Research on prosocial behaviour shows that help‐giving differs between the sexes. Gender role theory posits that males specialise in material aid, while females specialize in emotional support. Today, people increasingly help and support each other via online environments. The purpose of this study is to examine whether the genders of avatars used in online interactions influence help‐giving behaviour in similar ways as physical sex does in face‐to‐face situations.
An empirical study was conducted using a unique observational data set from a Japanese massively‐multiplayer online game. Instances of help‐giving were identified from conversation logs, coded, and analysed statistically to discover differences between male and female avatars and different help types.
Avatar gender is found to influence help‐giving in ways that deviate from expected gender roles: female avatars are more likely than males to provide assistance in the form of material support and labour, and no more likely than males to provide emotional support. Female avatars are more likely to give help to male avatars than other females. The results are explained using behavioural confirmation and self‐perception theory.
Physical sex was not controlled for, but most players were male. The study should be repeated in other environments.
The results suggest that designers can prime users towards prosocial behaviours by tuning the available line‐up of avatars.
The research question and use of observational data are novel. The study is valuable to online educators, companies seeking to reduce customer support costs through peer help, and policymakers dealing with ICTs and societal change.
