Online dating

Online dating signifies the range of matchmaking services offered on the Internet. Couples matches are usually search-based, providing users the ability to generate potential partner lists based on preferences such as age, location, ethnicity, and beliefs--from the mundane to the devout. A combination of computer dating, personal advertisements and the trademark Web collaborative spirit, Internet dating services have altered the face of matchmaking, its public perception and its techniques.

Online dating represents the single-most popular subscription service on the Internet. One study showed that eHarmony alone accounted for 2% of US marriages in a year, that is, 120 marriages per day. More than 40 million Americans have used online dating services and 120,000 marriages per year are attributed to dating matches from these sites. In 2006, US residents spent over $500 million and studies show spending will continue to increase.

History

The origins of online dating can be traced back to 1959 and two Stanford students. Jim Harvey and Phil Fialer were known for their notorious hijinks around campus. On the college radio show they ran, for instance, they were scolded for broadcasting at excessive power as well for playing radio programs that were less than tasteful. Stanford had recently begun renting the IBM 650 mainframe computer (with an instruction speed of 22-33 ips, for $3,750 per month.) As part of the programming course for which Harvey and Fialer registered, they were able to run their final algorithm on the new IBM computer. Their project consisted of matching 50 males students with 50 female students according to similarity on their responses to a survey. A brute force algorithm, testing all students against each other, the program computed difference scores from which could be determined the best matches. Only scheduled a short time period (due the cost of computer time) for each program, Fialer and Harvey were able only to run their program for a small percentage of students. However, being the students they were, the pair broke into the computing labs and ran the program for all students. They held a large party for the match-dates, however no long-term relationships developed. Harvey and Fialer received an A in the course.

Early experiments such as these occurred at a few campuses until 1965, when Harvard students Jeff Tarr and Merrill Vaughn launched Operation Match. Operation Match is considered the first commercial computer dating venture. Tarr and Vaughn were Pychology majors and developed a more sophisticated matching system than was Harvey’s unscientific “seems reasonable” method, though their approach was questionnaire-based as well. They hired a programmer for $100 to code their algorithm and charged student $3 for 5 guaranteed matches. By 1966, Operation Match was well-known among Ivy League campuses and brought Tarr and Vaughn minor fame. Their company was featured in Look Magazine, a widespread periodical, and they received radio attention as well. Computer dating services, modeled after Tarr and Vaughn’s Operation Match thereafter became a fad of the 1960s and 70s .

Jim Harvey's logical flowchart for Operation Match

Another fad that influenced online dating was personal advertisements. Popularized in the 1960s, these ads offered individuals seeking relationships space to contribute self-descriptions that would appear in the classified sections of newspapers. Video personals were an updated form of newspaper personals that gave a much more intimate aspect to the advertisement. Video personals utilized the spread of VHS and in this way foreshadowed online matchmaking's use of technology for dating consumerism. They would also foreshadow participants’ pursuit of more and more pre-knowledge about a potential date.

Research has also shown that a number of the online dating trends were anticipated by communication on bulletin board systems in the 1980s. These systems allowed single users at a time to post both personal and private messages as well as maintain profiles. America Online and Earthlink began in the bulletin board form. Although these systems did not formally offer dating services, numerous individuals attempted to court through the bulletin boards. Gender roles that would later be studied in depth in online dating sites can be shown to have existed here. Males’ descriptions of desired qualities in potential partners were generally more physical-based. Females’ descriptions tended more toward the emotional and the material, and generally the relationship on a whole. Like bulletin board systems, newsgroups served a similar function as unofficial dating forums.

The first online dating service (match.com) was created in 1995 to migrate newspaper personals online and it is currently the largest, with about 15 million users .

Popularity

Dating has altered drastically since its inception. It once consisted of first date proposals and generally the family head making deals with other family heads about whom their daughters will marry. But recently in America, mostly in the past century, dating has become more process-oriented than ends-oriented; during this process daters weed out individuals who they feel are not marriage potential even if the discarded numbers are large. Even more recently, the combination of these generational changes resulted in one type of service that caters to all types of daters, the Internet dating service, which gained prominence as traditional avenues closed.

By now, "Americans are marrying later and so are less likely to meet their spouses in high school or college. They spend much of their lives at work, but the rise in sexual harassment suits has made workplace relationships tricky at best. Among a more secular and mobile population, social institutions like churches and clubs have faded in importance. That often leaves little more than the bar scene as a source of potential mates."

Certainly, the popularity of online dating goes hand and hand with something else as well. Around the turn of the 21st century, a different image of the Internet began to be portrayed. Computers gained social capital, and altered as the tool of computer geeks and advanced research to be a new cosmopolitan (if escapist) social sphere. Peer-to-peer file-sharing programs, Napster the most prominent, were widely used. The proliferation of low-cost pornography, which is actually regularly acknowledged as the impetus towards numerous major Internet developments (e.g., broadband video, webcams) spread as well. America Online’s chat rooms and instant messaging capabilities also radically altered the Internet’s perception. By the time these major trends had taken effect, online dating was in a position to become the hot trend that it has.

Given their prevalence, it seems likely that online dating systems have begun to influence not only individual lives but also cultural notions of love and attraction with their overflowing catalogues of potential partners and their sometimes idiosyncratic choices of personal characteristics to highlight.

Algorithms

Online dating has turned searching for possible partners similar to other e-commerce websites, like online shopping. To enable this, many different algorithms have been created. Each online dating company boasts about their superior technology to show and create better matches. All possible approaches incorporating psychological to mathematical theories have been used.

Though each company may take a different approach, matching personalities is broadly divided into three stages which all companies use :

• Collection of User Information
• Matching Users
• Communication between Users

Collection of User Information

This step starts the moment a person joins an online dating service. Some people refrain from using online dating services for this very step of sharing private information about themselves. The data is amalgamated to create a 'user profile'. Such a profile is a webpage that provides information about a user and can be viewed by the other members of the dating service. The users indicate various demographic, socioeconomic, and physical characteristics, such as their age, gender, education level, height, weight, eye and hair color, and income. The users also frequently answer a question on why they joined the service, for example to find a partner for a long-term relationship, or, alternatively, a partner for a “casual” relationship. In addition, the users provide information that relates to their personality, lifestyle, or views. The data collected are basically of two types: characteristics data and criteria data. Characteristic data essentially describes the user; while the criteria data is what the user expects out of a possible partner.

Services collect data via various types of questions. There are some direct fundamental ‘yes’ or ‘no’ questions which all services use; ranking questions, where a user must rank various choices according to his preferences; questions where a user is given two possible extremes and he/she may choose anywhere between one extreme to the other; and indirect questions which calculate personality compatibilities. Some sites also offer options for users to ask their own constructed questions and provide self-definitions. Match.com and Yahoo! Personals use this form of data entry to describe themselves and what kind of person they want to meet.

Once all this data is collected, it has to be statistically converted to allow algorithms to generates matches.

Matching Users

The most basic matching is done by the users themselves, when they search the database manually. This gives a lot of control to the user, but it is tedious. So to do this automatically, all the responses of the users are converted into compatibility scores (which is the relative strength of each match) to mathematically match individuals. Services look for both complementing and similarity between users to be matched. Some services, like eHarmony.com, do not let users search the total database manually, and only allow users to search "compatible matches." However, match.com, Yahoo! Personals and other services do allow users to manually search the database.

All dating services claim trade secrecy of their particular matching algorithm. However, all matching consists of these vital steps that all the services incorporate. Firstly, the data is saved into the database and converted to numerical scores. Compatibility scores are then calculated. A range is then formed around a person’s compatibility scores to see which other users fall into his or her range. If a person is easy-going then his range will generally be higher than a difficult person. If there are fewer matches in a user's compatibility range, then he or she may be asked to change some answers to questions from the previous ‘user information collection’ stage, to get a higher volume of potentials.

During this phase, mathematical correlation is used too. Various types of factor analysis may be performed. All characteristic and criteria data are considered factors. All factors are given different mathematical values and then depending on how important that particular factor is for the user, it is multiplied by a weight.

Once potential matches are listed, communication between users is facilitated.

Communication between Users

Communication is started at a maximum privacy-protected stage or communication level. The users sometimes do not even reveal their names. Pseudonyms are regularly used. Instant messaging, emails, private voice messages, etc. are the commonly used tools for communication. Once the users show enough attraction to each other, they enter another level of communication where they reveal more of themselves. This process of communicating at different levels continues until at a very close level of communication the users decide to meet. At any communication level users may also decide to end contact if they so choose.

Marketing

One major reason why some dating sites are more successful than others is their marketing strategy. There are endless possibilities for market segmentation. For example, numerous users claim that eHarmony.com implicitly targets Christians who are looking for a relationship with someone with shared spiritual beliefs. This company is fairly well-known due to their unique marketing orientation. eHarmony has a fairly narrow market segment, and is considered more financially successful than most other sites. On the other hand, online dating services such as Seek4Love.com are perhaps not as well-known because they have not developed a specific target market, and are considered to be a more generic dating service. Thus, profitability appears to be more directly correlated with the ability to customize market segment while still taking advantage of the convenient broadcasting aspects of the Internet.

Consumer Spending for Online Dating.

Some companies, such as LavaLifeknow their target market and air commercials during television shows that their market is said to watch. This is a well-developed strategy, because they know who they are reaching and when. Internet matchmaking services show that knowing one's target market can be a key element to a company's success, and it also helps to build a stronger customer relationship between the business and the consumer.

Present Trends

• Niche dating sites have become incredibly common and they are expected only to increase in popularity.
  • Online Martimonials - Very common to eastern cultures where marriages between individuals must fulfill some socio-religious conditions. Shaadi.com, India's premier matrimonial website boasts 10 million members and a million marriages in 11 years. Also see Zawaj, Bharat
  • Religious-oriented sites, though less formal than matrimonials. See JDate, ChristianSingles.
  • Sites that offer short-term arrangements such as one-night stands and cybersex. These run the gamut of the sexually mild to the most adventurous. LonelyWivesAffairs.com
  • Sites that cater to gay, lesbian, bisexual and transsexual couples and encounters. See PlanetOut.com
• Virtual Dating – Incorporates the use of avatars in online encounters. Research has shown that interactive meetings are more strongly correlated with successful experiences. See OmniDate.
• Cellular Dating – Check profiles and request meetings over a cell phone connection. See Mobilove.
• Handwriting Analysis – Uses handwriting samples to assess compatibility. See HandWritingProfiles.
• DNA Testing – Preference-matching based on genetics. See ScientificMatch.
• Web 2.0 Technologies –Allows family and friends to browse for potential matches on dater’s behalf. Rating technologies are also being incorporating, such as date politeness and overall quality. Facebook and MySpace are already beginning to enter the online dating arena as well.
• Background Checkers - Perform checks to verify not only criminal background but personal profile information. See HonestyOnline.
• Code Builders – Script kits and hosting tools to build own sites. See SourceCodeWorld.

Issues

• Success:

It is questionable how successful online dating really is. One 2006 study found that 29% of online dating users reported pejorative experiences. Some studies also show that “the percentage of people browsing those sites is stagnating.“ Other studies have shown that online dating users typically only have face-to-face encounters with a miniscule percentage of the profiles they browse. One dating consultant states though, it could just be that the lookilous have gone away. Perhaps so, as the above statistics do not discount online dating’s success as much as dating in general’s difficulty.

• Reliability of Algorithms:

The scientific integrity of matching algorithms is only suspect at best. Scientific American has shown that, for instance, eHarmony’s algorithm is neither proven reliable (“roughly, that you can count on it to produce stable results”) nor valid (leading to successful matches). Additionally, no data from any online dating site algorithm has been submitted to academic journals for peer-review. .

• Safety and Illegality:

Online dating can lead to illegal and personal safety issues. Another major concern with these services is the divulgence of massive amounts of private information in profiles and online communication. Mildly, cases of identity theft have occurred. Of course more serious victimization presents itself as well, such as stalking and child predation. There have even been cases of online dating being used as a tool for prostitution. In Japan, online dating has been tainted by allegations of teenage prostitution through the services. The practice is known as enjo kosai, or ‘compensated dating,’ the victims offer sexual services in exchange for money or gifts. According to Japanese police, nearly 800 crimes in the first half of 2002 were related to online dating.

• Asymmetric Information:

Users cannot accurately determine the qualities and subjective attributes of a potential match through examination before matching is made. However, all potential matches can more accurately determine their own subjective attributes prior to matching . There are also problems relating to direct fallacies and information gaps. For example, “many online daters lie about their marital status."

• Adverse Selection:

Adverse selection describes a situation in which the mere fact that a good is available suggests its undesirability – the classic “market for lemons” problem. Consider a new car’s resale value, which plummets as soon as the car leaves the lot. Just as a prospective buyer has to wonder why the previous owner wants to sell, those of us fortunate enough to find ourselves online in a romantic pursuit might ask, "If this match is so great, why isn’t he already taken?"

• The Lake Wobegon Effect:

The Lake Wobegon Effect, is the human tendency to overestimate one's achievements and capabilities relative to others; this is also called the above-average effect. In the case of online dating, this would be a consequence of adverse selection and information asymmetry. Daters tend to lie on their profiles or steal other profiles to avoid adverse selection, which only exacerbates the effect. Interestingly, it has been found that 69% of men and 73% of women reported better than average looks on dating sites, and only 1% admitted to being below average.

• Lack of Subjective Knowledge:

Dating sites use a “shopping” interface like that used by other commercial sites. Thus, like any commodity, daters are classified by different searchable attributes like height, weight, age and race, which can be filtered in any way the shopper desire. The problem with this approach is that romance requires subjective knowledge (e.g. since of humor) about the potential match and not only objective knowledge, which the online dating experience a bit disappointing for some users.

• Social Stigma:

Though recently mitigated, the social stigma of online dating still exists. As has always been the case for matchmaking services, the question presents itself, "why can I not find love in the traditional way"?