Digital Inequality and Participation in the Political Process

Abstract

This study analyzes whether greater levels of Internet access closed the participation gap between individuals of lower and higher socioeconomic status (SES) in the early stages of the 2012 presidential campaign. Our analysis of data from the Pew Research Center demonstrates that greater levels of access to the Internet are significantly associated with greater political knowledge and engagement for low SES individuals, but not high SES individuals. We explain our results in the context of incidental learning among the disengaged public during high-profile political events, such as a presidential election.

Keywords

social inequality Internet digital divide political participation

Political voice is any activity used by an individual or group that is intended to influence government action (Schlozman, Verba, & Brady, 1995). In the normative context of establishing quality representative governance in the United States, one of the most revered democratic values is equality of political voice for each American (Dahl, 1971). However, the lack of equality of political voice across socioeconomic status (SES) persists in America (Bartels, 2010; Jacobs & Skocpol, 2005; McCarty, Poole, & Rosenthal, 2006). The wealthy participate more in the political process than those at the bottom of the economic ladder, and the result is a government comprised of elected officials more responsive to the policy preferences of the rich than everyday Americans (Gilens, 2005; Schlozman, Verba, & Brady, 2012).

Throughout the 20th century, many observers, researchers, and activists looked at technological advances in communication as a method of closing the gap in political influence across SES (Margolis & Resnick, 2000). As the century came to a close, many looked at the Internet as the next best democratic hope in this regard. As an open-access mass communication tool, the Internet created the possibility of a forum for any single person or group to be heard by all. One researcher suggested that “the political significance of [the Internet] lies in its capacity to challenge the existing political hierarchy’s monopoly on powerful communications media, and perhaps thus revitalize citizen-based democracy” (Rheingold, 1993).

Has the Internet realized this democratic potential? With Internet penetration rate currently topping 80% (Pew Research Center, 2012), the question of whether or not the Internet has helped equalize political influence in America remains important, particularly in context of a national presidential election campaign. Presidential campaigns now depend heavily on the Internet to organize, fundraise, and disseminate information. In addition, coverage of presidential campaigns on the Internet is extensive, outreaching simple mainstream political media. Our assertion is that access to this information via the Internet allows citizens to capitalize on the resources necessary to actively learn about the politics surrounding a presidential election and engage in the process. The more opportunity an individual has access to Internet technology, the greater their chance to express their political voice. Some research has found that access to the Internet associates with higher levels of political knowledge and engagement (Kenski & Stroud, 2006; Kobayashi, Ikeda, & Miyata, 2006; Shah, Cho, Eveland, & Kwak, 2005; Tolbert & McNeal, 2003; Weber, Loumakis, & Bergman, 2003). Other researchers are more skeptical (Bimber, 2001; Davis, 2005).

Our research question asks whether or not Internet access can serve to close the knowledge and participation gap between low and high SES individuals. We address this question within the context of the early stages of the 2012 Republican presidential nomination season. Forty years since the presidential nomination process was transformed from a convention-based system to the current primary system, scholars have focused heavily on the media’s role in primary campaigns (Ansolabehere & Iyengar, 1995; Orren & Polsby, 1987; Patterson, 1993), particularly on the coverage of the early contests in Iowa and New Hampshire (Adams, 1987; Belt, Just, & Crigler, 2012). Although some research has looked at the Internet’s role in political knowledge and engagement during nomination campaigns (Baumgartner & Morris, 2010), we are aware of no work that has considered how SES and levels of Internet access interact in presidential election campaigns.

The following section will outline the competing perspectives on whether or not differing levels of access to the Internet work to close or exacerbate the political knowledge and participation gap that exists between the wealthy and the less well-off. We then make the case that higher levels of Internet access may serve to close the knowledge and engagement gap between low and high SES individuals during a presidential election campaign. We test the interaction between Internet access and SES with the purpose of understanding their effect on knowledge of the primary campaign and political participation. We close with a discussion of the significance of our findings as they pertain to how the Internet and other forms of information technology may factor into the future political world, especially for those of lower SES in America.

Digital Divide: SES, Access, and Skill

As the era of the Internet began in the early 1990s, many speculated that the medium might serve to bridge territorial, demographic, ideological, and cultural divides by creating virtual communities and spurring social movements (Grossman, 1995; Kush, 2000; Rheingold, 1993; Schwartz, 1996). There is some evidence that confirms this potential has been partially realized in promoting democratization abroad (Hill & Sen, 2005; Ott & Rosser, 2000). However, evidence of unequal access across lines of culture, SES, race and ethnicity (Jones, Johnson-Yale, Millermaier, & Perez, 2009), gender (Wilson, Wallin, & Reiser, 2003), and geographic location (Sylvester & McGlynn, 2010) prompted many researchers to speculate that a “digital divide” had emerged (Norris, 2001).

Social scientists have also focused on whether the Internet has closed the participation gap between low and high SES individuals (DiMaggio, Hargittai, Celeste, & Shafer, 2004). While some observers suggest the digital divide continues to shrink as the costs of Internet access drops and supply grows (Compaine, 1988; Thierer, 2000), others contend that a gap in political knowledge and participation persist, and the Internet has failed to close it (Schlozman et al., 2012). Research on SES and the digital divide have focused on two dimensions: Internet access and Internet skill (DiMaggio et al., 2004; Hargittai, 2003; Hargittai & Hinnant, 2008; Mossberger, Tolbert, & Stansbury, 2003). The access gap has been characterized as the first-level divide, while the skill gap has been referred to as a second-level digital divide (Hargittai, 2002, 2008).

The growing accessibility of the Internet even among low SES individuals in America has prompted many scholars to look past the first-level access divide and develop explanations of a second-level divide of Internet skill. Those who research the second-level skill divide assume the access gap is closing and focus instead on individual motivations to gather political information as well as the ability to retrieve the information necessary to contribute to the political process (Min, 2010). Some work even suggests that a “usage gap” is developing between those who use the Internet for entertainment and those who use it for education and work (van Dijk, 2005). Low SES individuals trend more toward entertainment usage of the Internet and lack the skill to locate information relevant to democratic citizenship (Hargittai & Hinnant, 2008; van Dijk, 2005).

But has the first-level access divide really closed along the lines of SES? If one conceptualizes Internet access as a dichotomous variable, then the answer to this question would be yes. Recent data from the Pew Research Center (2012) show that as of 2012, 83% of Americans use the Internet at least occasionally. But, should we conceive of Internet access as a simple yes/no variable? We argue no. Indeed, the access divide has often been discussed in context of a scale, with one side of the spectrum being those who have no Internet at home or work and the other being those who are highly connected via multiple platforms (Schlozman et al., 2012; Wilhelm, 2000). Just as a speedometer on an automobile must offer more data points than simply “stop” and “go,” it is our contention that occasional Internet access must be differentiated from heavier levels of access through tablets, mobile devices, and social networking connectedness. We are interested in examining how various levels of Internet access might correlate with greater levels of political voice among those of lower SES in the context of the 2012 presidential election.

Why focus on Internet access as opposed to Internet skill? First, as mentioned above, it is our sense that access has been overlooked as a key independent variable in the last several years of research on the digital divide. Second, research into the second-level digital divide has used political motivation and skill accessing political information as key independent variables to predict political engagement. This is problematic in that elements of the dependent variable are too closely intertwined with the independent variables, thus bringing parts of the right-hand side of the equation into the left-hand side. In other words, we see political motivation and skill as elements of political engagement. It is our intention to keep political engagement on the right-hand side of the equation as the dependent variable and look only at levels of Internet access, not political Internet use, as the key independent variable. Third, as we discuss below, there is reason to believe that in and of itself, access can work to close the participation gap between low and high SES groups. We discuss this reasoning below.

Internet Access and Incidental Learning in Presidential Elections

How could greater levels of Internet access translate into closing the political participation gap between individuals of lower and higher SES? As the previous section discussed, researchers who have explored the notion of a second-level digital skill divide suggest that access will not translate into quality political participation unless the user has the political motivation and skill to access information (Hargittai, 2002, 2008; Hargittai & Hinnant, 2008). Theoretically, however, there is reason to believe that greater access can translate into greater political knowledge and engagement during large-scale political events, even when the motivation and skill may be lacking. This theory is grounded in the incidental by-product model of learning and the political engagement gateway hypothesis (Baum, 2003).

The incidental by-product model suggests that when media coverage of high-profile political events transcends traditional news media and spills over into nontraditional forms of media, those who know little about politics have the opportunity to learn (Baum, 2003). This learning process is considered a by-product because the individual is not seeking political information, but instead accidentally stumbles across it while (in context of Internet browsing) in search of entertainment or online social networking. Therefore, the more access an individual has to the flow of information via the Internet, the greater the chance that individual will accidentally come across coverage or discussion of political news regardless of whether they are seeking it. This is especially important for lower SES individuals who may have fewer informational or politically engaged social resources. For them, the Internet may serve as a unique gateway to political information and discussion.

The gateway hypothesis argues that individuals who accidentally learn about political issues or events through nonnews-related media sources may then gain motivation to pursue more detailed information (Baum, 2003). Thus, accidental exposure to nontraditional media coverage of a high-profile political event such as a presidential election has the potential to spur political interest, which can in turn lead to a greater willingness to express one’s political voice.

The process of incidental learning and the gateway hypothesis apply primarily to those who are disengaged from politics, particularly those of lower SES. Those of higher SES are more likely to be informed and engaged in the first place, so accidental exposure to political information will have little impact. Thus, we hypothesize that higher levels of Internet access among low-SES individuals will positively affect the willingness to express political voice, thus serving to close the gap between citizens of low and high SES.

Data and Method

We used data from the Pew Research Center for the People and the Press 2012 Political Communications Study. The study interviewed 1,507 respondents aged 18 and over. Of these, approximately 900 were contacted via landline and 600 were contacted via cell phone. Interviews were conducted in English and Spanish and were completed from January 4–10, 2012, during the height of the early 2012 Republican nomination season, a time period corresponding to coverage of the Iowa caucuses and the New Hampshire primary.

We were first interested in Internet access, a composite measure of access ranging from 0 to 5. Five dichotomous measures (0 = no, 1 = yes) were combined to create this measure. Respondents were asked: (1) “Do you use the Internet, at least occasionally?” (2) “Do you send or receive e-mail, at least occasionally?” (3) “Do you access the internet on a cell phone, tablet or other mobile handheld device, at least occasionally?” (4) “Do you ever use Twitter?” and (5) “Do you use social networking sites, such as Facebook, Google Plus or LinkedIn?” Our index ideally measures Internet use by capturing both access to the Internet and the overall degree of a person’s online presence. Those who have access through a variety of devices and use a variety of online resources are online more often, while those who do not have access or use such online resources are not.

SES, our primary independent variable, is a composite measure of respondent’s educational attainment and total family income in 2011. For educational attainment, eight choices were provided: (1) less than high school; (2) high school incomplete; (3) high school graduate; (4) some college, no degree; (5) 2-year associate or technical degree; (6) 4-year degree; (7) some postgraduate training; and (8) postgraduate or professional degree. For total family income, nine categories were provided: (1) Less than $10,000; (2) 10,000 to less than $20,000; (3) 20,000 to less than $30,000; (4) 30,000 to less than $40,000; (5) 40,000 to less than $50,000; (6) 50,000 to less than $75,000; (7) 75,000 to less than $100,000; (8) 100,000 to less than $150,000; and, (9) $150,000 or more. Each category of income was recoded to the midpoint of the range of income it represented (e.g., 1 = $5,000, 2 = $15,000). Recoding to the midpoint provides a sense of the distribution of the income variable used in our study. To create the measure for SES, both educational attainment and income were standardized to z-scores, combined into one measure, and standardized into a z-score.

Political knowledge is an index of responses to four questions measuring respondents’ knowledge of the candidates and processes involved in the 2012 Republican nomination. For each question, a correct answer was coded 1 and an incorrect answer was coded 0. Respondents were asked four questions: (1) “Was Mitt Romney governor of … ” (a = Massachusetts, b = Utah, c = Michigan, and d = Texas); (2) “Which Republican candidate served as Speaker of the U.S. House of Representatives?” (a = Newt Gingrich, b = Rick Santorum, c = Jon Huntsman, and d = Ron Paul); (3) “Which Republican candidate opposes U.S. military involvement in Afghanistan” (a = Newt Gingrich, b = Rick Perry, c = Rick Santorum, and d = Ron Paul); and (4) “After Iowa and New Hampshire, which state holds the next Republican primary?” (a = California, b = New York, c = Ohio, and d = South Carolina). The total number of correct answers given was totaled for each respondent to gauge political knowledge.

Voting regularity is an ordered 4-point scale that measures how often respondents vote. Respondents were asked, “How often would you say you vote?” (1 = never vote, 2 = seldom, 3 = part of the time, 4 = nearly always, 5 = always). Options 4 and 5 were collapsed into one category, so that the uppermost category captures people who regularly vote in elections. This was necessary because some might claim they “nearly always” vote only because they sometimes do not vote in special or certain local elections, for example, while others might display the same voting behavior but say that they “always” vote.

We included several control variables in our models as well. Interest in politics measures respondents’ attention to elections and politics and asks, “And how much do you enjoy keeping up with political news about campaigns and elections” (1 = not at all, 2 = not much, 3 = some, 4 = a lot). Age is a continuous measure of each respondent’s age. Race was recoded as four dummy variables: White (reference category), Black, Hispanic, and other race. Region was also recoded into dummy variables: South (reference category), Northeast, Midwest, and West.

To carry out our study, two types of regression analyses were used: negative binomial regression and ordered logistic regression. Negative binomial regression was used to analyze political knowledge, and ordered logistic regression was used to analyze voting regularity. The analysis of the relationship between Internet access and SES and both dependent variables includes three models: a bivariate model, a main effects model with controls, and an interaction model. Table 1 provides means and standard deviations (SDs) for all independent and dependent variables included in our study to establish average levels of Internet access, social standing, demographic background, and political knowledge and behavior for our sample. Sample weights were applied to more accurately resemble the U.S. population in terms of race, gender, and age.

Table 1.

Means and Standard Deviations.

	N	M	SD	Minimum	Maximum
Internet access	1,503	2.70	1.52	0	5
Socioeconomic status (SES)	1,292	0	1	−1.99	2.12
Interest in politics	1,495	2.71	1.02	1	4
Male	1,507	0.46	0.50	0	1
White (referent)	1,477	0.75	0.43	0	1
Black	1,477	0.10	0.29	0	1
Hispanic	1,477	0.09	0.29	0	1
Other race	1,477	0.06	0.24	0	1
Northeast	1,507	0.17	0.37	0	1
Midwest	1,507	0.24	0.43	0	1
South (referent)	1,507	0.35	0.48	0	1
West	1,507	0.25	0.43	0	1
Age	1,468	50.53	17.80	18	96
Political knowledge	1,507	2.06	1.52	0	4
Voting regularity	1,491	3.48	0.96	1	4

Results

Table 2 displays the results of the negative binomial regression analysis examining the relationship between Internet access and SES and political knowledge. Model 1 shows the bivariate relationship between Internet access and political knowledge. The two are significantly and positively correlated. Model 2 adds SES and controls to the analysis. Not surprisingly, SES is positively associated with a statistically significant increase in political knowledge (.22). Additionally, even after controlling for SES, interest in politics, age, and other background characteristics, we still see a statistically significant positive relationship between Internet access and political knowledge. This suggests that the positive link between Internet use and knowledge about politics is not an artificial relationship caused primarily by educational attainment, income, or interest in following political news. In other words, there does appear to be a genuine link between Internet access and political knowledge.

Table 2.

Political Knowledge Regressed on Internet Access and Socioeconomic Status (SES).

	Model 1	Model 2	Model 3
	Coefficient/SE	Coefficient/SE	Coefficient/SE
Internet access	.07*** (0.02)	.09*** (0.02)	.08*** (0.02)
SES		.22*** (0.02)	.31*** (0.05)
Interest in politics		.23*** (0.02)	.23*** (0.02)
Male		.30*** (0.04)	.29*** (0.04)
Black		−.34*** (0.08)	−.34*** (0.08)
Hispanic		−.62*** (0.13)	−.63*** (0.13)
Other race		−.28** (0.10)	−.28** (0.10)
Northeast		−.03 (0.06)	−.02 (0.05)
Midwest		.02 (0.05)	.01 (0.05)
West		−.09 (0.06)	−.09 (0.06)
Age		.01*** (0.00)	.01*** (0.00)
Access × SES			−.03* (0.01)
Intercept	.37*** (0.06)	−.98*** (0.13)	−.94*** (0.13)
ln alpha
Intercept	−1.40*** (.17)	−16.14*** (0.08)	−17.93*** (.34)
N	1,503	1,260	1,260
Log psuedolikelihood	−8159	−5949	−5942

Note. Numbers in parentheses are standard errors.

Reference categories: race = White, region = South.

*p < .05. **p < .01. ***p < .001.

Having demonstrated a positive link between Internet access and political knowledge and between SES and political knowledge, the next task is to examine whether the link between access and knowledge varies by SES. If access provides a unique gateway that helps compensate for a disparity in access to information between lower and higher SES individuals, then we would expect disadvantaged individuals to gain more political knowledge through their access. To examine this, Model 3 in Table 2 adds the interaction term between Internet access and SES, and it is statistically significant and negative. This suggests that an increase in SES is associated with lower returns on political knowledge from Internet access (−.03). In other words, lower SES individuals gain relatively more knowledge from access to the Internet. This contradicts previous studies that suggest the presence of a “digital reproduction of inequality” in which the ubiquity of Internet use is likely to increase the knowledge gap between the haves and have-nots (Hargittai, 2008; Zillien & Hargittai, 2009).

To provide an alternative manner in which to view these results, we examined predicted political knowledge based on the different degrees of Internet access for varying levels of SES. Figure 1 shows the increase in knowledge associated with access for the most disadvantaged respondents (2 SDs below the SES mean), those at the mean of SES, and the most advantaged respondents (2 SDs above the SES mean), while holding all other measures constant at their means.

Figure 1.

Political knowledge and Internet access by socioeconomic status (SES).

Figure 1 clearly shows that Internet access is positively associated with political knowledge, regardless of SES. However, the slope is steeper for the lower SES respondents compared to the highest SES respondents. Although higher SES people are more politically knowledgeable than their lower SES counterparts, the political knowledge gap shrinks as Internet access increases. This reduction in the knowledge gap is possibly due to two factors: (1) Lower SES individuals are exposed to new information online which increases their political knowledge and (2) higher SES individuals start at a high level of political knowledge, even with no Internet access, and therefore have very little room for an increase in political knowledge. Our contention is that a combination of both phenomena results in the reduction of the socioeconomic political knowledge gap.

Table 3 displays the results of the ordered logistic regression analysis examining the relationship between Internet access and SES and voting regularity. All coefficients are odds ratios. Model 1 shows the bivariate relationship between Internet access and voting regularity by presenting the likelihood that individuals always or nearly always vote compared to part of the time, seldom, or never. The two are significantly and positively related. Model 2 adds SES and controls to the analysis. SES is statistically significant and positively related to voting regularity (2.33). Individuals who are 1 SD higher on the SES scale are more than twice as likely to vote always or nearly always compared to their counterparts who are lower on the SES scale. In terms of Internet access, a one-step increase in access is associated with a 27% increase in the odds of voting always or nearly always, even after accounting for SES, interest in politics, and other controls.

Table 3.

Odds of Voting Regularly Regressed on Internet Access and Socioeconomic Status (SES).

	Model 1	Model 2	Model 3
	Coefficient/SE	Coefficient/SE	Coefficient/SE
Internet access	1.10* (0.05)	1.27*** (0.08)	1.11 (0.10)
SES		2.33*** (0.27)	3.82*** (1.07)
Interest in politics		1.78*** (0.15)	1.80*** (0.15)
Male		.75 (0.12)	.72* (0.11)
Black		.64 (0.15)	.65 (0.15)
Hispanic		.65 (0.17)	.65 (0.18)
Other race		.38** (0.13)	.39** (0.13)
Northeast		.81 (0.20)	.81 (0.20)
Midwest		1.02 (0.21)	1.00 (0.20)
West		1.14 (0.25)	1.15 (0.25)
Age		1.06*** (0.01)	1.06*** (0.01)
Access × SES			.86* (0.07)
Cut 1	.15*** (0.02)	3.02* (1.44)	1.87 (1.02)
Cut 2	.44*** (0.06)	13.77*** (6.58)	8.58*** (4.63)
Cut 3	.67** (0.09)	27.38*** (13.33)	17.12*** (9.37)
N	1,487	1,252	1,252
Log psuedolikelihood	4610	3220	3210

Note. Numbers in parentheses are standard errors.

Reference categories: race = White, region = South.

*p < .05. **p < .01. ***p < .001.

Does the relationship between Internet access and voting vary based on SES? If there are differential online payoffs in favor of higher SES individuals, then we would expect higher SES individuals to see an increase in voting regularity with increased Internet access compared to their lower SES counterparts. Alternatively, if increased Internet access provides a unique resource that helps compensate for a disparity in access to information and social networks between lower and higher SES individuals, then we would expect disadvantaged individuals to see an increase in voting patterns with increased access. To test the soundness of these conflicting outlooks, Model 3 adds the interaction between Internet access and SES to the analysis. The odds ratio for the interaction is statistically significant and below 1 (.86) indicating that lower SES individuals see a greater return on voting regularity with increased Internet access. As with the relationship between access and SES and political knowledge, this finding conflicts with a “digital reproduction of inequality” perspective, at least in the context of the early 2012 primary campaign season (Hargittai, 2008; Zillien & Hargittai, 2009).

Having established a variation in the relationship between Internet access and voting regularity based on SES, we examined predicted probabilities of voting regularly based on differing degrees of access for varying levels of SES. Figure 2 shows the increase in the probability of voting always or nearly always associated with access for the most disadvantaged respondents (2 SDs below the SES mean), those at the SES mean, and the most advantaged respondents (2 SDs above the SES mean), while holding all other measures constant at their means. The most socioeconomically advantaged individuals are most likely to vote regardless of the level of Internet access. However, when access is 0 for all respondents, the gap between the probability of voting for the most advantaged (97%) and least advantaged individuals (16%) is over 80%. However, when Internet access is at 5, the gap between the probability of voting for the most advantaged (93%) and least advantaged individuals (58%) is reduced to 35%. The reduction in the voting gap is primarily due to the increased probability of voting for lower SES individuals, although there is a somewhat decreased probability of voting for higher SES individuals. It is possible that because the most advantaged respondents are already very likely to vote even when they have no online connection, increased Internet activity provides a distraction or is indicative of a lifestyle that compromises political participation.

Figure 2.

Probability of voting and Internet access by socioeconomic status (SES).

Conclusion

We set out to investigate whether or not Internet access closes the participation gap among low and high SES individuals. Specifically, we were interested in learning whether those who are more advantaged socioeconomically get more politically from their Internet access than those who are less advantaged. Our findings clearly suggest that lower SES individuals see higher returns on political knowledge and voting regularity from increased Internet access compared to higher SES individuals, meaning that Internet access helps to reduce the SES gap in political knowledge and participation. This supports the idea that the Internet can help serve as an equalizing agent for citizen involvement in the political process, rather than serve as a source of greater inequality (Hargittai, 2008; Zillien & Hargittai, 2009).

Prior studies have argued that higher SES Internet users are more likely to use the Internet to enhance social networks and travel to websites that offer information about politics, travel, finances, and the economy (Hargittai, 2008; Zillien & Hargittai, 2009). Based on these findings, authors of such studies conclude that these disparities in usage are likely to lead to increased social and political inequality. Based on the theoretical notions of incidental by-product model of political learning and the gateway hypothesis, we disputed these findings and argued that greater levels of Internet access offer opportunities for lower SES individuals to close the gap in political inequality in the context of the most recent presidential election campaign.

We argued that that those who are more educationally and financially disadvantaged get more out of their online access. Internet access and use of various online resources can provide unique assets to those lower on the SES scale, making Internet access quite beneficial. Alternatively, online resources might offer very little new information and social networking opportunities for those higher on the SES scale, thus making greater levels of access a redundant resource for these users.

Our findings suggest that greater access to the Internet can close the participation gap in the context of a major political event. Much of the time, however, the political scene is somewhat mundane. Future research should examine whether or not our findings hold up during times in which political news does not dominate the media. Also, future studies could differentiate between unique dimensions of access, including different forms of social networking. Pursuing this line of research will offer greater insight into the power of the Internet to close the participation gap in American politics.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Adams

W. C.

(1987). As New Hampshire goes … In Orren

G. R.

Polsby

N. W.

(Eds.), Media and momentum: The New Hampshire primary and nomination politics (pp. 42–59). Chatham, NJ: Chatham House.

Ansolabehere

Iyengar

(1995). Going negative: How political advertisements shrink & polarize the electorate. New York, NY: Free Press.

Bartels

L. M.

(2010). Unequal democracy: The political economy of the new gilded age. Princeton, NJ: Princeton University Press.

Baum

M. A.

(2003). Soft news goes to war: Public opinion and American foreign policy in the new media age. Princeton, NJ: Princeton University Press.

Baumgartner

J. C.

Morris

J. S.

(2010). Myfacetube politics: Social networking web sites and political engagement of young adults. Social Science Computer Review, 28, 24–44.

Belt

T. L.

Just

M. R.

Crigler

A. N.

(2012). The 2008 media primary: Handicapping the candidates in newspapers, on tv, cable, and the internet. The International Journal of Press/Politics, 17, 341–369.

Bimber

(2001). Information and political engagement in America: The search for the effects of information technology at the individual level. Political Research Quarterly, 54, 53–67.

Compaine

(1988). Issues in new information technology. Norwood, NJ: Ablex Publishing.

Dahl

R. A.

(1971). Polyarchy: Participation and opposition. New York, NY: Yale University Press.

10.

Davis

(2005). Politics online: Blogs, chatrooms, and discussion groups in American democracy. New York, NY: Routledge.

11.

DiMaggio

Hargittai

Celeste

Shafer

(2004). Digital inequality: From unequal access to differentiated use. In Neckerman

(Ed.), Social inequality (pp. 355–400). New York, NY: Sage.

12.

Gilens

(2005). Inequality and democratic responsiveness. Public Opinion Quarterly, 69, 778–796.

13.

Grossman

L. K.

(1995). The electronic republic. New York, NY: Penguin.

14.

Hargittai

(2002). Beyond logs and surveys: In-depth measures of people’s web use skills. Journal of the American Society for Information Science and Technology Perspectives, 53, 1239–1244.

15.

Hargittai

(2003). The digital divide and what to do about it. In Jones

D. C.

(Ed.), New economy handbook (pp. 821–839). San Diego, CA: Academic Press.

16.

Hargittai

(2008). The digital reproduction of inequality. In Grusky

(Ed.), Social stratification (pp. 936–944). Boulder, CO: Westview Press.

17.

Hargittai

Hinnant

(2008). Digital inequality: Differences in you adults’ sue of the Internet. Communication Research, 35, 602–621.

18.

Hill

Sen

(2005). The internet in Indonesia’s new democracy. New York, NY: Routledge.

19.

Jacobs

L. R.

Skocpol

(2005). American democracy in an era of rising inequality. In Jacobs

L. R.

Skocpol

(Eds.), Inequality and American democracy: What we know and what we need to learn (pp. 1–18). New York, NY: Russell Sage Foundation.

20.

Jones

Johnson-Yale

Millermaier

Perez

F. S.

(2009). U.S. college students’ Internet use: Race, gender and digital divides, Journal of Computer-Mediated Communication, 14, 244–264.

21.

Kenski

Stroud

N. J.

(2006). Connections between internet use and political efficacy, knowledge, and participation. Journal of Broadcasting & Electronic Media, 50, 173–192.

22.

Kobayashi

Ikeda

Miyata

(2006). Social capital online: Collective use of the internet and reciprocity as lubricants for democracy. Information, Communication, & Society, 9, 582–611.

23.

Kush

(2000). Cybercitizen: How to fight for all the issues you care about. New York, NY: St. Martin’s.

24.

Margolis

Resnick

(2000). Politics as usual: The cyberspace ‘revolution’. New York, NY: Sage.

25.

McCarty

Poole

K. T.

Rosenthal

(2006). Polarized America: The dance of ideology and unequal riches. Cambridge, MA: MIT Press.

26.

Min

S. J.

(2010). From digital divide to the democratic divide: Internet skills, political interest, and the second-level digital divide in political internet usage. Journal of Information Technology & Politics, 7, 22–35.

27.

Mossberger

Tolbert

C. J.

Stansbury

(2003). Virtual inequality: Beyond the digital divide. Washington, DC: Georgetown University Press.

28.

Norris

(2001). Digital divide: Civic engagement, information poverty, and the internet worldwide. New York, NY: Cambridge University Press.

29.

Orren

G. R.

Polsby

N. W.

(Eds. (1987). Media and momentum: The New Hampshire primary and nomination politics. Chatham, NJ: Chatham House.

30.

Ott

Rosser

(2000). The electronic republic? The role of the internet in promoting democracy in Africa. Democratization, 7, 138–156.

31.

Patterson

T. A.

(1993). Out of order. New York, NY: Vintage Books.

32.

Pew Research Center. (2012). Cable leads the pack as campaign news source: Twitter, facebook play very modest roles. Retrieved January 4, 2013, from http://www.people-press.org/2012/02/07/cable-leads-the-pack-as-campaign-news-source/

33.

Rheingold

(1993). The virtual community: Homesteading the electronic frontier. Reading, MA: Addison-Wesley.

34.

Schlozman

K. A.

Verba

Brady

H. E.

(1995). Voice and Equality. Cambridge, MA: Harvard University Press.

35.

Schlozman

K. A.

Verba

Brady

H. E.

(2012). The unheavenly chorus: Unequal political voice and the broken promise of American democracy. Princeton, NJ: Princeton University Press.

36.

Schwartz

E. A.

(1996). Netactivism: How citizens use the internet. Sebastopol, CA: O’Reilly Media.

37.

Shah

D. V.

Cho

Eveland

W. P.

Kwak

(2005). Information and expression in a digital age: Modeling internet effects on civic participation. Communication Research, 32, 531–565.

38.

Sylvester

D. E.

McGlynn

A. J.

(2010). The digital divided, political participation, and place. Social Science Computer Review, 28, 64–74.

39.

Thierer

(2000. How free computers are filling the digital divided (The Heritage Foundation Backgrounder, Number 1361). Washington, DC: Heritage Foundation.

40.

Tolbert

C. J.

McNeal

R. S.

(2003). Unraveling the effects of the Internet on political participation. Political Research Quarterly, 56, 175–185.

41.

van Dijk

J. A. G. M.

(2005). The deepening divide. London, England: Sage.

42.

Weber

L. M.

Loumakis

Bergman

(2003). Who participates and why? An analysis of citizens on the internet and the mass public. Social Science Computer Review, 21, 26–42.

43.

Wilhelm

A. G.

(2000). Democracy in the digital age. New York, NY: Routledge.

44.

Wilson

K. R.

Wallin

J. S.

Reiser

(2003). Social stratification and the digital divide. Social Science Computer Review, 21, 133–143.

45.

Zillien

Hargittai

(2009). Digital distinction: Status-specific types of internet usage. Social Science Quarterly, 90, 274–291.