Abstract not found

In this work we examine the characteristics of social activity and patterns of communication on Twitter, a prominent example of the emerging class of communication systems we call “social awareness streams. ” We use system data and message content from over 350 Twitter users, applying human coding and quantitative analysis to provide a deeper understanding of the activity of individuals on the Twitter network. In particular, we develop a content-based categorization of the type of messages posted by Twitter users, based on which we examine users ’ activity. Our analysis shows two common types of user behavior in terms of the content of the posted messages, and exposes differences between users in respect to these activities.

Go to a section in the article: Social network sites (SNSs) provide a new way to organize and navigate an egocentric social network. Are they a fad, briefly popular but ultimately useless? Or are they the harbingers of a new and more powerful social world, where the ability to maintain an immense network—a social "supernet"—fundamentally changes the scale of human society? This article presents signaling theory as a conceptual framework with which to assess the transformative potential of SNSs and to guide their design to make them into more effective social tools. It shows how the costs associated with adding friends and evaluating profiles affect the reliability of users ' self-presentation; examines strategies such as information fashion and risk-taking; and shows how these costs and strategies affect how the publicly-displayed social network aids the establishment of trust, identity, and cooperation—the essential foundations for an expanded social world.

This paper examines the location traces of 489 users of a location sharing social network for relationships between the users ’ mobility patterns and structural properties of their underlying social network. We introduce a novel set of locationbased features for analyzing the social context of a geographic region, including location entropy, which measures the diversity of unique visitors of a location. Using these features, we provide a model for predicting friendship between two users by analyzing their location trails. Our model achieves significant gains over simpler models based only on direct properties of the co-location histories, such as the number of co-locations. We also show a positive relationship between the entropy of the locations the user visits and the number of social ties that user has in the network. We discuss how the offline mobility of users can have implications for both researchers and designers of online social networks.

YouTube is a public video-sharing website where people can experience varying degrees of engagement with videos, ranging from casual viewing to sharing videos in order to maintain social relationships. Based on a one-year ethnographic project, this article analyzes how YouTube participants developed and maintained social networks by manipulating physical and interpretive access to their videos. The analysis reveals how circulating and sharing videos reflects different social relationships among youth. It also identifies varying degrees of ‘‘publicness’ ’ in video sharing. Some participants exhibited ‘‘publicly private’ ’ behavior, in which video makers ’ identities were revealed, but con-tent was relatively private because it was not widely accessed. In contrast, ‘‘privately public’ ’ behavior involved sharing widely accessible content with many viewers, while limiting access to detailed information about video producers ’ identities. doi:10.1111/j.1083-6101.2007.00400.x

Abstract not found

Previous research has shown a relationship between use of social networking sites and feelings of social capital. However, most studies have relied on self-reports by college students. The goals of the current study are to (1) validate the common self-report scale using empirical data from Facebook, (2) test whether previous findings generalize to older and international populations, and (3) delve into the specific activities linked to feelings of social capital and loneliness. In particular, we investigate the role of directed interaction between pairs—such as wall posts, comments, and “likes”— and consumption of friends ’ content, including status updates, photos, and friends ’ conversations with other friends. We find that directed communication is associated with greater feelings of bonding social capital and lower loneliness, but has only a modest relationship with bridging social capital, which is primarily related to overall friend network size. Surprisingly, users who consume greater levels of content report reduced bridging and bonding social capital and increased loneliness. Implications for designs to support well-being are discussed. Author Keywords Social network sites, social capital, loneliness, computermediated communication

As social computing systems persist over time, the user experiences and interactions they support may change. One type of social computing system, Social Network Sites (SNSs), are becoming more popular across broad segments of Internet users. Facebook, in particular, has very broad participation amongst college attendees, and has been growing in other populations as well. This paper looks at how use of Facebook has changed over time, as indicated by three consecutive years of survey data and interviews with a subset of survey respondents. Reported uses of the site remain relatively constant over time, but the perceived audience for user profiles and attitudes about the site show differences over the study period. Author Keywords Social network sites, Facebook, audience, privacy,

The prevailing paradigm in Internet privacy literature, treating privacy within a context merely of rights and violations, is inadequate for studying the Internet as a social realm. Following Goffman on self-presentation and Altman’s theorizing of privacy as an optimization between competing pressures for disclosure and withdrawal, the author investigates the mechanisms used by a sample (n = 704) of college students, the vast majority users of Facebook and Myspace, to negotiate boundaries between public and private. Findings show little to no relationship between online privacy concerns and information disclosure on online social network sites. Students manage unwanted audience concerns by adjusting profile visibility and using nicknames but not by restricting the information within the profile. Mechanisms analogous to boundary regulation in physical space, such as walls, locks, and doors, are favored; little adaptation is made to the Internet’s key features of persistence, searchability, and cross-indexability. The author also finds significant racial and gender differences.

This research explores how cues deposited by social partners onto one's online networking profile affect observers' impressions of the profile owner. An experiment tested the relationships between both (a) what one's associates say about a person on a social network site via ''wall postings,'' where friends leave public messages, and (b) the physical attractiveness of one's associates reflected in the photos that accompany their wall postings on the attractiveness and credibility observers attribute to the target profile owner. Results indicated that profile owners' friends' attractiveness affected their own in an assimilative pattern. Favorable or unfavorable statements about the targets interacted with target gender: Negatively valenced messages about certain moral behaviors increased male profile owners' perceived physical attractiveness, although they caused females to be viewed as less attractive. doi:10.1111/j. 1468-2958.2007.00312.x Forming and managing impressions is a fundamental process, and one that has been complicated by new communication technologies. As computer-mediated communication (CMC) has diffused, successive technological variations raise new questions about interpersonal impressions. For example, with people meeting via text-based CMC-e-mail, discussion groups, or chat spaces of various kinds-a variety of questions arose about impression formation and management. These included whether and at what rate impressions are formed online The newest forms of online communication complicate matters in ways that are unique with respect to the kinds of information they offer for observers to draw impressions, and they raise different theoretical issues than those formerly applied to interactive uncertainty reduction online. Social networking technologies, such as Facebook (http://www.facebook.com), offer a blend of interactive and static features in any one individual's online ''profile. '' Moreover, what complicates these sites from an impression formation perspective is that people other than the person about whom the site is focused also contribute information to the site. Such postings may or may not include secondhand descriptions about the target individual and his or her conduct. More importantly, whereas postings by other people on one's own profile reflect the character of the individuals who made the postings, it is also possible that observers' reactions of those others may affect perceptions of the target profile maker as well, even though the profile maker his-or herself did not initiate or condone the postings. This makes participative social networking technologies different from Web pages, e-mail, or online chat because all those technologies allow the initiator complete control over what appears in association with his-or herself. The possibility that individuals may be judged on the basis of others' behaviors in such spaces prompts this question: Are we known by the company we keep? This research examined the question of how other individuals' contributions to one's own online profile affect observers' impressions and evaluations of the profile maker. We discuss several approaches to impression formation online with respect to personal characteristics and draw upon offline research specifically related to context effects on judgments of physical attractiveness and interpersonal evaluations in order to form hypotheses about online perceptions that can be tested on the basis of judgments resulting from different characteristics on Facebook profiles. An original experiment involved mock-up Facebook profiles that alternately featured attractive or unattractive elements surrounding invariant central profile material. These variations affected observers' impressions of credibility and attractiveness, all without the target of these judgments having changed. Facebook Facebook is a social networking Web site initially built for college communities. It is organized around social networks corresponding to schools and, recently, other institutions and locales. Like other online social networking sites (see for review, As we will show, garnering impressions from online information is nothing new, although the kinds of information social networking technologies present and the manner in which impressions form may be. Impression formation and online venues People spend considerable effort in order to form and to manage impressions, especially when anticipating or engaging in the initial stage of interactions Because online impressions are controllable, they are often suspect. Online users can organize the information flow and enhance self-image by strategically selecting how and what to convey to the receiver One way in which Facebook differs from other online sites for self-presentation has to do precisely with the degree to which some personal information is presented by means other than disclosure by the person to whom it refers. In many other communication settings, as Petronio (2002) notes, people make active decisions about when and how they will self-disclose. These decisions involve a complex process in which people set rules about how and when they will divulge private information, negotiate those rules with other people, and make decisions on disclosure based on violations of those rules. However, social networking sites to some extent obviate an individual's rules, negotiations, and disclosure decisions by placing discretion at the mercy of their social networks: ''While (individuals) may have control over the content they disclose on their university-housed webpages, friends . can post discrediting or defamatory messages on users' Facebook websites,'' according to Mazer, Murphy, and Simonds (2007, p. 3). Walther and Parks (2002) considered the interplay of online self-generated claims and other-generated clues in their formulation of the ''warranting'' value of information, or the perceived validity of information presented online with respect to illuminating someone's offline characteristics. The warranting value of information is hypothesized to be a function of the degree to which that information is perceived to be immune to manipulation from the target to whom the information pertains. Several forms of high-warrant information have been nominated that meet this classification. Information gained about someone through others in that person's social network is one example. People try to find out about one another via common nodes in overlapping social networks for several reasons: to reduce uncertainty about the partner, for example, and to do so without direct knowledge of the target that s/he is being inquired after People use various features to assess personalities of the Web site creators The Brunswikian lens model Brunswik's lens model Self-directed identity claims are ''symbolic statements made by occupants for their own benefit, intended to reinforce their self-views'' (p. 380). Other-directed identity claims are ''symbols that have shared meanings to make statements to others about how they would like to be regarded'' (p. 380). Gosling et al. Context effects In addition to the Brunswikian lens, research focusing on ''context effects'' in person perception also suggests that characteristics of other people who appear together with a target affect perceived attractiveness of the target. In this case, research has focused on physical attractiveness judgments. Rather than focus on global ''goodness'' in the present case, we will focus on common interpersonal evaluations that are sensitive to communication variations. Although any number of attributes might be examined, the nonphysical aspects of interpersonal attractiveness-social and task attractiveness-along with source credibility are common evaluative dimensions in impressions of interaction partners, and their factor-based measures have been used in communication research for over 3 decades (see With regard to the effect of friends' apparent physical attractiveness on a profile owner's task attractiveness and credibility, we hypothesize the opposite direction than we do for physical and social attractiveness. Physical attractiveness has been associated with lower observers' perceptions of competence. ''Physical beauty. can impose limitations on the kinds of performances others view as credible. Beautiful people may be considered unapproachable or unintelligent,'' according to a review by Burgoon and Saine (1978, p. 263). Once again, it is likely but not certain whether such an attribution of an individual's abilities is bestowed due to or in contrast to H2: The physical attractiveness of one's friends depicted on a virtual profile affects observers' perceptions of the profile owner on social evaluations (a) in the same direction as physical attractiveness, for social attractiveness and (b) in the opposite direction as physical attractiveness, for (i) task attractiveness and (ii) credibility. RQ1: Does the effect of physical attractiveness of friends' photos interact with the sex of the profile owner on (a) task attractiveness and (b) credibility? Hypotheses are also derived based on context effects and the Brunswikian lens conception of the utilization of cues. We extend cue utilization to include behavioral residues that appear in the wall postings that friends leave in a profile owner's (virtual) space. Although they are not created by the target, because the target allows specific others to leave these residues such cues will be utilized and affect observers' impressions of targets. We hypothesize: H3: Statements made by others on one's virtual profile influence observers' social evaluation of the profile owner with respect to (a) social attractiveness, (b) task attractiveness, and (c) credibility, in directions consistent with the valence of the statement content. Finally, a research question asks whether the association between attributions of goodness and physical attractiveness works in reverse, if positive qualities bestowed by friends' statements left on a profile owner's site might invoke superior perceptions of physical attractiveness for the profile owner. Thus, the overall design was 2 (attractive or unattractive wall poster) 3 2 (positive or negative wall message) 3 2 (sex of profile owner) 3 2 (sex of participant). Stimuli Wall poster attractiveness Researchers pretested a number of photographs containing head shots of collegeaged persons in order to create specific variations in the apparent attractiveness of wall posters (high and low attractiveness) and depict neutral attractiveness for the ostensible profile owners. Researchers gathered photographs from photo-rating Web sites (i.e., hotornot.com), from photos generated in previous studies where permission to use participants' likenesses was given, and by utilizing an Internet search engine with keywords such as ''pretty girl.'' Photos that seemed to capture the range of attraction were presented to a mixed-sex group of college-aged raters (N = 10) who evaluated each photo on a scale from 210 (physically unattractive) to 10 (physically attractive). Analyses of these scores yielded two attractive and two unattractive male, and two attractive and two unattractive female, photos, among which there was no overlap on raw attractiveness ratings. These photos were used to represent wall posters on the profile mock-ups. Analyses also revealed neutral photos, which were employed to represent the profile owners. Message value Focus groups and pretests also contributed to the development of verbal messages in the wall postings. These postings were designed to reflect positively or negatively on the profile owner by describing things the owner had done or would do, which were socially desirable or undesirable among the college population based on the results of focus group discussions. Messages were generated in part by actual postings observed on Facebook that reflected the behaviors of the profile owner. Rewrites followed focus group discussions that helped shape messages that were both believable and positive or negative to the prospective participants who would later observe them. Each pair had a male and female version. The focus group discussions sought to discern what types of behaviors discussed in Facebook postings connoted desirable and undesirable qualities or characteristics. These discussions employed college-aged discussants other than those who were later to become research participants. Discussions began with the question: ''What statements on Facebook would lead you to think that the profile owner was a 'loser,' or conversely, a really desirable person?'' Respondents suggested that statements reflecting both excessive and morally dubious behavior were unfavorable. For example, drinking, per se, was not negative, nor was drinking a considerable amount, but drinking to the point of illness was. Flirting was not negative but flirting with unattractive targets, and promiscuity, was alleged to be. Positive messages, on the other hand, generally connoted that the profile owner was a socially desirable individual. The person was, for instance, popular with people or was included in enjoyable past or future activity of some type. Prototype messages were reviewed, and the final set consisted of two excessive and negative or inclusive and positive statements per stimulus. The specific foci of these stimulus messages may have some effects on the ultimate and differential responses to them, as will be discussed below. At the point of their creation, though, they satisfied the criteria of arousing respectively negative and positive global evaluations. Message postings also reflected profile owners', rather than posters', behavior. Although spectators reading about friends' own behaviors might also attribute qualities to the profile holder (in the same way that photos affect judgments), verbal messages were designed to reflect only owners' behaviors. This approach is consistent with the behavioral residual principle in which clues pertain to a central target, which drove the hypotheses regarding messages. This approach was bolstered by a content analysis of the nature of wall postings on Facebook, which indicated that many postings do indeed discuss the target explicitly. One coder content analyzed 237 randomly selected Facebook wall postings on the college network where the research took place. The coder classified 27% of friends' postings as presenting information about the target (profile owner), 23% as presenting information about the friend/ poster his-or herself, 28% presenting information about both the target and the poster, and 21% pertaining to other foci or undeterminable. Thus, approximately 55% of wall postings referred to the profile owner alone or the poster and the profile owner. Future research may explore the effects of self-oriented wall poster messages on profile owner evaluations if such questions are warranted. The final set of negative messages consisted of the following: ''WOW were you ever trashed last night! Im not sure Taylor was that impressed.'' ''Hey, do you remember how you got home last night? Last I remember you were hanging all over some nasty slob. please tell me you didnt take [him/her] home..'' Positive messages consisted of these: ''Vegas Baby!!! Only 3 days and we are on our way. Im so pumped!'' ''Chris, I just gotta say you rock!!! u were the life of the party last night. all my friends from home thought you were great!'' Typographical errors in these messages were intentional and reflect common writing characteristics in Facebook postings. An example stimulus is presented in Procedure Researchers provided volunteers with a single www address and requested that they complete the research on their own. Participants therefore viewed stimuli in a natural setting in which they typically used the Internet. The www address directed users to a Web page that reflected informed consent information and a button with which to signal consent. Clicking the button activated a hidden Javascript code that randomly redirected a participant's Web browser to one of the eight stimulus conditions. Participants were instructed to view the Facebook profile and then to follow a link to a questionnaire where they would answer questions about the Facebook profile owner. Questionnaire measures The posttest questionnaire included measures of task, social, and physical attractiveness developed by Results H1 predicted that greater physical attractiveness of friends displayed on wall postings in one's Facebook profile raises perceptions of the profile owner's physical attractiveness. Given the directional nature of the hypothesis, it was tested by way of a onetailed t test, which was significant, t(341) = 2.37, p , .01, h 2 = .02. Participants who saw attractive friends' photos rated the profile owner significantly more physically attractive (M = 3.65, SD = 1.17, N = 183) than did those exposed to unattractive photos (M = 3.30, SD = 1.17, N = 160). This finding confirms an assimilation effect: Profile owners' attractiveness varied in the same direction as their friends' did. An omnibus F test showed no interaction effects involving the photos' attractiveness with any other factors on physical attractiveness ratings. on social attractiveness as well (H2a), but negative effects were anticipated for task attractiveness and credibility (H2b). Because the directions of the hypotheses differed for social attractiveness versus task attractiveness and credibility, hypothesis testing proceeded in two steps. The test for H2a examined the effects of friends' photo attractiveness only on social attractiveness ratings. Because the hypothesis predicted directional effects for the impact of physically attractive photos on social attractiveness perceptions, a one-tailed t test was conducted, which supported H2a, t(341) = 1.70, p , .05, h 2 , .01. H2b specified negative effects on evaluations that were assessed with several measures: task attractiveness and the three dimensions of credibility (safety, qualification, and dynamism). Bartlett's test of sphericity, x 2 (20) = 695, p , .01, indicated that these variables were sufficiently related to be treated as a group. (Simple correlations among all dependent variables are presented in RQ1 explored whether previously documented stereotypical negative associations between physical beauty and competence for women were prompted by friends' photos and, if so, whether it was confined to judgments of females. Specifically, analyses focused on whether the interaction of physical attractiveness of the friends' photos with the sex of the profile owner affected task attractiveness and credibility. As in the previous test, this analysis employed a multivariate test, with the Photo 3 Target Gender interaction (and main effect factors) as the independent variables. The omnibus test did not yield significance for the interaction effect, Wilk's l = 0.99, F(4, 326) = 0.74, p = .66. It appears that the physical attractiveness of one's Facebook friends does not affect observers' judgments of one's qualifications, either directly or in combination with the target's gender. H3 predicted that the valence of statements left by friends on a profile owner's Facebook wall affects social evaluations of the profile owner in a consistent direction across the dimensions of social and task attractiveness and credibility. A preliminary MANOVA examining the multivariate main effect of wall statements yielded significance, Wilk's l = 0.89, F(5, 338) = 8.99, p , .01, h 2 = .11. See RQ2 asked whether the valence of friends' wall postings affected the physical attractiveness perceptions of the profile owner on whose wall the postings appeared. A preliminary omnibus analysis of variance revealed a significant disordinal interaction of message valence by the apparent gender of the profile owner, F(1, 339) = 6.47, p = .01, h 2 = .02. Female profile owners were rated more physically attractive when their profiles showed positive comments from friends, M = 3.98, SD = 1.29, 43 N = 80, than when the statements were negative, M = 3.46, SD = 1.19, N = 100. This pattern, however, was reversed when the profile owner was male; positive comments yielded lower physical attractiveness ratings, M = 3.20, SD = 0.91, N = 88, whereas negative statements produced greater physical attractiveness perceptions, M = 3.31, SD = 1.17, N = 75. Due to the higher order interaction, no further analysis of statements on physical attractiveness was warranted. Discussion The goal of this research was to examine how information provided by and about people's friends in a Facebook profile impacts judgments about the profile owners, who did not themselves provide that information. This investigation addresses novel theoretical questions about the process of impression formation and the influence of different forms of communication in that process as these issues are revealed by new communication technologies. In this experiment, the physical attractiveness of one's friends' photos, as seen in the Facebook wall postings presented on another individual's profile, had a significant effect on the physical attractiveness of the profile's owner. Perceptions of physical attractiveness did not reduce task competence attributions, an effect associated with evaluations of women in other, offline domains. It behooves one to have goodlooking friends in Facebook. One gains no advantage from looking better than one's friends. Although photo appearance on a wall posting may directly reflect a quality of the posting friend, not the profile owner, the verbal statements within wall postings may describe behaviors of the profile owner more directly. This approach was reflected in the stimuli adopted in the present study. Results showed that complimentary, prosocial statements by friends about profile owners improved the profile owner's social and task attractiveness, as well as the target's credibility. In all analyses, the effect sizes were very modest. This should be expected given the small proportion of overall information in the stimulus profiles that reflected the independent variable manipulations. For instance, even though there were two friends' photos on each profile, both of which were respectively attractive or unattractive, each profile also featured a much larger photo representing the target person. These target photos were neutral in attractiveness, by design. Furthermore, stimulus profiles offered neutral yet visible information about the target's interests and favorite media, as is customary on Facebook. Thus, given the large amount of common, neutral information about the target against which the small manipulations in friends' photos and wall comments appeared, small effects on the target's physical attractiveness and personal characteristics are not surprising. In a sense, it is somewhat surprising to see any effect on physical attractiveness perceptions, given that the alleged target's photo was more prominent than anything else. An unanticipated interaction effect involving the sex of the profile owner and the nature of the wall statements was obtained with respect to the effect of friends' comments on perceptions of the targets' physical attractiveness. The negative statements depicted normatively undesirable behavior, as they involved sexual innuendo and insinuated that the target person was drinking excessively the previous night. These statements raised the desirability of a man's appearance among the subject population in this study, whereas the residues of such behavior rendered the target physically unattractive when she is female. These results reflect what has come to be known as the sexual double standard when making social judgments or forming impressions of others. The sexual double standard pertains to the differences in individuals' evaluations of men and women who engage in premarital sexual behaviors: Men who engage in such encounters receive respect or admiration, whereas women who engage in similar behavior are often shunned or denigrated by society. This effect has been shown to be a pervasive belief in Western cultural ideology and the subject of much scholarly research (see, for review, Alternatively, because the information may be perceived as unsanctioned by the profile owner, it may have particularly great impression-bearing value. The results are consistent with Walther and Parks's (2002) warranting hypothesis. The warranting principle suggests that other-generated descriptions are more truthful to observers than target-generated claims. Findings that friends' statements significantly altered perceptions of profile owners support this contention. It is less costly to alter or distort claims that one makes about oneself (e.g., one's own profile) than to modify or manipulate statements made by others (e.g., their pictures and wall postings). Thus, information reflected in others' ''testimonials'' should be of special value to an individual making a judgment about the profile owner, according to the warranting principle. Results with regard to participants' ratings of profile owners' social and It is a clear limitation of the present research that the negatively oriented statements it employed focused on behaviors that might arouse these stereotypes and double standards. These manipulations arose as a result of the inductive approach to identify realistic positive and negative statements among college students who were familiar with Facebook norms. Future research with different theoretical concerns about the kinds of messages Facebook users display might focus on other topics and may find less gender stratified effects. Future research may also explore the perception of who ''owns'' the comments that appear on one's Facebook wall: owners, friends, or both? As Web sites become more interactive and participatory, the question of textual authority becomes less clear. The degree to which impressions formulate or dissipate may depend on the perceived independence or collusion the material leading to them appears to have and whether it remains independent or consensual, affecting how much online we are judged by the company we keep.