Nonrecursive Models of Internet Use and Community Engagement: Questioning Whether Time Spent Online Erodes Social Capital
Shah, Dhavan, Schmierbach, Michael, Hawkins, Joshua, Espino, Rodolfo, Donavan, Janet, Journalism and Mass Communication Quarterly
Although some argue that Internet use may erode involvement in public life, the most common Internet behaviors, social communication and information searching, may actually foster social and civic participation. To examine this possibility, we test a series of non-recursive models using a national survey of nearly 3,400 respondents. Two-stage least squares regressions were performed to simultaneously test the reciprocal relationship between frequency of Internet use (i.e., hours per day) and three sets of community engagement behaviors: informal social interaction, attendance at public events, and participation in civic volunteerism (i.e., annual frequency). Time spent online has a positive relationship with public attendance and civic volunteerism. No evidence of time displacement from frequency of Internet use is observed.
A great deal of attention has been paid to the decline in Americans' participation in social and civic life. Available evidence suggests that while contributions to charitable groups are at all-time highs, face-toface encounters in our communities have slid to a forty-year low. Measures of informal socializing indicate that people are visiting friends, playing cards, having dinner parties, and going out to bars at substantially lower levels than a generation ago. At first glance, levels of volunteering counter this trend; however, cohort analyses suggest that older Americans bear a disproportionate amount of the service burden. And although attendance at public events has remained high-even increased-it cannot match the sharp rise in privatized entertainment, particularly television. It seems, then, that between 1960 and 2000, we have drifted from being a nation of joiners to a nation of watchers, with the youngest Americans the most detached from public life.1
Political participation has also declined, with fewer than half of Americans voting in recent national elections, and reduced numbers working for campaigns and running for political office.2 Research on social capital links these indicators of community health at the aggregate and individual level by conceiving of political participation as a "byproduct of activities engaged in for other purposes."3 Putnam, drawing on Coleman, defines social capital as "features of social life-networks, norms, and trust-that enable participants to act together more effectively to pursue shared objectives."4 Inglehart5 has found that aspects of social life, such as spending time with friends and participating in community life, may strengthen social networks and reinforce norms of reciprocity, Thus, both formal and informal community ties can improve the health of civil society. Individuals who are connected and confident about the return of their social investments feel a greater sense of belonging to their communities and take a more active role in politics.6
Scholars and pundits alike have looked to the media for clues on the production and destruction of social capital. Newspapers, with their focus on news and community events, produce pro-civic consequences; newspaper readers, especially those who pay attention to local news content, are more participatory and politically knowledgeable than nonreaders.7 Conversely, television has been blamed for civic disengagement because time spent with it supposedly displaces social activities and cultivates a psychological barrier to participation.8 Scholars have questioned whether television use, writ large, creates an obstacle to participation.9
Nonetheless, some technologists and social critics surmise that Internet users also become increasingly removed from meaningful social relationships and less likely to engage the community as they spend more time online.10 Field research-the little there is-provides some support for this pessimistic view; frequent Internet use has been related to withdrawal from family and community life within a pair of recent panel studies.11 However, these studies provided subjects with free Internet access and Net devices and then assessed social effects. …