In two related articles, Colley & Todd (2002) and Colley et al. (2004) explore whether the sex differences demonstrated in speech and writing transfer into the medium of email. In particular, they reference previous research showing that females utilize more emotional language, measured through the use of indicators of excitability (!!, very) and nonessentials ( ..., "weak"). In addition, the use of hedges (I guess) and tag questions (right?) led researchers to emphasize the "weak" nature of female language. In contrast, male language has been characterized as assertive and competitive, with more opinions given, and the use of longer monologues. These differences were summarized by Tannen as rapport talk (females) versus report talk (males).
Interestingly, these apparent sex differences, common in research conducting in the '70s, have begun to fade in more recent research. Perhaps this change is best explained by the shift in focus to the function of the language act and considering the context in which it is performed. For example, tag questions can function to include the listener in a conversation and are not, in such cases, a sign of uncertainty. In terms of context, the characteristics of both the speaker and the listener, as well as the setting of and purpose for the talk, can influence the type of language displayed. Finally, I'd point out that labeling relation-building language as "weak" is a reflection of inherent cultural stereotypes, apparent even in our own discipline-specific terminology.
In any event, the questions posed by Colley and colleagues are interesting because email, as a more informal medium, could serve to increase the use of feminine language style by both females and males. For example, Colley et al. (2004) argue that, "the informal conventions of e-mails may allow women to express emotional tone" (p. 376). And, Colley & Todd (2002) argue that when writers chose to diverge from (or converge towards) the recipient, increasing the use of feminine language would be an effective way to support rapport building (p. 390). The alternative, increasing or decreasing the use of masculine language would not be effective for this purpose, especially given its "adversarial" nature.
To consider the influence of context on language use in emails, Colley & Todd (2002) manipulated the sex of the recipient of the email message. Specifically, male and female students were asked to write an email to a male or female friend regarding a recent holiday. Note that participants were free to chose any friend of the assigned sex with whom they had an ongoing relationship. However, they knew that this friend would not actually receive the email. In addition, in electing to allow participants to select real friends, the researchers were not able to control the level of closeness to these friends.
Colley & Todd coded several aspects of the language used in these emails. First, elements of style were considered including the use of specific speech acts such as humor, apologies and self-disclosures as well as the use of nonessentials and markers of excitability. Second, the topic of the emails was examined. Finally, markers of intimacy were recorded including the use of personal pronouns (I/me/my/myself; we/our/ourselves; you/yourself), references to the recipient, mention of future contact, initial personal inquires, and affectionate sign-offs. As can be seen in Table 1, several sex differences were identified across these three categories. In addition, females used more second-person singular pronouns (you/yourself) and tended to use more first-person plural pronouns (we/our) than males.
However, the more important findings are the interactions between sender and recipient. In particular, both males and females self-disclosed more, and tended to use more first-person singular pronouns (I/me/myself), when writing to recipients of the opposite sex than to those of the same sex. Furthermore, females used several features more frequently in emails to males including humor, references to the recipient, multiple exclamation points, and nonessentials. Overall, the authors suggest that females displayed three broad features of feminine language use in their emails including interpersonal sensitivity, social warmth, and greater emotionality (p. 388). In addition, they suggest that the language changes seen in emails written by females to males are an indication of a "heightened arousal, intimacy, and desire to engage the recipient's interest" (p. 389).
Colley and colleagues (2004) conducted a very similar study in which they considered the medium of the message. Male and female participants either wrote an email or letter to a male or female friend discussing what they did for the summer. Two aspects of these messages were considered. First, they considered style by examining formality (e.g., incomplete sentences & abbreviations), excitability, nonessentials, and relational devices (e.g., questions). Second, they considered the content of the emails, categorizing the phrases in the emails as activities, relationships, and personal disclosures as well as noting the use of initial personal enquiries, mentions of future contact, and the use of affectionate signatures.
Overall, the emails were more informal in structure than letters (fewer words, more abbreviations, more incomplete sentences, and more multiple exclamations). In terms of sex differences, females used more positive intensifiers (sooo, really), multiple exclamations, initial personal enquiries, affectionate signatures, and expressed more positive emotions.
Once again, however, the interesting results are those that demonstrate the interaction between sender and recipient. Women used more positive intensifiers in emails to males than to females; more men used affectionate signatures in emails to females than to other males. In contrast to these opposite-sex differences, the results also showed that females used more humor and multiple exclamations in their emails to other females. These same-sex patterns are in contrast to those found in the earlier study. However, the authors argue that the difference could be due to the topic of the email--in the earlier study, the participants were focused on describing a place; in the later study, participants were focused on describing their personal experiences (p. 375).
In general, the results of these two studies should encourage researchers to consider situational factors and individual differences when exploring the language patterns used in emails. Of particular interest is the relation between the sender and the recipient as well as the purpose of the email message.
Colley, A., Todd, Z. (2002). Gender-linked differences in the style and content of e-mails to friends. Journal of Language and Social Psychology, 21(4), 380-392.
Colley, A., Todd, Z., Bland, M., Holmes, M., Khanom, N., & Pike, H. (2004). Style and content in e-mails and letters to male and female friends. Journal of Language and Social Psychology, 23, 369-378.