Second Screening for News and Digital Divides

Abstract

Second screening is a relatively new set of media practices that arguably empower audiences to shape public narratives alongside news organizations and political elites. But in developing countries such as Colombia, it is important to examine who participates in this process, as substantial inequalities in both access to and use of information and communication technologies (ICTs) persist. This study examines how socioeconomic status (SES) relates to the adoption of second screening practices in Colombia, a country in which the technological access and literacy necessary to engage in these practices are becoming widespread but are not yet ubiquitous. Based on a random sample of face-to-face interviews, results show evidence of persistent digital divides in Colombia in terms of ICT access, ICT use, and second screening for news. Additionally, results indicate that the relationship between SES and second screening for news is indirect, mediated through technological access and public affairs engagement.

Keywords

second screening dual screening news use social media social networking sites digital divides digital inequalities Colombia Latin America

Second screening for news refers to a relatively new set of media practices in which individuals use a second device to further engage with television or online television news content (Gil de Zúñiga, García-Perdomo, & McGregor, 2015), and these practices arguably give audiences greater ability to shape public narratives alongside journalistic organizations and political elites (Anstead & O’Loughlin, 2011). But while these practices are rapidly spreading on a global scale, it is important to examine whether their adoption reflects digital divides based on access to or use of information and communication technologies (ICTs), especially in developing countries such as Colombia.

Access to ICTs has become ubiquitous in the developed world. In fact, countries such as (South) Korea and Denmark have achieved almost total penetration with more than 95% access to high-speed broadband. To some extent, the ubiquity of ICTs in the developed world has abated concerns about digital divides based on access to technology and has given rise to “second-level” concerns over divides in civic skills and specific uses of ICTs (e.g., DiMaggio, Hargittai, Celeste, & Shafer, 2004).

But in the developing world, inequalities in both ICT access and use persist (e.g., Hilbert, 2016). These gaps are particularly relevant when it comes to the adoption of emerging media practices that require access to multiple devices and civic skills to understand and participate in the creation of public narratives about ongoing social and political issues. Therefore, while second screening may empower audiences to participate in public conversations, important questions remain about who gets to participate in those conversations in developing countries such as Colombia.

While prior research has examined second screening practices in the context of the developing world and, more specifically, Latin America (e.g., McGregor, Mourao, Neto, Straubhaar, & Angeluei, 2017), it has not specifically examined second screening for news, as it relates to digital divides. The current study fills that gap in the literature by testing how socioeconomic status (SES) is related to the adoption of second screening for news in Colombia, a country that has seen a rapid increase in the size of the middle class in the last decade (World Bank, 2015). Based on a random sample of face-to-face interviews, the paper tests the expected relationships between SES and second screening while accounting for the mediating roles of technological access and use of the internet and social media for public affairs information.

Second Screening

Second screening is an emerging set of technologically enabled communication behaviors in which “individuals watching television use an additional electronic device, or ‘screen’ to access the internet or social networking sites to obtain more information about the program or event they are watching or to discuss it in real time” (Gil de Zúñiga et al., 2015, p. 795). According to this definition, second screening takes the form of two related but distinct behaviors, information seeking and discussion, and these behaviors may occur in tandem as individuals watch-mediated events in real time (Wohn & Na, 2011).

With a specific focus on second screening practices for news, one body of research centers on the idea that these practices help to reshape the relationships among news audiences, news media, and social institutions (e.g., Gillespie & O’Loughlin, 2015; Selva, 2015). Second screening for news affords users the ability to publicly criticize mass media and political elites (D’heer & Verdegem, 2015; Freelon & Karpf, 2015; Pond, 2016). To a certain extent, news organizations and political elites respond to this criticism by reshaping their content or talking points (Anstead & O’Loughlin, 2011, 2015; Chadwick, Dennis, & Smith, 2016), even though second screeners represent a vocal minority (Bentivegna & Marchetti, 2015) who generally respond to mass media content (Kim, Lee, & Park, 2016; Vergeer & Franses, 2016; Wells et al., 2016). Thus, scholars have described second screening for news as a complex interplay between media industries and audiences (Barry & Doherty, 2017; Moe, Poell, & van Dijck, 2016), and this interplay implies that second screening behaviors may be especially meaningful in terms of the influence those behaviors have on media and political organizations or institutions.

One prominent example of this complex interplay between news organizations and social media users is the contemporary televised political debate. It is increasingly common for people to scan Twitter for additional information or “live Tweet” during a political debate (Bentivegna & Marchetti, 2015; Freelon & Karpf, 2015; Pond, 2016; Vergeer & Franses, 2016; Wells et al., 2016). These interrelated practices enable debate viewers the ability to seek additional information about the debate and discuss it with other Twitter users in real time. They also allow users to offer commentary about the debate as it occurs and news coverage about the debate as it is produced. Ostensibly, both news organizations and political strategists have the ability to react to this feedback in real time, reshaping content or talking points accordingly.

Another strand of literature suggests that second screening for news generally promotes political engagement. It is positively related to online political participation (Fitzgerald & Clarke, 2012; Gil de Zúñiga et al., 2015; Vaccari, Chadwick, & O’Loughlin, 2015), it provides alternative spaces for political discussion (D’heer & Courtois, 2016; D’heer & Verdegem, 2014; Iannelli & Giglietto, 2015), and it promotes online content creation (Freelon & Karpf, 2015; Giglietto & Selva, 2014). Additionally, second screening may enhance engagement with and enjoyment of news content (Auverset & Billings, 2016; Choi & Jung, 2016; Horning, 2017), and it may be influential in terms of helping people to form opinions about the event they are watching (Barnidge, Gil de Zúñiga, & Diehl, 2017). Thus, not only can second screening practices influence organizations and institutions, but they can also influence the individuals who engage in them. Second screening for news therefore provides new ways for people to engage with politics and public affairs, and it can also be influential in terms of shaping perception, opinion, and behavior.

Digital Divides in Developed Economies

ICT Access

Emerging from center-left social inclusion policy in the 1990s, the term “digital divides” became popular as a way to describe the gap between “information haves” and “information have-nots” within individual countries (Selwyn, 2004). One of the major themes of this public conversation is that the adoption of ICTs is regarded as a bellwether for the democratization of information, the empowerment of marginalized social groups, and the ability of citizens to bypass elite organizational structures to achieve meaningful political outcomes (e.g., Norris, 2001; Selwyn, 2004). Access to ICTs, these authors argue, enables a greater number of individuals to leverage communication networks for political, economic, or educational progress, forging a more inclusive society in the process.

Early research largely supported the idea that well-off individuals had more access to ICTs than less well-off individuals. In fact, the relationship between SES and ICT access became a relatively common finding across early digital divide studies (e.g., Ali, 1999; Gaziano, 1997; Hoffman & Novak, 1999; Katz & Apsden, 1997; Markus, 1987; Rice & Katz, 2003). Thus, these scholars concluded that ICT adoption is socially structured and that long-standing inequalities based on history, class, education, and economic opportunities are reproduced in the form of access divides whenever new technologies diffuse through a society. The result is a form of social exclusion, where older and poorer groups are most affected. Because the internet plays an increasingly important role in organizing the global economy, the lack of access can hinder a range of activities for low-SES groups, from job searches and health-care access to political engagement and learning opportunities (Ragnedda, 2017, p. 84).

But more recently, access to basic telecommunications services has become widespread in developed economies. To be sure, social inequalities in ICT access remain in these countries, especially between the young and old (Enoch & Soker, 2006; Harris, Staker, & Pollock, 2017; Livingstone & Helsper, 2007, 2010; Loges & Jung, 2001; North, Snyder, & Bulfin, 2008) and between suburban and other geographic areas (Van Dijk, 2005; Warren, 2007; Warschauer, 2004; Wei & Hindman, 2011). However, concerns about unequal access to ICTs have abated, to some extent, now that internet and mobile access has become almost ubiquitous in these contexts (Hilbert, 2014). According to webworldwide.io, 85.7% of people in the United States and 90.2% of Canadians now have high-speed broadband internet access (4 Mbps+), and these figures are even higher in the most “wired” countries in the world, including Denmark (95.7%) and (South) Korea (97%).

ICT Use

In response to diminishing inequalities in terms of access in developed economies, scholars began to investigate what they termed “second-level” digital divides that emphasized the ICT use rather than access (e.g., DiMaggio et al., 2004; Hargittai & Hinnant, 2008; Hilbert, 2014; López-Sintas, Filimon, & García-Álvarez, 2012; Van Deursen & Van Dijk, 2014; Van Dijk & Hacker, 2003; Wei, 2012; Yang & Grabe, 2011; Zillien & Hargittai, 2009). These scholars argue that while the access gap may have closed, high-SES individuals are more likely to possess civic skills and education, and that these skills-based inequalities manifest in differential patterns of ICT use for economic, community, and political engagement. For example, research shows that there is a relatively strong relationship between an individual’s level of education and their likelihood to visit capital-enhancing websites such as news, politics, government, and consumer web pages (Hargittai & Hinnant, 2008; see also Hargittai, 2007). Additionally, higher levels of education are associated with ICT use for work and educational purposes such as word processing and looking for classes online (Van Dijk & Hacker, 2003; Van Deursen & Van Dijk, 2014). By contrast, lower-SES individuals are more likely to use ICTs for entertainment purposes (Hargittai & Hinnant, 2008; Van Deursen & Van Dijk, 2014).

A gap in the application and use preferences along SES lines is potentially problematic because news consumption and public affairs communications play important roles in helping people identify and act upon opportunities to participate in politics and community life, such as joining civic organizations, volunteering for a political campaign, or voting (Norris, 2000; Putnam, 2000). In addition, news consumption online is often regarded as an antecedent to other prodemocratic communicative behaviors, such as expressing political opinions, participating in online political forums, mobilizing others for political action, and developing social capital through network building and maintenance (Gil de Zúñiga, Jung, & Valenzuela, 2012; Rojas & Puig-i-Abril, 2009; Shah, Cho, Eveland, & Kwak, 2005; Velasquez & Rojas, 2017).

Digital Divides in Developing Economies and the Case of Colombia

In comparison to developed economies, gaps in terms of both ICT access and ICT use are wider in developing economies (Hilbert, 2016; Norris, 2001; Notten, Peter, Kraaykamp, & Valkenburg, 2009; Rojas & Puig-i-Abril, 2009; Tirado-Morueta, Mendoza-Zambrano, Aguaded-Gómez, & Marín-Gutiérrez, 2017). For example, Notten, Peter, Kraaykamp, and Valkenburg (2009) found that gaps in access to ICTs are larger in low-income countries than in high-income countries. Additionally, in a study of 172 countries, Hilbert (2016) found evidence of greater access and use divides in low-income countries than in high-income countries.

Additional evidence comes specifically from the Latin American context. For example, in a study of Colombia, Rojas and Puig-i-Abril (2009) found that both income and education are significant predictors of internet access and cellular phone ownership and that these variables are also significantly related to news use and blog use. Meanwhile, Tirado-Morueta, Mendoza-Zambrano, Aguaded-Gómez, and Marín-Gutiérrez (2017) found similar evidence of access and use divides in Ecuador. Taken together, these results suggest that while gaps in use may be more persistent than gaps in access in developed economies, both levels of the digital divide persist in developing countries, particularly in Latin America.

That said, internet and mobile adoption are on the rise in the developing world, generally, and in Latin America, specifically (International Telecommunication Union, 2016; Poushter, 2016). Latin America’s internet penetration growth rate between 2012 and 2017 is one of the highest in the world (internetworldstats.com), and the historical gap between Latin America and the developed world has shrunk during that time span. Today, Central and South America claim approximately 10% of the world’s internet users, 10% of the world’s mobile phone subscriptions, and 13% of the world’s social media accounts (wearesocial.com). Colombia exemplifies these regional trends: From 2012 to 2015, the percentage of the population with a social media account increased from approximately 43% to 64% (Allin1Social, 2015; Colombia Reports, 2015).

However, despite this relatively rapid diffusion of technologies over the past five years, Colombia still lags behind the most connected countries in the world—approximately 70% of the Colombian population is online, as compared to approximately 86% of the U.S. population and 97% of the (South) Korean population. These two contextual factors—rapid growth on the one hand and a comparative lag on the other—make Colombia an ideal case to explore the digital divide as it relates to second screening, because the technological access and literacy necessary to engage in these practices are fast becoming widespread, but they are not yet ubiquitous.

Hypotheses

Several testable predictions can be derived based on the above literature, which establishes (a) the logic of digital divides in developing economies and (b) the continued importance of examining digital divides in developing economies. Based on these theoretical considerations, a process model of adopting second screening behaviors is developed in this section. First, early research on digital divides indicates that the ICT adoption is socially structured and that long-standing inequalities based on history, class, education, and economic opportunities will be reproduced in patterns of ICT access (e.g., Katz & Apsden, 1997; Markus, 1987). Individuals and groups with more economic and social resources will be more easily able to afford access to ICTs as they appear on the market. Moreover, empirical evidence suggests that these gaps in access persist in the developing world and in Latin America, specifically (Rojas & Puig-i-Abril, 2009; Tirado-Morueta et al., 2017). Therefore, we predict the following in the context of Colombia:

Hypothesis 1: Individuals with high SES will have more access to ICTs than individuals with low SES.

Additionally, scholarship on gaps in ICT use suggests that high-SES individuals will possess the civic skills and education necessary to use ICTs to engage with news and public affairs (e.g., DiMaggio et al., 2004; Hargittai & Hinnant, 2008; Hilbert, 2014; Van Deursen & Van Dijk, 2014). The logic behind this hypothesis also suggests a direct effect of SES, such that it influences not only the affordability of ICTs but also the ways in which people use them. This use gap is also highly relevant in the context of Latin America (Rojas & Puig-i-Abril, 2009; Tirado-Morueta et al., 2017). Therefore, we predict that in Colombia, this logic will manifest in three different forms of public affairs engagement via ICTs: internet use for public affairs communication, social media use for public affairs communication, and second screening for news.

Prior research has distinguished between engagement with public affairs content on the internet and social media. The specific affordances of social media including the articulation of social networks (Barnidge, 2017; boyd & Ellison, 2007), the ease of social interaction (Gil de Zuniga, Molyneux, & Zheng, 2014), incidental exposure to news content (Brundidge, 2010), and the development of social capital differ (Ellison, Steinfield, & Lampe, 2007; Gil de Zúñiga, Barnidge, & Scherman, 2017). Additionally, scholarship on second screening has argued that these behaviors, while they largely take place on social media, are unique in that they require the use of multiple communication devices simultaneously (Gil de Zúñiga et al., 2015). Thus, we also distinguish between these various forms of engaging with public affairs content via ICTs, and we propose the following hypotheses in the context of Colombia:

Hypothesis 2: Individuals with high SES will use the internet for public affairs communications more than individuals with low SES.

Hypothesis 3: Individuals with high SES will use social media for public affairs communications more than individuals with low SES.

Hypothesis 4: Individuals with high SES will second screen for news more than individuals with low SES.

According to the logic of the use divide, the effects of SES on public affairs use are largely indirect and mediated through civic skills and resources. These civic skills are developed, in part, by habitual use of ICTs from youth (Enoch & Soker, 2006; Harris et al., 2017; Livingstone & Helsper, 2007, 2010; Loges & Jung, 2001; North et al., 2008). Therefore, it stands to reason that higher levels of access to ICTs will also manifest in increased public affairs use because having more access overtime helps individuals develop the civic skills necessary to engage with public affairs communications via ICTs. Therefore, we predict an indirect relationship between SES and the three forms of public affairs engagement, mediated through ICT access.

Hypothesis 5: The relationship between SES and second screening will be mediated by access to ICTs.

Hypothesis 6: The relationship between SES and second screening will be mediated by internet use for public affairs communications.

Hypothesis 7: The relationship between SES and second screening will be mediated by social media use for public affairs communications.

In addition to SES, age itself is an important factor to consider. While no formal hypotheses regarding age are presented because it is not a primary focus of the study, prior research shows that younger individuals are more likely to adopt ICTs (Enoch & Soker, 2006; Harris et al., 2017; Livingstone & Helsper, 2007, 2010; Loges & Jung, 2001; North et al., 2008) and to second screen for news (e.g., Barnidge et al., 2017). Therefore, ICT access and use may cut across socioeconomic strata among younger individuals who are more likely to adopt and use new media technologies regardless of their social or economic situation. Thus, age may act as an important counterbalance to SES.

Based on these predictions, we outline a process model for the adoption of second screening practices, which is illustrated in Figure 1. Two external background factors (age and SES) predict technological access, which in turn predicts engagement with public affairs on social media and other online venues. Finally, these three intervening variables (technological access, internet use for public affairs, and social media use for public affairs) are the proximal predictors of second screening and act as mediators for the indirect relationship between SES and second screening.

Method

Sample and Data

This study relies on survey data collected between June 28 and July 29, 2014, in 10 cities in Colombia as part of a biennial study of communication and political attitudes. The sample was designed to represent Colombia’s adult urban population—76% of Colombia’s 48.2 million inhabitants live in urban areas (Departamento Administrativo Nacional de Estadística (DANE), 2014 [National Administrative Department of Statistics]).

Respondents were selected using a multistage stratified random sampling procedure that selected households randomly based on city size and census data. Once the number of households was allocated for a given city, a number of city blocks were selected randomly according to housing district and stratum. Then, individual households were randomly selected within each block. Finally, the study used the “adult in the household who most recently celebrated a birthday” technique to identify an individual respondent at random. Up to three visits to each household were made (if needed) to increase participation in the survey. A local professional polling firm, Deproyectos Limitada, collected the data and 1,102 face-to-face completed responses were obtained for a response rate of 55.5%. Removing missing cases listwise yielded 748 cases for analysis.

Endogenous Variables

Second screening

Based on previous research (Anstead & O’Loughlin, 2011; Choi & Jung, 2016), three six-point survey items (0 = never, 5 = frequently) were used to measure second screening. Questions asked respondents how often they (1) check what other people are saying online about the event that (they) are watching, (2) publish (their) own comments about something that (they) are watching, and (3) exchange text messages with people who are watching the same event but in a different place. These three items were averaged to create the final variable (Cronbach’s α = .91, M = 1.02, SD = 1.42). The variable was then logged to normalize the distribution (M = .50, SD = .62, min. = 0.00, max. = 1.79).

Social media use for public affairs

Engagement with public affairs content on social media has been shown to be an important antecedent of second screening (Gil de Zúñiga et al., 2015). Eight six-point survey items (0 = never, 5 = frequently) were used to measure social media use for public affairs. Items asked respondents how frequently they (1) express (their) views about current issues on (a) Facebook and (b) Twitter, (2) share news stories with (their) contacts on (a) Facebook and (b) Twitter, (3) read news articles posted by (their) contacts on (a) Facebook and (b) Twitter, and (4) read the opinions of (their) contacts about political issues on (a) Facebook and (b) Twitter. These eight items were averaged (Cronbach’s α = .91, M = .81, SD = 1.09).

Internet use for public affairs

Likewise, engagement with public affairs content in other online spaces is an important antecedent of second screening (Gil de Zúñiga et al., 2015). The construct was measured with seven six-point items (0 = never, 5 = frequently) asking respondents how frequently they use the internet to: (1) comment on current events and politics via e-mail, (2) discuss the news and opinion columns that appear in online media, (3) participate in online discussion forums, (4) express (their) opinion on social and political issues, (5) search for news information online, (6) seek government information, and (7) visit political blogs. These seven items were averaged (Cronbach’s α = .89, M = .70, SD = .97).

Technological access

According to prior literature, basic access to communication technology is necessary in order for people engage in emerging media practices, and more advanced access makes it more likely that people will engage in such practices (Livingstone & Helsper, 2010). Technological access was measured with four yes/no survey items that asked respondents whether (1) they have used the internet in the last 12 months, (2) whether their cell phone has data capabilities (respondents with no cell phones received a score of 0), (3) whether they are a member of online social networks such as Facebook, and (4) whether they use Twitter. Scores on these four items were added together (Cronbach’s α = .76, M = 1.52, SD = 1.41).

Exogenous Variables

SES

According to research on digital divides, SES could be an important predictor of second screening (DiMaggio et al., 2004; Hargittai & Hinnant, 2008). Three survey items were used to measure SES. Education was measured on an eight-point scale (0 = none, 7 = postgraduate degree), annual household income was measured on a nine-point scale (0 = less than CP500,000, 8 = more than CP10,000,000), and household energy stratum—which is a proxy measure of the respondent’s household energy consumption (e.g., of a similar approach to proxy measures of SES; see, e.g., Oakes & Rossi, 2003; Somi et al., 2008)—was measured on a six-point scale (min. = 0, max. = 5). These three items were standardized and then averaged (Cronbach’s α = .74. M = 3.09, SD = 1.11). A binary variable was created using median split (median = 3), high SES = 56%, SE = .50.

Age

Prior research shows that younger people are more likely to second screen (Barnidge et al., 2017), and therefore age was measured with a single survey item (M = 42.85, SD = 16.81).

Control Variables

Gender

Gender was measured with a single survey item (female = 55%, SE = .50).

Race

Race was also measured with a single survey item (2 = White, 1 = mestizo/mixed race, 0 = Black Caribbean/Indigenous/other). Reflecting the national population of Colombia, the average respondent is mestizo/mixed race (M = 1.19, SD = .57).

Political interest

Three survey items were used to measure political interest (0 = not at all, 5 = a lot). Items asked respondents how much interest they have in (1) local or regional politics, (2) national politics, (3) and international politics. These items were averaged (Cronbach’s α = .92, M = 1.91, SD = 1.47).

Political efficacy

Three survey items were also used to measure political efficacy (0 = strongly disagree, 5 = strongly agree). Items asked respondents how much they agree or disagree with the following statements: (1) people like me can influence what the local government does, (2) I believe that the national government cares about what people like me think, and (3) city government responds to the initiatives of individuals. These items were averaged (Cronbach’s α = .80, M = 1.74, SD = 1.29).

Ideological extremity

Respondents were asked to place their political ideologies on a left-right scale (0 = liberal, 5 = neutral, 10 = conservative). The scale was folded, so that high scores indicate extreme ideologies on either side (M = 1.50, SD = 1.68).

News use

News use was measured with five six-point survey items (0 = never, 5 = frequency). Items asked respondents how often they read, watch, or listen to (1) national daily newspapers, (2) national news magazines (e.g., Semana), (3) national television news, (4) online news, and (5) social media news. These items were averaged (Cronbach’s α = .68, M = 1.83, SD = 1.02).

Political talk network size

Respondents were asked how many family members, close friends, neighbors, and acquaintances with whom they talk politics. These items were added together (Cronbach’s α = .75, M = 10.17, SD = 15.58).

Political talk frequency

Respondents were also asked how often (0 = never, 5 = often) they talk politics with each type of person named above (Cronbach’s α = .86, M = 1.65, SD = 1.37).

Political talk diversity

Finally, respondents were asked how often they talk politics with people who (1) have very different ideas from (their) own, (b) are from a different social stratum than (them), and (c) are a very different age then (them). These items were averaged (Cronbach’s α = .87, M = 2.09, SD = 1.59).

Analysis

The analysis was conducted in three phases. First, independent samples t tests were used to assess the mean differences between high- and low-SES groups in technological access, internet use for public affairs communication, social media use for public affairs communication, and second screening. Then, the endogenous and exogenous variables were residualized on the control variables, and path analysis was conducted in the structural equation modeling framework. Finally, mediation tests were conducted to assess the indirect relationships between SES and second screening.

Results

The results show evidence of digital divides in Colombia in terms of both access to and use of ICTs. Significant differences between high- and low-SES groups were detected for all four endogenous variables in expected directions. Independent samples t tests were conducted using the binary SES variable to assess mean differences, and results show that the high-SES group was more likely than the low-SES group to have access to ICTs, to use the internet for public affairs communication, to use social media for public affairs communication, and to second screen. These results, which generally support Hypothesis 1 to Hypothesis 4, are summarized in Table 1 and visualized in Figure 2.

Table 1.

Mean Comparisons and Test Statistics for ICT Access and Use Variables.

Variable	High SES	Low-SES	t Value (df)
Technological access	1.98	.97	−12.56*** (1,056.20)
Internet use for public affairs	1.01	.32	−12.65*** (990.34)
Social media use for public affairs	1.12	.46	−10.47*** (1,006.00)
Second screening for news	0.63	.31	−8.20*** (857.57)

Note. Cell entries are group means and test statistics from independent samples t tests. ICT = information and communication technologies. SES = socioeconomic status.

***p < .001.

Results also indicate that SES is indirectly related to second screening. Path analysis was conducted using the correlation matrix in Table 2, and standardized results are summarized in Figure 3. The model fits the data well, χ²(1) = 2.37 = 8, confirmatory fit index (CFI) = .99, goodness-of-fit index (GFI) = .99, root mean square error of approximation (RMSEA) = .04, p = .43, standard root mean square residual (SRMR) = .01, and its explanatory value is relatively high (R² second screening = .30). While SES is not directly related to second screening (γ = .01, p = .865), it is positively related to all intervening variables including technological access (γ = .22, p < .001), engagement with public affairs on the internet (γ = .08, p = .012), and engagement with public affairs on social media (γ = −.08, p = .004). All of these intervening variables are directly related to second screening (for technological access, β = .20, p < .001; for internet for public affairs, β = .21, p < .001; for social media for public affairs, β = .15, p = .001). Technological access is also positively related to both engagement variables (β = .42, p < .001 for internet and β = .64, p < .001 for social media).

Table 2.

Correlation Matrix Used in Path Analysis.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
(1) Second screening for news	1.00	—	—	—	—	—
(2) Social media for public affairs	.45	1.00	—	—	—	—
(3) Internet for public affairs	.41	.53	1.00	—	—	—
(4) Technological access	.46	.66	.44	1.00	—	—
(5) Socioeconomic status	.12	.09	.19	.26	1.00	—
(6) Age	−.37	−.40	−.25	−.49	−.07	1.00

Note. N = 748. Cell entries are Pearson’s correlation coefficients. Second screening is logged. All variables have been residualized on the controls: gender, race, political interest, political efficacy, ideological extremity, news use, political talk network size, political talk frequency, and political talk diversity.

Finally, results suggest that the indirect relationship between SES and second screening is mediated through technological access and engagement with public affairs content online and via social media. Mediation tests were conducted to formally test the indirect relationships between SES and second screening. Results, which support Hypothesis 5–Hypothesis 7, are reported in Table 3. Bootstrapped indirect effects are estimated for technological access at .06 (SE = .01, p < .001), for internet use for public affairs at .04 (SE = .01, p < .001), and for social media use for public affairs at .02 (SE = .01, p < .001). Combined with a direct effect of .01 (SE = .01), the model therefore estimates a total effect of SES at .11 (SE = .02). Thus, respondents score 4% higher on the second screening measure (range = 2.89) for every 1 standard deviation increase in SES, and this change is facilitated by technological access and engagement with public affairs content.

Table 3.

Indirect Effects of Socioeconomic Status on Second Screening for News.

Variable	Sobel	Aroian	Goodman	Bootstrap B	Bootstrap SE
Technological access	6.41***	6.39***	6.42***	.06***	.01
Internet use for public affairs	4.84***	4.83***	4.86***	.04***	.01
Social media use for public affairs	2.44*	2.43*	2.45*	.02**	.01

Note. N = 748. Direct effect = .01, SE = .01, p = .747. Total effect = .13, SE = .02, p < .001. 1,000 bootstrap iterations. The dependent variable is logged. All variables have been residualized on the controls: gender, race, political interest, political efficacy, ideological extremity, news use, political talk network size, political talk frequency, and political talk diversity.

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

Discussion

This study finds evidence of substantial digital divides in Colombia, both in terms of ICT access and use. Specifically, it finds gaps in the adoption of second screening behaviors, and SES is a significant factor in creating these gaps via its indirect relationship with ICT use for public affairs communications. These conclusions relate to several important discussion points.

First, these findings highlight the role of digital divides in reinforcing social inequality in Colombia, which is in the process of democratization in the wake of more than 40 years of political and drug-related conflict (Rojas, Barnidge, & Puig-i-Abril, 2016). Despite the fact that a vibrant sphere of political communication has emerged online in which Colombian political and social leaders seek to engage with citizens in order to reinvigorate democratic institutions, Colombia has recently witnessed various scandals involving corruption in government contracts, human rights violations, and illegal monitoring of opposition parties (Rodriguez & Seligson, 2008). In as much as second screening for news gives ordinary citizens the ability to pushback against this corruption by reshaping public narratives about important social and political issues (e.g., Anstead & O’Loughlin, 2011), the ability to do so may be limited to high-SES individuals. Thus, the problem of including those from lower-SES backgrounds in these processes of democratization is a fundamental challenge to Colombian democracy. And while this study has focused on digital inequalities in urban areas, where most Colombians live, the country’s Indigenous and Black populations may pose the biggest challenges to political inclusion, as these groups generally live in underdeveloped, coastal areas with low levels of access to basic education and communication infrastructure (Telles & Paschel, 2014). Therefore, while this study highlights social inequalities embedded in Colombia’s mestizo population, further research is needed on these Black and Indigenous groups to fully investigate digital inequalities in that country.

Second, the results of this study may not be generalizable to other national contexts. However, they are perhaps more representative of developing economies than they are of developed economies. These results also speak to the general state of democratization in Latin America, more broadly. Democracy in Latin America has a checkered past, and at times, it has been undermined by the residual legacy of colonialism, foreign intervention by the United States, and corruption at the highest levels of social and political institutions (e.g., Karl, 1990). Second screening for news may provide citizens in Latin America with a new tool to influence these institutions and to counteract political corruption in order to reshape their societies (Anstead & O’Loughlin, 2011). However, it is important to recognize that not all citizens of Latin America are currently participating in these processes, and this recognition should be a primary consideration for public policy both at the national and international levels. Specifically, policies that promote the expansion of telecommunications infrastructure and civic education—processes that are already underway—may enhance democracy in Latin America.

Third, and more generally, these findings highlight the continuing importance of studying digital divides in the developing world. While ubiquitous access to ICTs may have obviated the need to assess digital access divides in more developed countries (e.g., Poushter, 2016), these divides persist in Colombia and other developing nations despite recent trends toward growing internet access (e.g., Hilbert, 2016), particularly in Latin America (wearesocial.com). In fact, as access to online technologies reaches different social groups at different rates, these divides may exacerbate existing social cleavages, hindering the inclusivity of emerging democracies in the developing world (Norris, 2001). Continuing to investigate these trends, as the adoption of technology becomes more ubiquitous is of primary importance to those interested in the health of democracy in Latin America and elsewhere in the developing world.

Finally, the findings also imply that emerging media practices such as second screening may exacerbate information inequalities, especially in developing world contexts. Second screening for news may be a “weapon of the strong” (Barnidge et al., 2017). That is, the individuals who engage in second screening practices are often the individuals who are already in the best position to take advantage of their democratically beneficial outcomes including discursive engagement (e.g., D’heer & Verdegem, 2015), political content creation (e.g., Giglietto & Selva, 2014), social and community empowerment (e.g., McGregor et al., 2017), and political participation (e.g., Vaccari et al., 2015).

Several important limitations must be noted. First, the design of this research is cross-sectional, and therefore it cannot be used to make causal inferences. Although the process model developed herein is theoretically and empirically grounded in prior research, future research should employ an overtime design to be able to make causal claims about second screening and digital divides. A second set of limitations relates to the measurement strategy. Although all measurement scales reach acceptable levels of reliability, measurement of online behaviors and technological access are based on metaconcepts that combine several more specific concepts. We made this decision intentionally, trading a degree of conceptual precision for conceptual parsimony. However, future studies may prioritize precision over parsimony and opt to disaggregate some of the measures constructed in this study. Another limitation relates to certain findings that run counter to the literature. Specifically, the negative relationship between SES and social media use for public affairs goes against the study’s (informal) expectations based on prior research on digital divides. Further research is needed to clarify this relationship in multiple national contexts. A final limitation stems from the analysis. While the goal of this study was to examine a specific process of technological adoption and engagement in emerging online behaviors, there are certainly more variables (many of which were controlled) that could play an important role in predicting who second screens and who does not, and future research may opt to focus on examining these variables.

Despite these limitations, this study has provided relatively strong evidence of digital divides in Colombia both in terms of ICT access and use. Furthermore, SES has a direct influence on second screening for news, mediated through technological access and public affairs engagement. Thus, substantial socioeconomic divides in the adoption of second screening behaviors have developed in Colombia, potentially limiting the democratic impact of these emerging media practices in that context.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: Funding for this research was provided by the University of Wisconsin-Madison Graduate School Research Committee and also by Departamento Administrativo de Ciencias, Technología, e Innovación de Colombia (COLCIENCIAS). [Translation: Colombia Administrative Department of Science, Technology and Innovation (COLCIENCIAS)].

References

Ali

A. J

. (1999). Digital divide: A challenge that must be faced. Journal of Competitiveness Studies, 7, i–ii.

Allin1Social. (2015). Facebook statistics worldwide. Retrieved from http://www.allin1social.com/facebook-statistics/countries/

Anstead

O’Loughlin

(2011). The emerging viewertariat and BBC question time: Television debate and real-time commenting online. The International Journal of Press/Politics, 16, 440–462. doi:10.1177/1940161211415519

Anstead

O’Loughlin

(2015). Social media analysis and public opinion: The 2010 UK Election. Journal of Computer-Mediated Communication, 20, 204–220. doi:10.1111/jcc4.12102

Auverset

L. A.

Billings

A. C.

(2016). Relationships between social TV and enjoyment: A content analysis of The Walking Dead’s story sync experience. Social Media + Society, 2, 1–12. doi:10.1177/2056305116662170

Barnidge

(2017). Exposure to political disagreement in social media versus face-to-face and anonymous online settings. Political Communication, 34, 302–321. doi:10.1080/10584609.2016.1235639

Barnidge

Gil de Zúñiga

Diehl

(2017). Second screening and political persuasion on social media. Journal of Broadcasting & Electronic Media, 61, 309–331. doi:10.1080/08838151.2017.1309416

Barry

Doherty

(2017). What we talk about when we talk about interactivity: Empowerment in public discourse. New Media & Society, 19, 1052–1071. doi:10.1177/1461444815625944

Bentivegna

Marchetti

(2015). Live tweeting a political debate: The case of the ‘Italia bene comune.’ European Journal of Communication, 30, 631–647. doi:10.1177/0267323115595526

10.

Boyd

Ellison

N. B.

(2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13, 210–230. doi:10.1111/j.1083-6101.2007.00393.x

11.

Brundidge

(2010). Encountering “difference” in the contemporary public sphere: The contribution of the Internet to the heterogeneity of political discussion networks. Journal of Communication, 60, 680–700. doi:10.1111/j.1460-2466.2010.01509.x

12.

Chadwick

Dennis

Smith

A. P.

(2016). Politics in the age of hybrid media: Power, systems and media logics. In Bruns

Enli

Skogerbo

Larsson

A. O.

Christensen

(Eds.), The Routledge companion to social media and politics (pp. 7–22). New York, NY: Routledge.

13.

Choi

Jung

(2016). The effects of second-screen viewing and the goal congruency of supplementary content on user perceptions. Computers in Human Behavior, 64, 347–354. doi:10.1016/j/chb/2016/06/048

14.

Colombia Reports. (2015). Social media dominates internet usage in Colombia. Retrieved from https://colombiareports.com/social-media-dominates-internet-usage-in-colombia/

15.

DANE. (2014). Colombia’s National Department of Statistics—DANE. Retrieved from http://www.dane.gov.co

16.

D’heer

Courtois

(2016). The changing dynamics of television consumption in the multimedia living room. Convergence, 22, 3–17. doi:10.1177/1354856514543451

17.

D’heer

Verdegem

(2014). Conversations about the elections on Twitter: Towards a structural understanding of Twitter’s relation with the political and media field. European Journal of Communication, 29, 720–734. doi:10.1177/0267323114544866

18.

D’heer

Verdegem

(2015). What social media data mean for audience studies: A multidimensional investigation of Twitter use during a current affairs TV programme. Information, Communication, & Society, 18, 221–234. doi:10.1080/1369118X.-2014.952318

19.

DiMaggio

Hargittai

Celeste

Shafer

(2004). From unequal access to differentiated use: A literature review and agenda for research on digital inequality. In Neckerman

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

20.

Ellison

N. B.

Steinfield

Lampe

(2007). The benefits of Facebook “friends:” Social capital and college students’ use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143–1168. doi:10.1111/j.1083-6101.2007.00367.x

21.

Enoch

Soker

(2006). Age, gender, ethnicity and the digital divide: University students’ use of web-based instruction. Open Learning, 21, 99–110.

22.

Fitzgerald

Clarke

(2012, 8). How multi-screen consumers are changing media dynamics. Coalition for Innovative Media Measurement. Retrieved from http://www.comscore.com/dut/Insights/Presentations-and-Whitepapers/2012/HowMulti-Screen-Consumers-Are-Changing-Media-Dynamics

23.

Freelon

Karpf

(2015). Of big birds and bayonets: Hybrid Twitter interactivity in the 2012 Presidential debates. Information, Communication, & Society, 18, 390–406. doi:10.1080/1369118X.2014.952659

24.

Gaziano

(1997). Forecast 2000: Widening knowledge gaps. Journalism & Mass Communication Quarterly, 74, 237–264. doi:10.1177/107769909707400202

25.

Gil de Zúñiga

Barnidge

Scherman

(2017). Social media social capital, offline social capital, and citizenship: Exploring asymmetrical social capital effects. Political Communication, 34, 44–68. doi:10.1080/10584609.2016.1227000

26.

Gil de Zúñiga

García-Perdomo

McGregor

S. C.

(2015). What is second screening? Exploring motivations of second screen use and its effect on online political participation. Journal of Communication, 65, 793–815. doi:10.1111/-jcom.12174

27.

Gil de Zúñiga

Jung

Valenzuela

(2012). Social media use for news and individuals’ social capital, civic engagement and political participation. Journal of Computer-Mediated Communication, 17, 319–336. doi:10.1111/j.1083-6101.2012.01574.x

28.

Gil de Zúñiga

Molyneux

Zheng

(2014). Social media, political expression, and political participation: Panel analysis of lagged and concurrent relationships. Journal of Communication, 64, 612–634. doi:10.1111/jcom.12103

29.

Giglietto

Selva

(2014). Second screen and participation: A content analysis on a full season dataset of tweets. Journal of Communication, 64, 260–277. doi:10.1111/-jcom.12085

30.

Gillespie

O’Loughlin

(2015). Editorial introduction: International news, social media and soft power: The London and Sochi Olympics as global media events. Participations: Journal of Audience & Reception Studies, 12, 388–412. Retrieved from http://www.participations.org/-Volume%2012/Issue%201/24.pdf

31.

Hargittai

(2007). Whose space? Differences among users and non-users of social network sites. Journal of Computer-Mediated Communication, 13, 276–297. doi:10.1111/j.1083-6101.2007.00396.x

32.

Hargittai

Hinnant

(2008). Digital inequality: Differences in young adults’ use of the internet. Communication Research, 35, 602–621. doi:10.1177/-0093650208321782

33.

Hoffman

D. L.

Novak

T. P.

(1999). Bridging the racial divide on the Internet. Science, 280, 390–391. doi:10.1126/science.280.5362.390

34.

Harris

Straker

Pollock

(2017). A socioeconomic related “digital divide” exists in how, not if, young people use computers. PLoS One, 12, e0175011. doi:1371/journal.pone.0175011

35.

Hilbert

(2014). Technological information inequality as an incessantly moving target: The redistribution of information and communication capacities between 1986 and 2010. Journal of the Association for Information Science and Technology, 65, 821–835. Retrieved from https://doi.org/10.1002/asi.23020

36.

Hilbert

(2016). The bad news is that the digital access divide is here to stay: Domestically installed bandwidths among 172 countries for 1986–2014. Telecommunications Policy, 40, 567–581. doi:10.1016/j.telpol.2016.01.006

37.

Horning

M. A.

(2017). Interacting with news: Exploring the effects of modality and perceived responsiveness and control on news source credibility and enjoyment among second screen viewers. Computers in Human Behavior, 73, 273–283. doi:10.1016/j.chb.2017.03.023

38.

Iannelli

Giglietto

(2015). Hybrid spaces of politics: The 2013 general elections in Italy, between talk shows and Twitter. Information, Communication, & Society, 18, 1006–1021. doi:10.1080/1369118X.2015.1006658

39.

International Telecommunication Union. (2016). Key findings: Measuring information society report 2016. Geneva, Switzerland: Author. Retrieved from http://www.itu.int/-en/ITU-D/Statistics/Documents/publications/misr2016/MISR2016-KeyFindings.pdf

40.

Karl

T. L.

(1990). Dilemmas of democratization in Latin America. Comparative Politics, 23, 1–21. Retrieved from http://www.jstor.org/stable/422302

41.

Katz

Aspden

(1997). Motives, hurdles, and dropouts. Communications of the ACM, 40, 97–102. doi:10.1145/248448.248464

42.

Kim

Lee

Y. O.

Park

H. W.

(2016). Delineating the complex use of a political podcast in South Korea by hybrid web indicators: The case of the Nakkomsu Twitter network. Technological Forecasting and Social Change, 110, 42–50. doi:10.1016/j.techfore.2015.11.012

43.

Livingstone

Helsper

(2007). Gradations in digital inclusion: Children, young people and the digital divide. New Media & Society, 9, 671–696. doi:10.1177/1461444807080335

44.

Livingstone

Helsper

(2010). Balancing opportunities and risks in teenagers’ use of the internet: The role of online skills and internet self-efficacy. New Media & Society, 12, 309–329. Retrieved from https://doi.org/10.1177/1461444809342697

45.

Loges

W. E.

Jung

J. Y.

(2001). Exploring the digital divide: Internet connectedness and age. Communication Research, 28, 536–562. doi:10.1177/009365001028004007

46.

López-Sintas

Filimon

García-Álvarez

M. E.

(2012). A social theory of internet uses based on consumption scale and linkage needs. Social Science Computer Review, 30, 108–129. doi:10.1177/0894439310390611

47.

Markus

M. L.

(1987). Toward a “critical mass” theory of interactive media: Universal access, interdependence and diffusion. Communication Research, 14, 491–511. doi:10.1177/009365087014005003

48.

McGregor

S. C.

Mourao

R. R.

Neto

Straubhaar

J. D.

Angeluei

(2017). Second screening as convergence in Brazil and the United States. Journal of Broadcasting & Electronic Media, 61, 163–181. doi:10.1080/08838151-.2016.1273924

49.

Moe

Poell

van Dijck

(2016). Rearticulating audience engagement: Social media and television. Television & New Media, 17, 99–107. doi:10.1177/-1527476415616194

50.

Norris

(2000). A virtuous circle: Political communications in postindustrial societies. Cambridge, England: Cambridge University Press.

51.

Norris

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

52.

North

Snyder

Bulfin

(2008). Digital tastes: Social class and young people’s technology use. Information, Communication & Society, 11, 895–911. doi:10.1080/13691180802109006

53.

Notten

Peter

Kraaykamp

Valkenburg

P. M.

(2009). Research note: Digital divide across borders cross-national study of adolescents’ use of digital technologies. European Sociological Review, 25, 551–560. doi:10.1093/-esr/jcn071

54.

Oakes

J. M.

Rossi

P. H.

(2003). The measurement of SES in health research: Current practice and steps toward a new approach. Social Science & Medicine, 56, 769–784. doi:10.1016/S0277-9536(02)00073-4

55.

Pond

(2016). Twitter time: A temporal analysis of Tweet streams during televised political debate. Television & New Media, 17, 142–158. doi:10.1177/-1527476415616190

56.

Poushter

(2016, 2 22). Smartphone ownership and internet usage continues to climb in emerging economies. Retrieved June 6, 2017, from http://www.pewglobal.org/2016/02/22/smartphone-ownership-and-internet-usage-continues-to-climb-in-emerging-economies/

57.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon & Schuster.

58.

Ragnedda

(2017). The third digital divide: A Weberian approach to digital inequalities. Abingdon, England: Taylor & Francis.

59.

Rice

R. E.

Katz

J. E.

(2003). Comparing internet and mobile phone usage: Digital divides of usage, adoption, and dropouts. Telecommunications Policy, 27, 597–623. doi:10.1016/S0308-5961(03)00068-5

60.

Rodriguez

Seligson

M. A.

(2008). Cultura política de la democracia en Colombia, 2008: El impacto de la gobernabilidad [Political culture of democracy in Colombia, 2008: The impact of governability]. Bogotá, Colombia: LAPOP.

61.

Rojas

Barnidge

Abril

E. P.

(2016). Egocentric publics and corrective action. Communication and the Public, 1, 27–38. doi:10.1177/2057047315619421

62.

Rojas

Puig-i-Abril

(2009). Mobilizers mobilized: Information, expression, mobilization and participation in the digital age. Journal of Computer-Mediated Communication, 14, 902–927. Retrieved from https://doi.org/10.1111/j.10836101.2009.01475.x

63.

Selva

(2015). Social television: Audience and political engagement. Television & New Media, 17, 159–172. doi:10.1177/1527476415616192

64.

Selwyn

(2004). Reconsidering political and popular understandings of the digital divide. New Media & Society, 6, 341–362. doi:10.1177/1461444804042519

65.

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. doi:10.1177/0093650205279209

66.

Somi

M. F.

Butler

J. R.

Vahid

Njau

J. D.

Kachur

S. P.

Abdulla

(2008). Use of proxy measures in estimating socioeconomic inequalities in malaria prevalence. Tropical Medicine & International Health, 13, 354–364. doi:10.1111/j.1365-3156.2008.02009.x

67.

Telles

Paschel

(2014). Who is black, white, or mixed race? How skin color, status, and nation shape racial classification in Latin America. American Journal of Sociology, 120, 864–907. doi:10.1086/679252

68.

Tirado-Morueta

Mendoza-Zambrano

D. M.

Aguaded-Gómez

J. I.

Marín-Gutiérrez

(2017). Empirical study of a sequence of access to internet use in Ecuador. Telematics and Informatics, 34, 171–183. doi:10.1016/j.tele.2016.12.012

69.

World Bank. (2015). Colombia—Systematic country diagnostic. Washington, DC: Author. Retrieved from http://documents.worldbank.org/curated/en/-142801468188650003/Colombia-Systematic-country-diagnostic

70.

Vaccari

Chadwick

O’Loughlin

(2015). Dual screening the political: Media events, social media, and citizen engagement. Journal of Communication, 65, 1041–1061. doi:10.1111/jcom.12187

71.

Van Deursen

A. J.

Van Dijk

(2014). The digital divide shifts to differences in usage. New Media & Society, 16, 507–526. doi:10.1177/1461444813487959

72.

Van Dijk

(2005). The deepening divide: Inequality in the information society. Thousand Oaks, CA: Sage.

73.

Van Dijk

Hacker

(2003). The digital divide as a complex and dynamic phenomenon. The Information Society, 19, 315–326. doi:10.1080/01972240309487

74.

Velasquez

Rojas

(2017). Political expression on social media: The role of communication competence and expected outcomes. Social Media + Society, 3. doi:10.1177/2056305117696521

75.

Vergeer

Franses

P. H.

(2016). Live audience responses to live televised election debates: Time series analysis of issue salience and party salience on audience behavior. Information, Communication, & Society, 16, 1390–1410. doi:10.1080/-1369118X.2015.1093526

76.

Warren

(2007). The digital vicious cycle: Links between social disadvantage and digital exclusion in rural areas. Telecommunications Policy, 31, 374–388. doi:10.1016/j.telpol.2007.04.001

77.

Warschauer

(2004). Technology and social inclusion: Rethinking the digital divide. Cambridge: MIT Press.

78.

Wei

(2012). Number matters: The multimodality of internet use as an indicator of the digital inequalities. Journal of Computer-Mediated Communication, 17, 303–318. doi:10.1111/j.1083-6101.2012.01578.x

79.

Wei

Hindman

D. B.

(2011). Does the digital divide matter more? Comparing the effects of new media and old media use on the education-based knowledge gap. Mass Communication and Society, 14, 216–235. doi:10.1080/15205431003642707

80.

Wells

Van Thomme

Maurer

Hanna

Pevehouse

Shah

D. V.

Bucy

(2016). Coproduction or cooptation? Real-time spin and social media response during the 2012 French and US presidential debates. French Politics, 14, 206–233. doi:10.1057/fp.201

81.

Wohn

(2011). Tweeting about TV: Sharing television viewing experiences via social media message streams. First Monday, 16. doi:10.5210/fm.v16i3.3368

82.

Yang

Grabe

M. E.

(2011). Knowledge acquisition gaps: A comparison of print versus online news sources. New Media & Society, 13, 1211–1227. doi:10.1177/-1461444811401708

83.

Zillien

Hargittai

(2009). Digital distinction: Status-specific types of internet usage. Social Science Quarterly, 90, 274–291. doi:10.1111/j.1540-6237.2009.00617.x