During signup, the player has to specify his or her gender. It has been noticed that the freedom of gender selection may lead some players to experiment with gender roles. A survey on 8,694 players of the Massive Multiplayer Online Game (MMOG) Everquest found that 15.5% players are gender-swapped, 17% of the males and 10% of the females.
In Pardus, players can anonymously mark others as friends or enemies. Private Messages (PM) are the prevalent form of communication in the game. Communication events, friend and enemy markings, trades, attack and bounty placements are recorded as networks. The edges in the friend and enemy networks are binary whereas in the case of networks like communication, trade etc. the edges are weighted proportional to the number of private messages sent, trades executed etc.
The quantitative study of gender-specific behavior and network organization in a society of humans engaged in a virtual world of an online game led to some pretty interesting observations.
- Men take more risks but females are on an average wealthier.
In Pardus, players can earn virtual money in the currency of credits by economic actions such as trade or by attacking others (aggression), collection of bounty placed on other players etc. It is evident that females accumulate significantly more wealth than males. Male players on the other hand experience significantly more deaths due to more risk-taking and/or aggressive behavior. This points at a much larger engagement of females in economic, rather than destructive activities.
- Females attract positive behavior.
Females are much more active in performing actions such as making friends, writing private messages or initiating a trade. At even higher significance levels females receive positive actions with respect to males: being marked as friend, receiving a private message, and being on the receiving end of a trade than their male co-players. Unfortunately, for all negative action types males are both more active in initiating and receiving than females.
- Males are heterophiles but females show homophily
Homophily is the tendency of individuals to associate and link with similar others. A straightforward way to measure homophily in the muti- plex data is to compare the numbers of directed links between all gender-combinations in all networks (MM, MF, FM, FF) to the corresponding numbers surrogate networks where the gender of nodes is randomized. Female to female trading and communication links are over-represented link types. Similarly, male to female trades and communication are also strongly over-represented. However, female to male trades and PMs are much less substantial. Male to male trades and communication are highly under represented.
- Gender differences in networking
1) Males respond fast (slow) to female friendship (animosity) initiatives.
2) Females have 15% more trading and communication partners than males.
3) Female trading networks show 25% higher clustering coefficient than that of males. Also, the clustering of female friendship networks is significantly higher than those of males showing a preference for stability.
4) Males prefer well connected communication partners. The communication partners of males have more communication partners than the communication partners of females. The typical enemy of a male player has more enemies than the typical enemy of a female.
5) Females spend a lot of time in reciprocating links.
It was found that links between pairs of males are less likely than expected by chance. This tendency might be caused by the male-dominated game environment in which females may be treated far better than males, which is a known phenomenon, and therefore also receive more messages. The only way to determine to what extent MMOGs are good models for real society is the direct comparison of the MPNs in games with those observable in the real world. The evolution of network measures and network growth patterns are shown to strongly overlap with data from real-world societies.
1. M. Szell, S. Thurner: How women organize social networks different from men, Scientific Reports 3, 1214 (2013)
2. Stoica, A., Smoreda, Z., Prieur, C. & Guillaume, J. Age, gender and communication networks. In NetMob - Analysis of Mobile Phone Networks 2010 communication proposal (2010).
3. Leskovec, J., Backstrom, L., Kumar, R. & Tomkins, A. Microscopic evolution of social networks. In Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, 462–470 (ACM, 2008).