Differences in Women's and Men's Usage of E-mail

Nelson Tang

IS 246

Prof. Howard Besser

December 8, 1999


E-mail usage is rapidly becoming a preferred communication medium for many people. Businesspeople of all kinds routinely use e-mail, both for work purposes as well as socializing and even finding mates online (Johnson, 1994). Patients feel a new sense of connection to their physicians when they can send e-mail nearly any time of the day to ask questions or seek advice (Fein, 1997). Neighbors use e-mail to communicate with other neighbors, augmenting their physical community with a virtual community (Kugel, 1999). Additionally, families are taking advantage of the medium to bring together family members who are distant, both geographically as well as emotionally (Harmon, 1998). E-mail is being used in by people in many diverse areas, for many diverse reasons.

It is no wonder that e-mail usage is skyrocketing. E-mail has quite a number of advantages. Compared to conventional letter-writing, e-mail is fast and convenient; in fact, postal mail is referred to as "snail mail" by frequent e-mail users. An e-mail message usually takes only a few minutes to be transmitted to the recipient's mailbox, even if the recipient is halfway around the country or even the world. E-mail can be sent and responded to in an asynchronous manner, to the convenience of both the sender and the receiver. And since e-mail is already in digital format, it is easy to save, copy, and/or forward to others, creating simple mailing lists.

Yet it seems that not everyone is equally enamored with this new technology. Specifically, men and women do not seem to respond to e-mail in the same way. In general, men seem to embrace e-mail usage and culture in a much more comprehensive way, while women use e-mail for fewer, more stringent communications needs. What accounts for these differences in attitudes and usage patterns?

I will explain these differences in two ways. First, I believe these differences can be explained by taking communications theory and gender differences in conventional communications media and applying them to the specific domain of e-mail communications. The theories I believe have direct relevance to answering this question are as follows: the use of "report-talk" versus "rapport-talk"; conversation as a means to an end versus as an end in itself; turn-taking versus cooperative overlap; dominating a conversation through the controlling technique of silence; and the prevailing attitudes of assigning gender to technology.

Additionally, anecdotal evidence I have gathered in informal conversations suggests further areas of research to explain the differences of e-mail usage. These ideas have not been explicitly stated in the research, but they expand on the concepts found in the research. They suggest alternate approaches to understand why women and men use e-mail differently, and I will explain these ideas and how they apply to the problem.

Though every person is unique, there seems to be a general trend that women and men use e-mail in different ways to communicate. Through the theories which are grounded in both existing communications and gender theory as well as my own speculations, I will explain in this paper these differences in e-mail usage.

Communications and gender theories applied

Report-talk versus rapport-talk

The first theory to explain the differences in e-mail usage are based on the difference between "report-talk" and "rapport-talk." These two styles of talk are used by men and women, respectively, in their conversations. Tannen describes the difference between report-talk and rapport-talk in the following way:

For most women, the language of conversation is primarily a language of rapport: a way of establishing connections and negotiating relationships.... For most men, talk is primarily a means to preserve independence and negotiate and maintain status in a hierarchical social order. This is done by exhibiting knowledge and skill.... (Tannen, 1990, p. 77)
In other words, women talk to build rapport and men talk to impart knowledge. Women desire rapport because they seek to make networks of relationships with other people. For them, it is important to emphasize sameness and equality with their peers with whom they are establishing rapport. However, for men, the world is seen as a hierarchy, with people in a higher or lower status position with respect to the man. Exhibiting knowledge and skill is necessary to determine relative status.

But how does this theory apply to determining gender usage differences in e-mail? E-mail is a medium which is very efficient at transmitting knowledge and information. Given that this is the case, it should then come as no surprise that men respond positively to using e-mail as a natural medium through which to "talk," since they can easily exchange information to determine hierarchical status. On the other hand, e-mail is not able to convey emotional tone easily. (This is changing, as people gradually adapt the technology to convey emotional content through usage of "smileys", "emoticons" and other conventions. See, for example, "Useful Internet Emoticons" at http://www.randomhouse.com/features/davebarry/emoticon.html.) Bateson, as cited by Tannen (1990), refers to the emotional content and the context of a message as the "metamessage" (p. 32). E-mail does not transmit metamessages easily, stripping away all but the literal message. What this means is that women who attempt to use e-mail to converse are likely to be frustrated by the medium's inability to effectively transmit metamessages and paralinguistic hints which are necessary to establishing feelings of rapport.

This theory also supports another observed trend: women do use e-mail to exchange factual information, such as a woman telling a family member what time she is arriving at the airport and from which gate to come pick her up. In this case, the woman is not using e-mail for rapport-talk, and thus the poor metamessage capacity of an e-mail channel is not a hindrance in her goal of conveying unemotional, factual information. However, such information exchange would not be considered "talk" by the woman; it does nothing to further her goal of building rapport with her conversation partner.

This difference between rapport-talk and report-talk is a very significant difference between women's and men's conversation preferences. Traces of this difference can be seen in many of the other theories which are described below.

Conversations as means to an end versus an end in itself

Another result of communications theory is the observation that women and men use conversation for different means. Ebben (1994) states that "[f]or many women, conversation is an activity in itself, while for many men, conversation is merely a means to an end" (p. 87). Why, then, would e-mail conversations not be valued by women as an activity? Again, I believe the lack of metamessages in e-mail leaves women feeling they are not having a conversation, but instead typing words into a void. For men, however, an e-mail "conversation" could be for the express purpose of exchanging information, perhaps in a competitive way, subconsciously attempting to define hierarchy and relative status. What men consider good conversation can differ significantly from what women consider good conversation. This is clearly illustrated by the following quote, recounting an anecdote regarding linguistic students who asked others to tape conversations for them to analyze:

Another woman's husband delivered a tape to her with great satisfaction and pride. "This is a good conversation," he announced, "because it's not just him and me shooting the breeze, like 'Hi, how are you? I saw a good movie the other day,' and stuff. It's a problem-solving task. Each line is meaningful." When the woman listened to the tape, she heard her husband and his friend trying to solve a computer problem. Everything they said was technical and impersonal. Not only did she not consider it "a good conversation," she didn't really regard it as a conversation at all. His idea of a good conversation was one with impersonal, factual, task-focused content. Hers was one with personal content. (Tannen, 1990, p. 103)
Clearly, the husband in this example felt that his and his friend's problem-oriented talk made for a good conversation. Given this belief, it is not hard to imagine these two friends solving their computer problem via an e-mail exchange and feeling that the exchange was also a "good conversation." The men's task-driven discourse plays very well to e-mail's strength in conveying factual information, thus supporting men's acceptance of e-mail usage, while women's use of conversation as an end in itself exposes e-mail's difficulty in conveying metamessage information.

Turn-taking versus overlap

Just as women and men value conversation for different reasons, they may approach the structure of a conversation in different ways. It has been observed that many men prefer a turn-taking approach to conversation, where the person speaking "has the floor" until ceding it to the next speaker. This conversational model fits naturally with the idea of using conversation as a competition to establish hierarchy, and competition, argumentativeness, and disagreement are well-established in communications theory as being stereotypical behaviors of men (Rancer and Dierks-Stewart, 1987; Gayle, Preiss, and Allen, 1994).

On the other hand, many women prefer a conversation style with frequent cooperative overlap, to show support for the speaker and to help jointly develop the conversation (Coates, 1988; Tannen, 1990). Note that this overlap is different from an interruption. Interruptions are uncooperative overlaps; they "have the potential to disrupt turns at talk, disorganize the ongoing construction of conversational topics, and violate the current speaker's right to be engaged in speaking" (West and Zimmerman, 1983, p. 105). Cooperative overlap, on the other hand, indicates agreement, sameness, commonality, and rapport between the conversation partners.

However, given e-mail's asynchronous nature, it is clearly difficult to overlap (cooperatively or not) the speaker during a conversation. Instead, the best that can be done is only an approximation. When responding to a particular portion of an e-mail message, a common practice is to quote that portion and then respond directly below it. Ebben (1994) suggests that through this practice, a more synchronous feel can be simulated:

This convention of participants citing the previous contributor's specific words and inserting their own words in response serves to create, in a sense, a "conversation" among the participants. By contextualizing new remarks beside previously stated remarks, the practice restructures the flow of computer-mediated "talk" from a series of "monologues" to a more complex, and interconnected stream of "dialogue." (p. 138)
Despite this convention, overlap is still practically impossible to achieve. Instead, Ebben goes on to suggest that this simulated synchronicity "is most closely adhered to when participants are in disagreement with each other" (p. 141). Thus women wanting to format an e-mail message that allows them to cooperatively overlap will be frustrated, but men looking for a formatting that allows them to argumentatively challenge an e-mail message can simply using this interweaving strategy. Once again, the medium lends itself more naturally to men's communications preferences but not to women's preferences.

Silence as control

Communications theory also tells us men tend to control the conversation more than women do. One way this control is exercised in is through judicious use of silence. Sattel (1983) states that "inexpressiveness is related to men's position of dominance; that inexpressiveness works as a method for achieving control both in male-female and in male-male interaction" (p. 123).

This strategy of silence can also be implemented in e-mail conversations. Application of such inexpressiveness can range from deft cut-and-pasting in order to ignore certain statements or topics, to completely ignoring a message by not responding at all. This is not to suggest that only men can use this tactic; women too can cut-and-paste and ignore e-mail messages. However, since the e-mail medium can easily be used in this manner to achieve inexpressiveness, men may be able to use it more naturally and thus feel more at ease with the medium overall. Meanwhile, women may resist using these ignoring techniques precisely for the feel of dominance it lends over the conversation, and thus that provides another reason to women to feel that e-mail is unnatural for conversations.

Technology as inherently masculine

Communications theory is not the only body of research that can be applied to this question of e-mail usage. Feminist theory can also be applied to help explain the imbalance of usage. Notably, technology and computers specifically are clearly associated as predominantly men's area. It is the case that "our culture labels computer technology as belonging to the male domain" (Ebben, 1994, p. 49). The technology and the medium itself are considered to be inherently coded to men and men's values. By valuing efficiency, speed, and other easily quantifiable and comparable metrics, technology fits with men's tendency to organize by hierarchy one-up/one-down positions.

Nowadays, female computer programmers and are still the exception; it is not uncommon for a software design team to be composed entirely of males. Given this enormous slant towards male attitudes and experiences in the design and production phases of computer products, feminist theory argues that it clearly follows that women would have difficulty feeling comfortable with such products. For example, feminist theory states that new technology is designed to suit the needs of those in power, which would be men. Ebben (1994) goes on to say that computer-mediated communication "may similarly embody technological design choices which may be antithetical to women's conversational preferences" (p. 214). In other words, women may respond negatively to e-mail usage as a general communications medium because they subconsciously do not feel a connection to the technology. The technology was not designed by women, it does not take into account women's values, and therefore it is no surprise that women do not feel their internal needs are met by the technology. On the other hand, men may subconsciously feel an attachment and a validation from the technology, which was designed by other men, for other men, and with men's values in mind.

Further explanations

These theories from communications and gender theory can all be used to help explain the difference between women's and men's usage of e-mail. However, in researching this problem, anecdotal evidence has suggested further explanations for this difference. These concepts are based on and extend many of the theories already documented in the research literature, and they are corroborated through my informal discussions with a few women friends in informal discussion. There are three ideas to further explain men's and women's e-mail usages: the unsatisfying malleability of digital communications; the subconscious uncertainty of the e-mail exchanger's identity; and devaluing of e-mail due to its ease and simplicity. I will explain these ideas in further detail.

The malleability of digital communications

Digital communications, by their very nature, are malleable. By this I mean that only the literal message is guaranteed to be transmitted to the recipient, whereas the physical manifestations of digital communications are alterable and can take many different forms. Naturally, this also holds true for e-mail messages. When a person sends e-mail, he or she enters the message in some format, using some editor or mail agent. However, the recipient of the message receives it in some format based on his or her preferences. This format may be entirely different than the sender's format: the message may be in a different font or differently-sized; it may have different margins, especially if the sender's mail agent does not forcibly insert line breaks at certain columns; the recipient may even use text-to-speech software to read the message aloud, such as if he or she is visually impaired! There is absolutely no guarantee that the receiver of the message will experience the message in the same way that the sender sent it, due to this flexibility of digital communications.

What this means in terms of gender preferences for e-mail usage is related to the earlier discussion of paralinguistic cues and metamessages. According to Ebben (1994), citing Mehrabian, "93 percent of a communicator's intent [is] conveyed by paralinguistic cues" (p. 24). Since digital communications are so malleable, they can be transformed into many different formats which, though they preserve the literal message, are likely to distort or filter out any paralinguistic cues, which are not usually carried by the literal message itself. For example, the sender of a message may be using an e-mail agent with a certain font that looks very serious, but the recipient's e-mail agent might display the message in a large, informal-looking typeface.

This malleability is unique to digital communications. For other traditional forms of communication, there is some level of physical traces in the message which is difficult to alter. Conventional letters are a perfect example; when the recipient receives a letter, it is guaranteed that the letter itself is physically the same letter the sender wrote on, folded, and placed in the envelope. Even telephone communications embody a unique aspect of the sender: the sender's voice, though digitized and transmitted as pure information, is reconstructed with enough accuracy and fidelity that it is generally considered to be truly the speaker's voice. These unique physical traces are where paralinguistic cues are usually transmitted: the style of handwriting, the choice of paper and color of ink, the tone and speed of voice, etc.

But e-mail and other forms of pure digital communications can not claim to have this physical and unique trace of the sender in what the recipient receives. Without this guarantee, paralinguistic cues can not be assuredly sent, and this impacts women who rely on their use to have an emotionally-fulfilling conversation. Men may not be as impacted if they rely more on the literal message itself. For this reason, the malleability of digital communications leads to a discomfort and rejection of the use of e-mail for women's conversational needs.

Subconscious uncertainty of the e-mail exchanger's identity

Another theory on the differences in e-mail usage is what I believe is a subconscious uncertainty of the identity of the person with whom e-mail is being exchanged. As mentioned in the previous section, digital communications is very malleable, and the way in which it is presented to the receiver is not necessarily the same as the way in which the sender inputs the message. The computer interfaces to the users' e-mail agents insert a layer of anonymity and indirection (of an emotional nature) between the two users. This can lead to the problem of users being not quite certain, if even at a subconscious level, of who exactly their recipients are.

This is illustrated by the following quote:

In addition to conveying less of the richness of face-to-face communication, computer-mediated communication is also said to lead to mechanomorphism, the perception of the computer-mediated communication partner as machine-like. Shamp (1989) found that perceptions of computer-mediated communication partners were more similar to perceptions of the computer than were perceptions of a person with whom the individual communicated through media other than the computer. These results suggest that the lack of paralinguistic cues in online communication may play an influential role in users' perceptions of online communication partners as being more machine-like than human. (Ebben, 1994, p. 25)
This "disinhibition" is further described by Gackenbach, as quoted by Turner (1998): "Even if two parties know each other, such as in a discussion via corporate e-mail, Gackenbach says that there is a sense of emotional anonymity brought on by the lack of nonverbal communication, cues that would be a part of a normal conversation. 'There are many examples in which we know something intellectually but still behave based on our emotions. This is just one of them'" (p. B4).

What can happen is the user finds the recipient emotionally anonymous because of the introduction of the computer interface. The user can then confuse the computer with the actual human recipient of the e-mail exchange. This is similar to a famous hypotheses posited by the famous computer scientist Alan Turing, known as the Turing Test for intelligence. This test is as follows: suppose a person is connected via a terminal to either a computer or another human. The person is allowed to type questions to attempt to determine if the entity on the other side of the connection is a computer or a human. If the interrogator cannot tell the two apart, then it can be said that the computer is intelligent (Turing, 1950).

Thus, given the layer of anonymity and the tendency for people to mechanomorphise the computer in e-mail exchanges, a subconscious and emotional uncertainty for the true identity of the e-mail recipient is developed. As mentioned above, the lack of paralinguistic cues is an important cause of this tendency. Because women are generally more attuned to such paralinguistic cues, they would be more sensitive to the absence of such cues from this communications medium. The effects of disinhibition and mechanomorphism would then be more pronounced on women than men. Women may prefer the emotional certainty of knowing who their communication partners are, and this would mean relying on another, less anonymous, means of communication.

Devaluing e-mail for its ease

Throughout this entire paper, much has been made of the speed and efficiency of e-mail communications, and the ease and convenience with which people can send e-mail. Perhaps these traits actually lead to a devaluation of e-mail as a personal communications medium. In my informal interviews with women asking them what they perceive as the difference between using e-mail or physical letters for personal communication, many of the responses included valuing physical letters precisely because they require some amount of effort.

Clearly, it takes time and effort to find paper, write a letter, get an envelope and a stamp, assemble the letter, and take the letter to a mailbox or to the post office. The recipient is fully aware of the amount of effort involved in sending a physical letter, and for this reason he or she values the letter at some level. E-mail, on the other hand, is generally quite a bit easier to send. A person can sit down in front of any computer with a network connection, start up an e-mail agent, and type away. When the person is finished, he or she clicks on a button marked "send" or invokes the sending in some other way, and usually without any more effort on the user's part, the e-mail message is sent. Typically, the user is not charged any extra money for sending the message (above that which he or she spends for the network connection itself, and even that cost is often absorbed by the company for which the user works or the university the user attends).

This simplicity and convenience certainly may be attractive to those who dislike the overhead of physical letter writing; this may be conducive to allow such people to communicate more regularly than not at all if they did not have access to such an inexpensive medium. However, it is precisely this ease that may lend people to not value e-mail communications as much as physical letters. Telephone conversations, too, involve money, and they are usually much more expensive than sending e-mail. For this reason, they too are often valued more highly than e-mail messages (though their synchronous nature also provides an important different feature than e-mail). A recent article in the UCLA student newspaper also supports this differing valuation of e-mail versus physical letters: "E-mail can be quicker, easier and cheaper.... For many, however, true emotion is stuffed inside an envelope in the form of a hand-written letter. 'When writing a letter, you put more into it. You actually address the person.... It's more personal.'" (Shah, 1999).

The very features of e-mail which are valorized can be held against it as a devalued communications medium. This would be primarily true with women, since it is masculine and not feminine values that e-mail's advantages maximize, as discussed previously. For this reason, women may prefer not to use e-mail for their personal correspondence.


Women and men tend to use e-mail in different ways, for different reasons. Communications, gender, and feminist research provide a number of theories which can be applied to explaining these differences: report-talk versus rapport-talk; conversation as a means to an end versus an end in itself; turn-taking versus cooperating overlap; dominating a conversation through simulated silence; and feelings of technology as inherently masculine. Furthermore, other explanations can be offered for these differences, based on anecdotal evidence: the unsatisfying malleability of digital communications; subconscious uncertainty of identity; and devaluing e-mail for its ease and simplicity.

Women and men are different, and this includes their preferences for using e-mail for personal correspondence. With the ideas presented in this paper to explain this difference in attitude, perhaps both women's and men's preferences can be understood and accommodated to foster smoother conversations in e-mail or other media.


Coates, J. (1988). Gossip revisited: language in all-female groups. In J. Coates and D. Cameron (Eds.), Women in speech communities: New perspectives on language and sex (pp. 94-122). New York: Longman.

Ebben, M. M. (1994). Women on the net: An exploratory study of gender dynamics on the soc.women computer network. Ph.D. dissertation, University of Illinois at Urbana-Champaign.

Fein, E. B. (1997, November 20). For many physicians, e-mail is the high-tech house call. New York Times, A1(N), A1(L).

Gayle, B. M., Preiss, R. W., and Allen, M. (1994). Gender differences and the use of conflict strategies. In Turner, L. H., and Sterk, H. M. (Eds.), Differences that make a difference: Examining the assumptions in gender research (pp. 13-26). Westport, Connecticut: Bergin & Garvey.

Harmon, A. (1998, March 26). Guess who's going on line: to the studied horror gen-x'ers, mom is moving in on their cyberspace, and binding the net with family ties. New York Times, D1(N), G1(L).

Johnson, K. (1994, March 26). On-line romance; office workers feel cupid's byte. New York Times, 13(N), 21(L).

Kugel, J. F. (1999, August 17). The electronic block party. New York Times, A19(N), A17(L).

Rancer, A. S., and Dierks-Stewart, K. J. (1987). Biological and psychological gender differences in trait argumentativeness. In Stewart, L. P. and Ting-Tooney, S. (Eds.), Communication, gender, and sex roles in diverse interaction contexts (pp. 18-30). Norwood, New Jersey: Ablex Publishing Corporation.

Sattel, J. W. (1983). Men, inexpressiveness, and power. In B. Thorne, C. Kramarae, and N. Henley (Eds.), Language, gender, and society (pp. 119-124). Rowley, Massachusetts: Newbury House.

Shah, A. (1999, November 15). Return to sender: In an age dominated by efficiency and computers, the art of letter writing lends a personal and emotional feel to human interaction. The Daily Bruin, 3.

Tannen, D. (1990). You just don't understand: Women and men in conversation. New York: Ballantine Books.

Turing, A. M. (1950). Computing machinery and intelligence. In MIND (Journal of the mind association) 59(236), 433-460. Available: http://www.abelard.org/turpap/turpap.htm

Turner, J. (1998, October 29). Internet's illusion of intimacy leaves a lot of red faces. The Christian Science Monitor, B4.

West, C. and Zimmerman, D. H. (1983). Small insults: A study of interruptions in cross-sex conversations between unacquainted persons. In B. Thorne, C. Kramarae, and N. Henley (Eds.), Language, gender, and society (pp. 103-118). Rowley, Massachusetts: Newbury House.