“Tastes, ties, and time: A new social network dataset using Facebook.com” – Kevin Lewis, Jason Kaufman, Marco Gonzalez, Andreas Wimmer, Nicholas Christakis
Central issue: “Scholars have long recognized the potential of Internet-based communication technologies for improving network research (…). In the first half of this paper, we introduce a new public dataset based on manipulations and embellishments of a popular social network site, Facebook.com. (…) In the second half of this paper, we present descriptive findings from our first wave of data.” (LKGWC, p.330)
> “In line with the work of Pierre Bourdieu (1984), previous research suggests that cultural proclivities play an important role in shaping social boundaries (…) and that ‘culture and social relations empirically interpenetrate with and mutually condition one another so thoroughly that it is well-nigh impossible to conceive of the one without the other’ (Emirbayer and Goodwin, 1994, p.1438).” (LKGWC, pp.333-334)
> “(…) we analyze how students conceptualize their tastes when unrestrained by closed-ended survey questions (…) [and] we explore the intersection of tastes and ties by calculating the extent of taste similarity between two students sharing various kinds of social relationship.” (LKGWC, p.334)
> Potential question: “Do individuals form ties with one another on the basis of shared preferences (selection)? Or are tastes instead transmitted through ties (socialization)?” (LKGWC, p.338)
> Remark: “The two possibilities are not mutually exclusive (see Kandel, 1978) (…).” (Ibid.)
Focus on sns and Facebook because: “Individuals can enter information on their background (e.g. high school, hometown), demographics (e.g. birthday, gender), ‘interests,’ political views, and group affiliations, as well as on their cultural tastes (e.g. ‘favorite’ books, movies, and music). Additionally, users can enter ‘friendship’ [1] relationships with other registered users and share photo albums that can be linked to the profiles of those present in a picture[2].” (LKGWC, p.331)
> Features of the dataset (LKGWC, pp.331-332):
- Natural research instrument
- Sociocentric nature of the environment: complete network data
» “(…) it is possible to accurately locate individuals within the network – determining their role or position vis-à-vis peers and the interconnectedness of actors beyond first-degree ties or ‘direct contacts’.” (LKGWC, p.332)
»> Note: “The disadvantage of using complete network data is that they are not representative of some larger population.” (Ibid.)
> Perhaps the current saturation rate of Facebook would make this less of a problem if the study were to be repeated today.
- Longitudinal data
» “(…) allowing researchers to observe how students’ networks, tastes, and group activities evolve over time.” (Ibid.)
- Multiplex data: multiple social relationships
- Data include: demographic, relational and cultural information
» Note: “Naturally, not all students provide information on all available variables; but even the response rate for cultural tastes is reasonably high (66.2% for movies, 67.5% for music, 65.6% for books) …” (LKGWC, p.331)
»> Remark: “Taste responses, (…), are undoubtedly not only a product of respondents’ ‘true’ preferences but also involve strategic presentation of self.” (LKGWC, pp.331-332)
Focus on social relations involves:
- Network size
- Network density = “(…) an indicator of the extent to which individuals identify with those around them [or] the potential strength of normative pressures toward conformity.” (LKGWC, p.334)
- Heterogeneity of personal networks
> Note: “Network heterogeneity (…) has been found to be positively associated with such outcomes as cultural awareness (Antonio, 2001), reduced ingroup bias and intergroup anxiety (Levin et al., 2003), and continued [heterogeneous] contact in the future (Emerson et al., 2002).” (Ibid.)
- Betweenness centrality = “(…) an index measuring one’s potential to control communication in a given network (…)” (Freeman, 1979; in: LKGWC, p.334)
Observations:
Taste and genres…
“In fact, students in our dataset rarely indicated a preference for a genre; and when they did, they often qualified this preference by giving examples of the particular subtype they preferred. Instead, students tended to list particular titles for ‘favorite movies’; particular artists for ‘favorite music’; and either authors or titles for ‘favorite books’.” (LKGWC, p.339)
> Signifies the advantage of the format of sns-research:
“The template provided by Facebook is completely open-ended such that no a priori assumptions are made regarding the form (or even the quantity) of tastes. (…) This allows us the rare opportunity to see what cultural preferences actually look like.” (LKGWC, p.338)
Taste and ties
“Two students involved in any of the friendship relations we examined share significantly more tastes in every category of tastes than we would expect from chance alone.” (LKGWC, p.339)
> In addition:
“(…) proximity is unimportant for taste similarity, controlling for friendship (…). In other words, co-residence may be an important predicator of friendship; but it is this emergent social affinity, not mere proximity, that is associated with cultural likeness – in the absence of which students may actually distance themselves from one another by adopting (or at least professing) discrepant tastes.” (LKGWC, p.340)
General insights:
“As the Internet in general and contemporary SNSs in particular play ever-greater roles in everyday life, virtual and ‘actual’ communications, relationships, and identities become virtually indistinguishable.” (LKGWC, p.341)
“(…) students differ tremendously in the extent to which they ‘act out their social lives’ on Facebook: both the level of SNS participation and the meaning of this activity undoubtedly vary across individuals and settings.” (Ibid.)
[1] Remark: “Mayer and Puller (2008) report (…) that only 0.4% of the Facebook friendships they studied appeared to reflect ‘merely online interactions’. This finding is supported by other research indicating that Facebook is used primarily to maintain or reinforce existing offline relationships rather than to meet new people (Ellison et al., 2007).” (LKGWC, p.332)
[2] Why the focus on posting photos?
“The act of publicly posting a photo of someone suggests that ego wishes her relationship with alter to be socially recognized, rather than simply enumerating her ‘friends’ or ‘close confidants’ to an interviewer in a private setting.” (LKGWC, p.333)
