Users specify their preference for the age of a potential partner as a range. To facilitate comparisons with the user's own age, however, I took the midpoint of that range as an approximation of the age they seek. Figure 2.4 shows the distributions for men and women of the difference between the user's age and the age they sought. Women indicated they were seeking men older by a median 2.5 years than themselves (mean = 3.0). Similarly, men said they were looking for women a median 2.5 years younger (mean = 2.3).
In fact,men actually contacted women who were a median 4.0 years younger (mean = 4.2, s.d. = 6.3)., while women initiated contact with men who were a median 2.0 years older (mean = 1.8, s.d. = 5.3). Across all exchanges, the man was a median 3.0 years older than the woman.
So, women overestimated how much older they wanted men to be, contacting younger men than they initially claimed they were seeking. Men, on the other hand, underestimated how much younger they wanted women to be; they contacted younger women than they initially said they were seeking.
Age Fitness Metric
Age sought relative to self
Figure 2.4. Self-reported age sought relative to user's own age, in years.
To better assess the impact of age preferences on communication between users, I synthesized a metric that integrates whether a user's age falls into a conversational partner's preferred range with the self-reported importance of age to the partner, which ranges from 1 (unimportant) through 5 (very important). Specifically, if person A's age falls within person B's preferred range, I define A's age fitness with respect to B as 1.0 * B_age_importance, yielding a value between 1 and 5. If person A's age falls outside person B's preferred range, A's age fitness is defined as —1.0 * B_age_importance, yielding a value between -5 and —1.
Figure 2.5. At left, a graph of how well the Recipient's age fits the Initiator's preferences. At right, how well the Initiator's age fits the Recipient's preferences. Initiators pay more attention to how well the Recipient's age fits their preferences than to how well they fit the Recipient's preferences. These figures depend entirely on the actions of Initiators, because Recipients cannot control who contacts them.
Users who initiated conversation were naturally more attentive to how well their recipient's age fit their preferences than to how well their age fit their recipient's preferences. Thus, the age fitness of the receiving person (Person B) with respect to the preferences of the initiator (Person A) had a median value of 3 (mean 2.30, s.d. = 2.37), whereas Person A's fitness with respect to Person B's preferences was lower, with a median of 2 (mean 1.23, s.d. = 3.23).This makes sense, because Person B did not choose to be contacted by Person A. Figure 2.4 shows the distributions of fitness scores for initiators and recipients.
Continue reading here: Bounding Characteristics
Was this article helpful?