FAYETTEVILLE, Ark. – With an estimated 40 percent of the 100 million U.S. singles trying online dating, researchers at the University of Arkansas caution users that some Web sites’ claims of scientific justification may be “junk science.”

Psychology professor Jeffrey Lohr and two psychology graduates, Aimee King and Deena Austin-Oden, analyzed several leading dating Web sites and found that promotional claims were more self-serving opinion than legitimate psychological science.

The researchers explored the advertising tactics of matchmaking that existed long before the invention of online dating sites and the Web. For example, many of the matchmaking sites use anecdotes and personal testimonies in advertising in the hope that consumers will accept the endorsement as fact.

Consumers must be cautious of customer satisfaction testimonies in online dating advertisements because the matchmaking sites have most likely pre-selected only satisfied customers, rather than a representative sample, the team said.

Even when the dating services cite scientific evidence, consumers don’t always get all the facts. In an eHarmony comparison, the researchers found that the site neglected to reveal that they compared their couples, who were married only an average of six months (the “honeymoon period”), to couples in the control group who were married an average of two years. The researchers said that opinions expressed during the honeymoon period should not be compared to the opinions of couples after the honeymoon is over.

eHarmony asserts that its matchmaking model is based upon measurement and compatibility. The eHarmony technique is the only one of the leading matchmaking sites that has a patent on its compatibility tests. Thus, eHarmony does not reveal to users the characteristics of their key attributes or the way in which those attributes are related to the people with whom they are matched. 

PerfectMatch.com uses a compatibility system based on “over 35 years of research” developed by “a chief relationship expert,” while Match.com simply claims its method works by the volume of success stories.

Some sites reveal how the compatibility tests work.

The Mate Choice Study is a matching system used by another site, Chemistry.com. Its structure is similar in form and content to the well-known Myers-Briggs Type Indicator, which divides people into psychological types. For the Chemistry.com survey, there are four different personality types a person might be assigned according to how the person responds to the 56-item questionnaire. There are Explorers and Builders, who tend to gravitate toward their own type based on similarity, and Directors and Negotiators, who gravitate on the basis of complementarities.

 Lohr read some of the research analyzed by the leading expert of Chemistry.com, anthropologist Helen Fisher.

“The validity of the Chemistry.com matching strategy is on the threshold of scientific legitimacy,” Lohr said.

The team also researched to see if online relationship services give the appearance of matching in the more traditional way – work, church, the tennis court or bars. The outcome?

“The evidence is not yet in,” Lohr said.

Relations initiated online may not be as enduring as traditional face-to-face relationships because there are few barriers to breaking up when the cost exceeds the rewards of the relationship, the team said. And another mate is only one click away and easy to find.

Many Web sites make claims that they cannot substantiate. For instance, Match.com claims that they are responsible for “twice as many marriages as any other site in the world.” The site measures success according to the number of marriages. However, Match.com does not use divorce to measure failure and thus cannot offer scientific research to support the usefulness of their claim.

The researchers referenced four specific variables that determine the likelihood that two people will come together and form a relationship in the traditional face-to-face way: proximity, physical attractiveness and attitudinal similarity, a sense of rapport and similarities or self-disclosure.

The Internet has minimized the importance of proximity, making it possible for distant strangers to get to know each other through computer interactions.

Also, many of the Web sites do not allow users to see a picture of the match in the initial stages of the face-to-face communication, taking the second variable of interpersonal relationships – physical attractiveness – out of the picture.

Self-disclosure, the final variable that determines the likelihood of a relationship forming, can be deceiving, although advocates of Internet dating claim that the anonymity of computer-mediated communication accelerates intimacy through increased openness about aspects of the self. Critics of cyber-dating say that self-disclosure online is often less honest because of the increased opportunities for identity manipulation.

University of Arkansas alumni Austin-Oden and King were both students in an applied psychology research course with Lohr. Austin-Oden and King, a former honors college student, wrote papers about online matchmaking and evaluated it using the critical analysis techniques that Lohr taught. After reading both of the students’ papers, Lohr suggested that they continue the research and collaborate in compiling the information.

It took a total of two years to complete the project. The result is an article in Volume 15 of Skeptic Magazine with the students evaluating the scientific procedure of some of the major online matchmaking sites.

“It was a great experience for me personally, because I plan on pursuing a research career and submitting many manuscripts for publication,” King said.

Being able to evaluate research, compile papers, collaborate with others on projects and work with reviewers during the submission process during my undergraduate career was a good first step, King said.

“It has prepared me for graduate school and my future career in academia,” she said.

Austin-Oden said that the class helped her be more skeptical and proactive about what she believes in advertising.

Lohr is a professor of psychology in the J. William Fulbright College of Arts and Sciences at the University of Arkansas.