In this study we use personality dimensions as developed in Venkatsubramanyan and Hill (2007) to assess the perceiver’s impression of a target ePersona as a potential project teammate. Ven – katsubramanyan and Hill (2007) began with the traditional five-factor model of personality traits (Watson, 1989). The five-factor model comprises a hierarchical organization of five basic person­ality dimensions: Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to Experience. These five basic factors break down further into 106 personality dimensions. Through pilot testing the authors reduced the original personality dimension list down to a set of nine dimensions deemed by subjects as the

most applicable for assessing potential teammates based solely on Web-based perceptions. These include (a) Commitment to Excellence, (b) ability to work as an Effective Team Member, (c) ability to Manage Multiple Tasks, (d) ability to Handle Conflict, (e) having a strong interest in Working with People, (f) Managing Anger, (g) ability to Take Direction, (h) Curiosity, and (i) ability to Adapt to New Situations. The current study em­ploys these same personality dimensions to assess perceivers’ impressions of ePersonas.

Our experimental design used university students to evaluate ePersonas for the purpose of selecting potential team members for a class project. Subjects were recruited from several up­per division undergraduate business courses and offered course credit for their participation. 210 participated in the study including 104 males and 106 females and a wide range ofbusiness special­ties (Finance, Management, etc.). The task of selecting and evaluating a potential teammate was chosen for two main reasons. First, most college students have experience working in teams on course projects, and in many cases had to select those teammates themselves, thus the task was an appropriate one for our subjects. Second, we sug­gest that the process ofvetting potential teammates parallels the decision making processes required to assess and select one or more individuals from a pool of candidates and that these processes are largely independent of context. Examples of such decisions include selecting members for work – based teams, creating a short list of candidates for job interviews and choosing among professional service providers. The specific task of selecting a potential teammate is merely the scenario we used to develop a better understanding of how we form impressions of others based upon Web searches and we suggest that this scenario does not greatly limit the generalizability of our findings.

The experiment was conducted online with each subject randomly assigned to one of four manipulations representing the different potential teammates: John Doe 1, John Doe 2, Jane Doe 1 and Jane Doe 2. Each subject was provided a list of Google-style search results pertaining to his or her assigned potential teammate ePersona and was informed that the individual’s name had been changed to preserve anonymity. The search result links were disabled so that subjects’ information regarding the potential teammate was limited to the search results lists alone. If any link was clicked upon by the subject, a pop up message saying “DNS server is down. Please try again later.” was generated. Search results for John Doe 1 and Jane Doe 1 each contained 10 results, 8 of which were links to social networking sites. The results lists for John Doe 2 and Jane Doe 2 also contained 10 links though none of the results for these two ePersonas were to social networking sites. Please see Appendices A and B for screen captures of the search results for Jane Doe 1 and Jane Doe 2.

After looking at the search results page, sub­jects were asked to report the number of minutes spent reviewing the results to get an estimate of the amount of cognitive effort exerted in assessing the ePersona. Subjects were then asked to score the target as a potential team member on the nine factors described above. A five point Likert (Likert, 1932) scale ranging from “-2” (very unfavorable impression) to “+2” (very favorable impression) was used for each of these factors. Additionally, subjects were asked to rate the desirability of the target as a potential team member and their con­fidence in these desirability ratings. Also included in the rating was the question:

“If you are forced to make a decision at this point with no further information, would you select this person to be on your team – yes or no?”

Answers to this yes or no question were coded in the form of 1 or 0 for analysis purposes. On a scale of 1 to 5, subjects were asked to indicate how confident they felt about their decision. Subjects were asked to provide qualitative feedback as answers to these two questions:

1. What kind of teammate would the student

be?

2. What do you think influenced your decision?

Demographic information was also collected about the subjects including factors such as age, gender, major, number ofteam projects performed in the past, number of computer related courses, number of years of computer experience, number of years of experience searching the Internet, frequency of Web search, and a self-rating of his or her own level of Web search skills.