Antecedents and Civic Consequences of Choosing Real versus Ersatz Social Activities

Abstract

Social capital (Coleman, 1990), democracy's wellspring (Putnam, 1993; Tocqueville, 1835/2000), must now accrue within a social fabric that has been changed by the ubiquitous adoption of technologies such as television and the Internet. Three studies (N = 313) of college students supported a theory of ersatz social behavior, which proposed that: a) the use of technological (replacement/ersatz) alternatives to real social interaction may be less conducive to the skill building and opinion formation that foster social capital; b) ersatz activities involve fewer costs (effort, risk of rejection) than real interactions; and c) ersatz alternatives are regularly chosen, even though real interaction is ostensibly preferred. The results provide increased understanding of when and how mediated interaction can be detrimental to social capital formation.

ACKNOWLEDGMENTS

We thank Marilynn Brewer and Philip Tetlock for their insightful suggestions regarding the research. We are indebted to Charles Bond Jr., Marilynn Brewer, Sara Kiesler, Stephen Livingston, Darcy Reich, John Sabini, and Sri Kalyanaraman for comments on an earlier draft.

Notes

1. Social capital's essential role in sustaining robust democracy has been underscored by Putnam (1993). Structures alone (constitutions, voting rights) are not enough (consider the plight of Iraq in 2007); appropriate societal conditions—including adequate social capital—are also needed for democratic governance to thrive. Whereas Tocqueville (1835/2000) viewed the United States experiment with democracy as an excellent exemplar of a new kind of governance, he did not insist that democracy was the only viable form. Later writers did so insist: “man's inclination to injustice makes democracy necessary” (Reinhold Niebuhr quoted in Schlesinger, 1998, p. 179). Recent statistical compilations of the magnitude of murder of civilians by their own governments (such killing has been termed “democide” by Rummel, 1994) have appeared to close the case in favor of democracy. Citizens targeted by a nondemocratic regime could not rely on international organizations or superpowers to save them (Rummel, 1994): The only sure bulwark against government annihilation has been indigenous, well-functioning democracy.

2. “Social capital” is protean (e.g., Coleman, 1990). Our definition and our operationalizations of civic aspects of social capital are driven by the writings of R. D. Putnam (e.g., Putnam, 2000; see also Ray, 1999; Sullivan & Transue, 1999).

3. Putnam (2000) suggested that electronic entertainment accounted for roughly 25% of the decline in social capital (p. 283). Putnam also highlighted generational change, work pressures, and suburban sprawl as other important contributors to the decline in social capital, but we will not further address these factors here.

4. Putnam also warned about “cyberapartheid,” namely, “the most important cultural impact of computer-mediated communication could be the reinforcement of the culturally dominant social network” (Castells, 1996, p. 363); and about “cyberbalkanization”—the Internet enables us to confine communication to people who share precisely our own interests; in contrast, “real-world interactions often force us to deal with diversity” (Putnam, 2000, pp. 175–177).

5. For a detailed discussion of similarities and differences between organizational and informal activities, see Green & Brock, 2005.

6. Two independent coders classified the activities. Disagreements were resolved by the first author. Overall agreement between the original two coders across all activities listed was 81%. The majority of disagreements occurred on a small set of ambiguous activities (e.g., shopping, lifting weights) that were consistently coded into different categories by the two coders.

7. We did not have sufficient numbers of cases to test differences between interactive and noninteractive ersatz in Study 2 (only four participants listed an interactive ersatz activity in the first position.)

8. A similar finding emerged in another series of studies we conducted, in which participants actually selected between real, ersatz, and purely nonsocial activities. Participants (365 undergraduate students) completed a brief personality survey and then, to reflect upon their survey experience, chose one of three options: interacting online with real other participants, interacting with “chatterbots,” or private typing. Selecting “real participants” meant that they would take part in an online discussion with other students like themselves. The chatterbot option was described as an interaction with “semi-intelligent computer simulations that carry on interactions with humans. These simulations are programmed with a large vocabulary, and look for words or phrases in sentences to provide appropriate responses.” Private typing was an option for those who wished to organize their thoughts on their own rather than interacting with either machines or humans. Although most chose real participants (42%), and fewer chose private typing (20%), more than a third of participants (38%) chose machine (chatterbot) conversation. Follow-up analyses showed that the robust machine preference was stable and could not be attributed to participants' mood, to their perception of text-messaging difficulty, or to novelty of chatterbot. See Green, 2000 for details.

9. A complete list of alteration options for all activities is available from the authors on request.

10. Because participants could only choose one option for each activity, these tests are not independent. We also analyzed these data with frequency tables (real versus ersatz as rows; risk, effort, immediacy, and no change as columns), and this analysis also supported the finding of differences across conditions, χ²(3) = 119.31, p < .001. The lack of independence may also be a less serious problem because aggregate scores were used. A participant assigned an alteration option (effort, risk, immediate gratification, none) to each of ten activities (1–10). Whereas assigning an option to activity 1 (“effort”, say) precluded using “risk” and “immediacy” for activity 1, the option chosen for activity 1 did not determine option choice for other activities (2, 3, etc.). Rather than predilections for a kind of alteration (e.g., “immediate gratification”), the data showed that chosen alterations were driven by the nature of the activity, ersatz versus real (Figure 3).