«NETWORKING IN EVERYDAY LIFE by Bernard J. Hogan A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy ...»
Even presidential candidates have a direct line regardless of place, as 2008 Republican aspirant Rudolph Giuliani illustrated with notorious (and possibly staged) cell phone calls from his wife during campaign speeches (Dobbs, 2007). But cell phone numbers are not listed in a comprehensive and searchable directory. Neither are emails nor instant message handles. Moreover, even if one’s landline number is listed, one might be at the ofﬁce under a different number. From one perspective, this is a complete mess. Gone are the earlier days where everyone’s name was listed neatly in a phonebook found on every desk, or in every kitchen. Instead, people rely on a hodgepodge of address books, post-its, memory, past email messages, numbers scribbled on backs of used envelopes, and infrequent calls to others asking “do you happen to have her number?”1 From another perspective though, this is a sensible response to a need to regulate social accessibility. The unsolicited contact industry is big business. Easily searched directories mean easily exploited targets for direct mail, telemarketers, and See pages 184-187 of Wellman and Hogan (2006) for a more complete discussion of the use of such tools to manage contact information.
spam. Both the Canadian government and the American government have put in place standards to ensure this privacy (Katz, 2006; Gatehouse, 2005). Now privacy, vis-´ -vis the choice to whom people give their number, has greater value than the dia rectory. Now, individuals selectively give out their contact numbers to others on the assumption that it helps regulate their availability. For example, Licoppe and Heurtin (2001) show that increased exchange of mobile phone numbers between individuals leads to a very clear increase in communication between those individuals.
Within this context of shifting accessibility and the unsettling adoption of new technologies, I can ask if there are any guiding themes to the way individuals organize their access to each other, and by what media. Several scholars have made compelling contributions to this question. Many of these contributions rest on qualitative assertions of actors as ‘rational’ in how they decide which media to use, and to a lesser extent who they use it with (Madell and Muncer, 2005; Kling, 1980). Among them, Haythornthwaite has made an especially relevant contribution vis-a-vis her doctoral work and subsequent study of distance learners. Along with Wellman, she coined the term “media multiplexity” to stand for the way individuals have multiple points of access across several media (Haythornthwaite and Wellman, 1998). She contends that there is a strong relationship between the tie strength between ego and alter and the number of media shared between ego and alter. In short, the stronger the tie, the more media used.2 This hypothesis offers a powerful explanation for the usage patterns of media within personal networks. Moreover, the relationship between media use and tie strength was presaged in Zerubavel’s original discussion of social accessibility.
[G]iving others one’s telephone number—especially if it is not listed in the A note on language is apt here. When I discuss tie strength, I am referring to socioemotional closeness, the datum used throughout this dissertation. However, given that I discuss both socioemotional closeness and spatial proximity (physical closeness), it is perhaps easier on the reader if I simply refer to tie strength and proximity, and use the word “closeness” as little as possible.
CHAPTER 7. MEDIA USE WITH SPECIFIC NETWORK MEMBERS 180telephone directory—is a most signiﬁcant act of displaying accessibility to them. Aside from the practical signiﬁcance of granting them actual access to one, it also serves the function of symbolically incorporating them into a selective and exclusive social circle of intimates (Zerubavel, 1985, 145-6).
However, it is really the case that individuals give greater access to those with whom they have stronger ties? Since tie strength is a multidimensional construct combining frequency, reciprocity, affect, and self-disclosure (Granovetter, 1973; Marsden and Campbell, 1984), do these factors congeal into a single force acting on the way individuals regulate their accessibility? Or rather, is it the case that media multiplexity has previously been under-speciﬁed. Granted, tie strength may correlate with increased access and use of media, but that is not the whole story. Rather, several factors such as the embeddedness of the alter, their proximity to ego, their role and even the age of ego and alter can play a part in explaining differences in the number of media used. Thus, I conclude this chapter with a discussion of a more general paradigm that incorporates many of Haythornthwaite’s ideas (such as latent tie theory) into a more general understanding of networking as the regulation of social accessibility.
This chapter again uses data from the Connected Lives project. Like the last chapter on the organizing principles of networks, this chapter relies on the social networks captured during the interviews. For each interview we performed a ‘minisurvey’ of selected alters in the networks. These alters were carefully selected using a speciallydeveloped algorithm that ensures there is equal weight given to a diverse set of alters that were both very close and somewhat close. Details about this particular sampling technique have been addressed in Chapter 4 as well as Hogan et. al. (2007). In short, I assert that these sampled individuals fairly represent the social network as a whole.
The minimum number of minisurveys per network is 3 (which was every individual included in one respondent’s small network) up to 15 (which was the speciﬁed cut-off
age, spatial proximity and media use.
7.2 Media multiplexity hypothesis explained In “Social Networks and Internet Connectivity Effects”, Haythornthwaite (2005) lays out a coherent and extensive set of theoretical propositions about the ways that the Internet may shape personal networks and networking. This is not done with the Internet as the focal point, but as one of many ways in which individuals communicate.
This is an example of the sort of holistic analysis of communication media that I seek to pursue. These propositions are based on prior work she has done with Wellman and other collaborators about the role of media use within an organization. Three
propositions are as follows:
• Media multiplexity: Those more strongly tied use more media.
• A unidimensional scale of media use: There is a single unidimensional scale of contact with peers akin to a Guttman scale of media use. Guttman scales are ways of ordering preconditions. For example, one would have a telephone before using email, and then use email before using instant messenger.
• Latent tie theory: The addition of a new media into an existing media ecology enables individuals to access alters that they would not otherwise access.
These three propositions lead to a perspective of media use that I term the “Fort Knox theory of social connectivity”. Everyone who works at Fort Knox gets a key to the front door. Likewise, everyone in the personal network gets to interact with ego by one keystone medium/context, presumably in-person interaction. Those who are not merely weak ties also get access to the inner areas of the vault (i.e., they can access ego by the email) while those who are the strongest ties actually have access to
telephone and via email. Adding another media is like building another wall around Fort Knox—more people can get inside the ﬁrst door or maybe an alternate door, but it is still just as difﬁcult to get into the control room and the vault.
I believe there is a lot of truth to this metaphor. Individuals do guard their privacy and seek to maintain social accessibility. However, as was shown in Chapter 5 on media styles, it is difﬁcult to suggest that in everyday networks there is a single scale of media use, at least one that unilaterally applies to all sorts of networking styles.
For the group in that chapter that used all media infrequently (termed the “all light” group), there was an inverse correlation between frequency of use of mobile phones and email. Those who supplement the core of in person and telephone contact did so in different ways, suggesting a certain media preference. Also those individuals who mainly networked in person used cell phones more than landlines to supplement their planning. Hence, one can see that people prefer to use a select combination of media that is conditioned by the alters they need to access, but is neither determined by them, nor by a single unifying scale. So in that chapter I referred to an ecology of media styles, rather than a nesting of media styles. Similarly, in everyday networks, as opposed to academics or students, the mere reason for being in the network can vary signiﬁcantly. In Haythornthwaite’s studies, as well as the study by Koku et al.
(Haythornthwaite, 2005; Koku et al., 2001), individuals are all at least ‘peers’, either as academics or students. In everyday life, however, there is not the same base reason for being included in the network—some are friends, others family and others still are organizational members. This means that the institutional norms that might encourage email as a ﬁrst point of contact are not so obvious in everyday life; some alters may love email while others do not own a computer. On one hand this makes Haythornthwaite’s studies perfectly suited to testing the media multiplexity hypothesis for they are relatively homogeneous populations all starting from a reasonably static baseline.
that are present in everyday life which might either be absent or muted given the scope conditions of a more controlled organizational setting. Finally, there is the matter of a rather dyadic sentiment, both in the metaphor and in the original work. In Haythornthwaite and Wellman (1998), dyads in the network were considered, although the analysis did not control for the relative prominence of individuals.
One issue with the Fort Knox metaphor, and the theory it represents, is that it seems very dyadic. It is like everyone has their own speciﬁc keychain for Fort Knox, and ego is the only person who decides which keys a person holds. However, individuals are not merely accessed through speciﬁc pre-ordained media. They are also accessed through triads. A triad means that one party can relay messages between the other two. It also means that one party can relay contact information between the other two. For example, it is common for me to ask my mother for the current contact information of many of my relatives. This is also borne out in the data as 72 percent of the survey respondents report that they ask someone else for a phone number at least monthly. Thus, it seems likely that there is an under-analyzed structural dimension to media multiplexity and that by specifying this dimension I can more fully explain media multiplexity.
Bearing in mind the above theoretical discussion, I present ﬁve hypotheses about
the media multiplexity hypothesis:
• Hypothesis 1: Individuals will have role-speciﬁc values for media multiplex
As mentioned above, Haythornthwaite’s original work was done among individuals who shared very similar roles, generally as academics. However, in everyday life, individuals are kin, friends, neighbours, and organization members.
This should lead to differences in levels of accessibility.
Above I referred to this analysis as moving beyond dyads towards network structure. Highly connected alters may be gatekeepers to one’s personal network. Regardless of the strength of the relationship between ego and alter, if alter is highly connected, then alter may serve as a gatekeeper in that network.
As a gatekeeper, ego may rely on them for contact with other network members,
This suggests that media have a primarily instrumental purpose for in-person social coordination. Those who are nearer will beneﬁt more from using many media, since it means more ways to coordinate with these individuals. This is especially relevant when people need to access each other in a short time frame— which is more common among individuals who are trying to jointly coordinate in-person social activity. For example, if I am waiting for a friend at a coffee shop and if I cannot reach that friend by calling her landline, then it helps to have her cell number or email address in the hopes that I can reach her.
• Hypothesis 4: Alters in more frequent in-person contact will share greater numbers of media with ego.
This hypothesis works with the above hypothesis that the more frequent individuals see each other in person, the more they will want to reinforce their meetings with mediated contact.
• Hypothesis 5: Controlling for the above factors, tie strength will not be a signiﬁcant predictor of the number of media used with alter.
works. If they sufﬁciently capture the relevant dimensions of tie strength as it relates to the number of media used, then tie strength will no longer be significant. However, if the signiﬁcance of tie strength persists, then there are other aspects of tie strength that are not adequately speciﬁed in this model.