair in the bottom section is ‘similar’ because both people score highest on Family Orientation and lowest on Openness – i.e. the two sets of scores are ranked about the same.
…
Also, we found that using both types of similarity simultaneously better captures the people’s individual characteristics and the dynamics of a couple. Therefore, our similarity algorithm combines the two types of indices in a proprietary fashion. ”
GRAPH
http://img100.imageshack.us/my.php?image=pofcpsimdisfh8.jpg
Using the high precision quantitative method I had invented to compare similarity between quantized patterns:
The patterns of the couple shown in the top section have a high degree of similarity between them
and
The patterns of the couple shown in the bottom section have a very low degree of similarity between them
exactly opposite as POFCP calculates similarity!
By the way, Dr. Houran, could you disclose all the tests/assessments/instruments used in actual online dating sites you had constructed/co_authored/peer_reviewed ?
Like:
True Compatibility Test (constructed / co_authored),
True Sexploration Test (constructed / co_authored),
Cybersuitors (improved, re_constructed using IRT),
Chemistry (peer_reviewed)
WeAttract’s Compatibility Test used in Yahoo!Personals (peer_reviewed)
POFCP PlentyOfFish Chemistry Predictor (constructed / co_authored)
PlentyOfFish Needs Assessment (constructed / co_authored)
Kindest Regards,
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
Fernando,
Just for the record, Cybersuitors uses the CQ (Compatibility Quotient) Test devised and validated by Dr Glenn Wilson and I.
You are right to note that James Houran and his team subjected our data to further welcome investigation, but it is not correct to say that the CQ Test was improved or re-constructed in any way following their analysis.
Thanks
Jon Cousins