Betsy’s last post on the power of self-efficacy perceptions reminded me of a recent study by Professor Ellen Langer of Harvard University. I am fascinated by Prof. Langer’s ability to conduct provocative research and have used her ideas in my research on the illusion of control in spreadsheet-based decision making. Prof. Langer’s recent study, which showed that perceptions can be more important than reality, is equally provocative. She surveyed hotel maids and found that even though their daily physical activity at work exceeded the U.S. surgeon general’s recommendation for daily exercise, 67% of those surveyed did not see themselves as exercising. In fact, the hotel maids’ body fat, waist-to-hip ratio, blood pressure, weight and body mass index matched their perceived rather than their actual amount of exercise.

Prof. Langer then set out to see if perceptions were influencing these health indicators. She broke up 84 maids into two groups. Maids in one group were explained how many calories they were burning when they did different tasks during the day. They were told that their level of activity already met the surgeon general’s definition of an active lifestyle. The other group was not given any information. Physical measurements of the maids taken a month after this surprised Prof. Langer and her team: the blood pressure, weight, and waist-to-hip ratio of maids who were given information dropped! Since there was no change in the maids’ work routines, Prof. Langer concluded that the physical change was due to a change in their mindset. The physical body was simply responding to the maids’ perceptions (altered by education) that they were exercising as they went about doing their work. Thus, what mattered was not the actual level of exercise (i.e., the real thing) but rather how it was perceived.

In virtual teams too, perceptions matter and can be made to matter more than reality. As in any encounter, virtual team members need some information about other team members. They use this information to size up others in their team. This sizing up determines future interaction by influencing the trust that is formed and by acting as a filter through which the behaviors and the work of others is interpreted. Eventually, it determines the team’s outcomes, including how well the team executes its project or task. If there is no past or current interaction to help members of a virtual team build this information, they rely on information provided by the stereotypes that they associate with other members of the team. The basis of stereotyping is likely to come from demographic and occupational information about team members. Thus, team members’ perceptions about others based on their demographic and occupational profile is likely to play a significant role in a team’s process and outcomes and may even supersede the individual abilities of team members to contribute to the team’s project.

Much like Prof. Langer, who tried to control how the physical body responded by influencing the perceptions about exercise, the leader of a virtual team can cue the perceptions of team members to build team spirit, cohesion, and trust. One often hears that it is not possible or very difficult to build these in virtual teams. In fact, a recent study also indicated this difficulty. It found that virtual teams displayed lower cohesion and satisfaction and greater leader emergence than face-to-face teams. The authors attributed the higher level of leader emergence in virtual teams to perceived failure of communal decision making. But we know from other studies (see a review of some of these studies) that people do form strong, trusting bonds with others in online groups/forums and, in many cases, even fall in love with someone online. So what can a leader do to build cohesion and trust in a virtual team? The Social Identity Model of Deindividuation Effects (SIDE) provides an answer.

A short form of the SIDE model is that when members of a virtual team don’t have visual information about other team members, they become sensitive to identity cues provided by the team’s interaction. If the cues consistently highlight the team’s social identity, then that social identity becomes salient to team members and they behave in accord with the norms associated with that identity. For example, consider a group of information technology (IT) specialists who assemble for an online discussion. If these specialists repeatedly refer to their professional background to make a point (e.g., we as IT professionals should do this, we as IT professions should do that, etc.), SIDE says that the specialists will pay more attention to their identity as IT professionals and they will conform to expectations of how IT professionals should behave. I have observed support for this model in my research, as have several other researchers. Incidentally, the social identity does not have to be a publicly recognizable identity. Studies have shown that even labeling the team members as TeamA_M1 (Member 1 of Team A), TeamA_M2 (Member 2 of Team A), etc. makes their identity as members of Team A salient to them.

This model tells us that the leader of a virtual team who wants to build team spirit, cohesion and trust should consistently cue team members to focus on a shared identity. For instance, team members may have experience with new product development. When team members cannot see each other, their attention is taken away from individual differences and they latch on information that highlights their shared identity. They begin to see others in terms of this shared identity. Team members are likely to see this shared identity positively because they are themselves associated with it. Furthermore, to be consistent with how they feel about having this identity, they are also likely to have a more positive view of their team mates who share this identity. In such a climate, team members are likely to cooperate with each other, trust each other, and be more influenced by each other. In my own research, I have found that when individual identities were downplayed (by keeping members’ input as anonymous) and group membership was made salient, group members agreed more with each other at the end of their task. Essentially, anonymity took members’ attention away from individuals and, therefore, from individual differences, in the group. This increased the salience of the group membership information provided to team members and they became more aware of themselves as part of a team, which apparently led to greater agreement with others in the team.

What does this mean for leading virtual teams? Here are some implications. All of them are based on this idea: emphasize the team and how its members are an important part of it. The focus should be on the whole (the team and its purpose). Even when you talk about the parts (the members), emphasize how the parts are related to the whole. Always keep the whole in perspective — this will help the members identify themselves in terms of the team and its purpose.

1. Think about how to focus attention during introductions and initial meetings. Team members will of course introduce themselves, and some virtual team experts recommend allowing each member to discuss him/herself a bit in order to make each team member feel like a person. During these introductions, the team leader should be steering the conversation towards experiences, skills, and abilities that can be related to the team project and team goals. This might even mean asking team members ahead of time to present themselves with team building in mind. It certainly means the team leader should be prepared for the meeting, knowing something about each member that can be related to the group. If your team is meeting for the first time face-to-face, you may want to share the team’s purpose and and how each member’s skills and experiences fit with that purpose before the meeting. This means the first impression formed about team members will be relating to group’s goals and its ability to accomplish them so that stereotypes based on demographics are less salient during the first meeting.
2. Use collaboration technology that makes the group identity salient. Email is not suitable because it relies on individual addresses. If you have to use email, see if there is a way for you to set up a group address that highlights the shared identity (e.g., NewProductDevGroup) and which you and others can use to address email to the team. This would be similar to what you can do with Google Groups. Google Groups allows you to set up a group consisting of all the members of your virtual team and you can use the email address it sets up to send email to all. When an email is sent to this address, others who it is from and the name of the group. Ideally, you will set up a collaboration workspace where you can share documents and calendars, conduct discussions, and manage your project. The ideal workspace would identify the team by its name and present the team’s purpose, its members and their roles (Ann Majchrzak, Arvind Malhotra, Jeffrey Stamps, and Jessica Lipnack provide an example of such a workspace in their paper “Can Absence Make a Team Grow Stronger?“). Seeing the names together can be the beginning of thinking of oneself as a team. These cues may be subtle, but can have an effect if they are pervasive. The point is that the team leader should take every opportunity to make the team salient.
3. For teams meeting or working in a virtual world, it may also help to minimize the differences between avatars. This might mean pre-designing avatars that are fairly similar, and that don’t have any pronounced demographic characteristics that would make members feel different or excluded. In order to make this feel fun, you might find creative ways to make the avatars represent the organization’s overarching identity or the team’s purpose and goals. For example, the avatars might all wear Tshirts with the organization’s logo on them, or even wear clothing that the organization manufactures. Or they might all wear colors that are associated with the organization (e.g., blue for IBM). Since they are avatars, this can be used to mask what might normally separate into demographic groups – perhaps the avatars all have blue hair to represent the company and NOT clearly represent any human population. In a virtual world, you will have a lot of creative leeway to use avatar design to focus attention on what team members have in common, rather than how they differ or what each person looks like in “real” life.
4. In many of our latest blog entries, we mention the benefits of creating a team compact. In this case, the benefit is that it asks team members to carefully consider how they fit into the team, what the team’s purpose will be, and how the team will do work. All of this focuses attention on aspects of the group rather than aspects of each individual.

What we suggest in this post requires leaders to use some very careful judgment. Chances are that the members of your team were brought together because they have a variety of skills needed for a project. There is no question that this kind of diversity is useful, particularly for innovative or creative tasks. But remember as a leader to keep focusing attention, at every opportunity, on how these skills complement each other to produce a capable team. Ultimately, your goal is to prevent team members from thinking about themselves as individuals with different skills than their teammates. Help them think about how they are a team that possesses a broad range of abilities and experiences.