Media Participation in Midterm Elections: Creative and Consumptive Influences on Political Engagement

Abstract

Building upon the media participation hypothesis, this study examines the relationship between creative and consumptive media use during two pivotal midterm elections in the U.S.A series of regression analyses utilizing original, nationally representative data from 2014 and 2018 were modeled across dimensions of campaign participation, crossover political talk, and political system efficacy. Overall, support is found for increased creative media activity in comparison to more passive consumption in predicting political engagement. The results introduce additional nuance to the media participation literature while contextualizing the evolving nature of the uses made of interactive media for civic purposes.

1. Introduction

This chapter uses survey data across two points in time to investigate two longstanding concerns of political communication scholars. The first is the nature of media use during election campaigns and the effects that different forms of news following an expressive engagement have on political outcomes and evaluations. Questions about media influence on campaigns are foundational to the field and constitute some of the classic works on public opinion and American politics (see [1, 2]). A more recent, though no less pressing, concern has to do with Donald Trump’s impact and norm-busting influence on the American political system [3]. The chaotic reign that Trump’s election unleashed on normal government functioning and lack of adequate checks and balances has been cast as nothing short of a crisis of democracy [4].

From a research perspective, Trump’s election and conduct in office provides a political rupture significant enough to serve as an environmental shock worthy of before and after analysis. Despite general satisfaction with then-President Obama, the midterm elections of 2014 were a lackluster affair that did not attract much voter interest [5] but nevertheless set the stage for the Republican Party’s reascension, as GOP lawmakers regained control of the Senate and effectively blocked any meaningful legislative action—and Supreme Court appointments—during the last 2 years of the Obama Administration. Four years later, the situation could not have been more different—and dire. Pundits, mainstream news organizations, and former government officials were sounding the alarm that the project of American democracy itself was under such attack that not restoring partisan balance to Congress (by voting Democratic) would represent a potentially irreversible step towards authoritarian rule (e.g., [6]). Hearing these calls, voters responded with turnout during the 2018 midterms at levels normally only seen during presidential election years—49% nationwide [7].

These contrasting election contexts provide the backdrop for an examination of passive and active media use and its role in both political involvement and assessments of system functioning when the political stakes are seemingly routine on the one hand and extraordinary on the other. Under such disparate circumstances, we would expect differing modes of media use to be associated with varying levels of political interest and concern while significantly impacting depth of campaign attention and involvement. At the same time, attention and engagement with news that places citizens in conversations about politics should shape overall evaluations of how well the political system is serving the needs of citizens, or the sense of system efficacy.

In this paper, we analyze data from two national surveys conducted during the two national midterm elections in the U.S., the first in fall 2014 when for various reasons (e.g., the number of problematic police shootings caught on video, the rise of the Black Lives Matter movement that some equated with urban unrest, and Barack Obama’s second term as president) the country seemed to be drifting rightward but seemed relatively stable politically, and the second in fall 2018 when political institutions were under such consistent attack from the Trump White House, which itself was engulfed in scandal, that the project of democracy seemed increasingly precarious. These contrasting years allow us to investigate the role of media use in promoting civic engagement and resilience under varying political conditions. In examining media use, we distinguish between passive news consumption (consumptive use) that entails exposure to traditional media without any interactivity, and active media use (creative activity) that entails more proactive engagement with digital platforms and fellow citizens through technology.

In the democracy under threat context (the 2018 midterms), we predict that different forms of media use (consumptive vs. creative) will have more influence on democratic outcomes (political system efficacy, campaign participation, and talking about politics across party lines) than in the more stable context of moderate urban unrest (the 2014 midterms). Because active engagement with digital media for political purposes (i.e., “media participation”) is tantamount to active political participation, creative media activity should also positively predict campaign involvement. And as an affirmation of the legitimacy and functioning of the system, higher levels of creative media activity should positively influence the sense of political efficacy. We expect these effects to be contingent, however, on perceptions of political urgency and what’s at stake in a given election.

2. Media use and political involvement

In the analog era, the influence of different forms of print and broadcast media was considered independently and compared for their influence on such outcomes as voting intention, political knowledge, interest, and attitudes (e.g., [8, 9]). With the rise of cross-platform news brands and content, and concurrent development of interactive features on media platforms that support civic and political involvement [10, 11], it has become more important to track media behaviors and engagement with content than insisting on arbitrary distinctions between legacy distribution channels [12]. Traditional media use of course still exists, but when describing citizen engagement with politics through media, platform distinctions are increasingly fluid and a multiscreen culture now prevails [13, 14].

In the contemporary mediascape, therefore, the emphasis should be placed on varying forms of engagement that transcend delivery platforms and instead consider different modes of user activity as a reflection of one’s engagement with politics (see [15]). Here, a more passive mode of attention-paying would suggest a spectator view towards politics, where the political sphere is regarded as something to observe from a distance rather than attempt to engage with proactively. Fundamentally, passive spectatorship implies a consumptive orientation towards the political process, where different candidates, policies, and choices are regarded as something to intellectually consider and informationally consume at arm’s length but not necessarily partake in beyond that.

By contrast, an active mode of engagement suggests a gladiatorial view of citizen involvement at the media/politics interface, where through a combination of online and social media activity, efforts to acquire political knowledge, clarify and persuade take on a more assertive and goal-oriented cast. Such activity may occur across different platforms (e.g., social media, news sites, and topic-driven user communities) and devices (e.g., mobile media, laptops, and desktops) and at varying levels of analysis depending on the scope of the audience. Indeed, on distributed platforms, it is increasingly important to recognize new hybrid forms of political engagement that are simultaneously forms of both interpersonal and mass communication—or “masspersonal” communication [15, 16].

As media and technology researchers are showing with growing clarity, engagement with participatory media formats increases in different ways with each passing election cycle, showing generally positive outcomes on political attitudes and perceptions [17, 18, 19], although most notably for people who are already politically active [20]. A key explanation for this effect is the feeling of individual empowerment and efficacy that networked technologies afford (see [11]). Indeed, this is a core appeal of social media that lack institutional gatekeepers and which rely on user-generated content. Whether media participation affects actual change (see [21]), online platforms make accessible to citizens a political system that otherwise seems highly orchestrated, professionalized, and out of reach. At the individual level, use of interactive tools and technologies may thus offer a feeling of participatory empowerment capable of producing various social or civic “rewards,” including knowledge gain, proximity to important people or events, or a heightened sense of system satisfaction [22].

Though unlikely to influence an election or change policies directly, mediated forms of participation are important because they provide a ready avenue of active involvement that transcends consumptive surveillance of the political environment. Indeed, although considered dysfunctional for democratic stability, emotional objections to Donald Trump’s 2020 election loss and protests following the presidential election were largely amplified and organized through social media [23]. By allowing people to enact their civic role and engage with others in politically relevant spaces, networked platforms and technologies satisfy the need for popular involvement in civic life by delivering a continuous stream of daily opportunities for active citizenship—something the political system absent electronic media and digital platforms is unable to do. Moreover, civically consequential media behaviors are often highly consistent with, and overlap, “real world” political involvement [20, 21, 24].

Thus, media participation works to enhance the perception of political accessibility and openness by, first and foremost, giving citizens the opportunity to act as citizens [25], even if that energy is at times misdirected.

2.1 Predictions

Based on this discussion, the following three hypotheses are posed, each predicting a positive relationship between creative media activity and political involvement. The analysis they support add to our understanding of active versus passive engagement with media use, particularly in the understudied context of midterm elections:

H1: Increased participatory media use in the form of creative media activity will be positively associated with higher levels of campaign participation compared to (a) consumptive media use and (b) this tendency will manifest over time.

H2: Increased participatory media use in the form of creative media activity will be positively associated with higher levels of crossover political talk compared to (a) consumptive media use and (b) this tendency will manifest over time.

H3: Increased participatory media use in the form of creative media activity will be positively associated with higher levels of political system efficacy compared to (a) consumptive media use and (b) this tendency will manifest over time.

3. Method

3.1 Data

This study is based on analysis of two separate, nationally representative surveys that measure the same behavioral, cognitive, and affective variables before the 2014 and 2018 U.S. midterm elections. Demographically, the surveys were fielded using stratified sampling methods that carefully constructed national samples reflecting Census data distributions. The 2014 survey was fielded online between October 20 and November 3 (Election Day was November 4) by Social Survey International (SSI) and the 2018 survey was fielded online between October 28 and November 5 (Election Day was November 6) by Simple Opinions using the Prodege Market Research panel. After removing incomplete responses, the final datasets were compiled (N₂₀₁₄ = 1142, N₂₀₁₈ = 1169). Among other variables, this study takes into account partisan differences among Democrats (N_2014-D = 474, N_2018-D = 380), Republicans (N_2014-R = 259, N_2018-R = 422), and Independents (N_2014-I = 358, N_2018-I = 336). In both years, just a small number of respondents identified with other parties (N_2014-Other = 51, N_2018-Other = 31).

3.2 Independent variables

The key independent variables in this study were the use of different types of media, in particular the use of such media for what can be considered consumptive or creative activities. Consistent with earlier analyses [12], these different uses can be placed on a continuum of increasing participation with media. In the contemporary media environment, a “traditional” versus “emerging” view of technology may not fully reflect the role that interactivity now plays in people’s daily media practices. Based on this conceptualization, we created two blocks of activities, each consisting of six different self-reported behaviors to reflect consumptive (i.e., non-participatory) and creative (i.e., participatory) media use.

Consumptive activities: Consumptive activities include reading a newspaper, watching television, listening to radio, using social media for information gathering, using a news aggregator, and paying attention to other online sources for political and social issues. These six items act as the baseline for understanding the extent to which users were receiving information from various media platforms but not necessarily actively participating with that media.

More specifically, participants in both rounds of survey collection responded to the question “On average, how many days per week do you consume news using each of the following?” and indicated their weekly use frequency (0 = none to 7 = everyday) for each of the following media. Options here included reading a newspaper in print or online (M₂₀₁₄ = 3.29, SD₂₀₁₄ = 2.81; M₂₀₁₈ = 2.09, SD₂₀₁₈ = 2.56), watching television news (M₂₀₁₄ = 5.36, SD₂₀₁₄ = 2.41; M₂₀₁₈ = 3.88, SD₂₀₁₈ = 2.72), listening to talk and news radio (M₂₀₁₄ = 2.93, SD₂₀₁₄ = 2.65; M₂₀₁₈ = 2.34, SD₂₀₁₈ = 2.47), using social media for news (M₂₀₁₄ = 2.64, SD₂₀₁₄ = 2.89; M₂₀₁₈ = 3.01, SD₂₀₁₈ = 2.80), and using news aggregators such as Google News (M₂₀₁₄ = 2.94, SD₂₀₁₄ = 2.80; M₂₀₁₈ = 2.74, SD₂₀₁₈ = 2.64). Another consumptive activity was simply “paying attention to news online” measured using a 7-point scale, where 1 = none and 7 = everyday (M₂₀₁₄ = 4.21, SD₂₀₁₄ = 1.98; M₂₀₁₈ = 4.54, SD₂₀₁₈ = 2.01).

Creative activities: On the other hand, creative activities refer to those forms of engagement where media users participate in the original development of media content or actively share or elaborate on media content provided by others. In the data used for the study reported here, creative activities include searching, forwarding, or circulating information, “liking” a post, adding a post, creating original content, and participating in discussions.

To clearly identify how these measures were operationalized, participants in both the 2014 and 2018 surveys provided responses to the question “Please indicate how often you have participated in any of the following online activities in the past month.” Here again, respondents indicated their frequency of partaking in each activity using a 7-point scale. Options included searched for information about the candidates (M₂₀₁₄ = 3.57, SD₂₀₁₄ = 2.05; M₂₀₁₈ = 4.02, SD₂₀₁₈ = 2.09), forwarded or circulated information about them (M₂₀₁₄ = 2.95, SD₂₀₁₄ = 2.08; M₂₀₁₈ = 2.86, SD₂₀₁₈ = 2.03), “liking” a post about them (M₂₀₁₄ = 2.73, SD₂₀₁₄ = 2.09; M₂₀₁₈ = 3.36, SD₂₀₁₈ = 2.25), posting original content about them (M₂₀₁₄ = 2.51, SD₂₀₁₄ = 2.01; M₂₀₁₈ = 2.72, SD₂₀₁₈ = 2.04), creating original content about them (M₂₀₁₄ = 1.92, SD₂₀₁₄ = 1.69; M₂₀₁₈ = 2.03, SD₂₀₁₈ = 1.72), and participating in a discussion about them (M₂₀₁₄ = 2.66, SD₂₀₁₄ = 2.00; M₂₀₁₈ = 2.85, SD₂₀₁₈ = 1.99).

3.3 Dependent variables

Based on earlier work [12, 20, 21], we examined the relationship between media engagement and three key outcome measures shown to be consequential for enacting citizenship.

Campaign participation: To model how participating in the campaign related to participating with media, respondents provided self-reports to a five-item, 7-point scale that captured the frequency of their participation in pre-election campaign activities ranging from, for example, “displayed a political button, sticker, or sign” to “volunteered for a campaign to help get a candidate elected,” where we have 1 = not at all and 7 = very frequently as options. Participation scores were generated by calculating the mean for each individual’s participatory activity frequency (M₂₀₁₄ = 1.86, SD₂₀₁₄ = 1.50; M₂₀₁₈ = 1.67, SD₂₀₁₈ = 1.33). While it may be counterintuitive that our measure of participation was not found to be increasing over time, this may well reflect how citizens generally view midterm elections as being less pivotal than presidential election years and their contributions less clearly pronounced. 2018 was also a highly polarized year in American politics and the overall climate could have discouraged certain voters from participating.

Political conversation: Respondents were asked to estimate how often in a typical week they have in-person or mediated conversations about politics with different social groups including friends, family, coworkers, and acquaintances (0 = none, 7 = everyday). An average was calculated as an individual’s score for general political conversation (M₂₀₁₄ = 1.49, SD₂₀₁₄ = 1.77; M₂₀₁₈ = 2.79, SD₂₀₁₈ = 1.64). In addition to the questions about general political conversation, respondents were also asked using the same 0 to 7 scale as to how frequently, in a typical week, they engaged in conversation with people who “express political views you disagree with” (M₂₀₁₄ = 1.39, SD₂₀₁₄ = 1.91; M₂₀₁₈ = 2.21, SD₂₀₁₈ = 1.74).

Political system efficacy: For this construct, respondents provided input to a four-item, 7-point Likert scale adapted from a widely used scale [26] about their political system efficacy. Measures here included questions, such as “There are many legal ways for citizens to successfully influence what the government does” on a 7-point agreement scale, where 1 = strongly disagree to 7 = strongly agree scale (M₂₀₁₄ = 3.99, SD₂₀₁₄ = 1.19; M₂₀₁₈ = 3.99, SD₂₀₁₈ = 1.23).

3.4 Control variables

We also control for germane demographic variables including gender (Male₂₀₁₄ = 55%, Male₂₀₁₈ = 49%), age (M₂₀₁₄ = 48.43, M₂₀₁₈ = 53.76), education (M₂₀₁₄ = some college or 2-year degree, M₂₀₁₈ = some college or 2-year degree). We also control for individual partisanship strength (M₂₀₁₄ = 4.85, SD₂₀₁₄ = 1.77; M₂₀₁₈ = 5.01, SD₂₀₁₈ = 1.63).

4. Results

The first hypothesis expected that increased participatory media use in the form of creative media activity will be positively associated with higher levels of campaign participation. In comparison with consumptive media activities, creative media activities were not only a consistent and positive predictor of campaign participation, they also produced the largest regression coefficients in both 2014 (β = 0.454, SE = 0.028, p < .001) and 2018 (β = 0.365, SE = 0.024, p < .001). There is thus evidence to support H1a, which is summarized in Table 1.

	2014		2018
Variables	B	(SE)	B	(SE)
Newspaper	.065	(.012)**	.063	(.014)*
TV	−.026	(.014)	.032	(.013)
Radio	.044	(.013)	.047	(.014)
News Aggregator	−.053	(.013)*	.016	(.014)
Social Media	−.029	(.014)	.011	(.014)
Online Pay Attention; Online Search	−.023; .052	(.024); (.028)	−.087; .066	(.011)*; (.023)
Online Share	.077	(.027)*	.035	(.026)
Online “Like”	.081	(.026)*	−.008	(.022)
Online Post	.052	(.033)	.080	(.027)
Online Discuss	.072	(.028)	.134	(.023)***
Online Create	.454	(.028)***	.365	(.024)***
N	1142		1169
Adjusted R^2	.510		.376

Table 1.

Regression analyses of campaign participation, 2014 compared to 2018.

^*

p ≤ .05.

^**

p ≤ .01.

^***

p ≤ .001.

Note: Coefficients reported are standardized and results summarize the final media blocks in a hierarchical regression model where R² change was significant for demographic, partisanship, consumptive, and creative blocks.

When looking at comparisons over time, while the relative strength of the creative media use measure declined slightly from 2014 to 2018, the effect of discussing politics online became significant and weighed heavily on the model in 2018 (β = 0.134, SE = 0.023, p < .001). Other factors that were significant in 2014 dropped from statistical significance by 2018, namely sharing and liking political posts online. The overall takeaway from comparing these models in terms of campaign participation is that the influence of more participatory, creative activities is variable over time and contributes more, not less, than consumptive activities, where the net effect is either stagnant or declining over time.

Moving forward, the second hypothesis predicted that increased participatory media use in the form of creative media activities will be positively associated with higher levels of crossover political talk. Compared to consumptive media activities, a similar pattern emerged. Creating media online, discussing politics online, and posting about politics online all had consistently strong standardized β coefficients that, as shown in Table 2, seemed to drive crossover political talk in both 2014 and 2018 to a greater extent than consumptive activities. It is, however, worth noting that political talk radio use was significant in both 2014 (β = 0.093, SE = 0.019, p < .001) and 2018 (β = 0.076, SE = 0.019, p < .01), which connects meaningfully to previous conceptualizations of the media participation hypothesis [22, 25]. On the whole, there is general support for H2a.

	2014		2018
Variables	B	(SE)	B	(SE)
Newspaper	.054	(.017)*	.051	(.019)
TV	.034	(.020)	.108	(.018)***
Radio	.093	(.019)***	.076	(.019)**
News Aggregator	.073	(.019)**	.023	(.019)
Social Media	−.03	(.020)	.058	(.019)
Online Pay Attention; Online Search	.087; −.049	(.034)*; (.039)	−.037; .091	(.032); (.032)*
Online Share	.131	(.038)**	−.007	(.037)
Online “Like”	−.015	(.037)*	−.016	(.030)
Online Post	.159	(.047)**	.167	(.038)***
Online Discuss	.180	(.039)***	.194	(.032)***
Online Create	.180	(.040)***	.135	(.038)***
N	1142		1169
Adjusted R^2	.398		.302

Table 2.

Regression analyses of crossover political talk, 2014 compared to 2018.

^*

p ≤ .05.

^**

p ≤ .01.

^***

p ≤ .001.

Note: Coefficients reported are standardized and results summarize the final media blocks in a hierarchical regression model where R² change was significant for demographic, partisanship, consumptive, and creative blocks.

When looking at the contribution of more participatory media use between 2014 and 2018, there are again general gains. For example, online searches for political information were significant in 2018 (β = 0.091, SE = 0.032, p < .05) and the action of online “liking” went from negative and significant in 2014 to non-significant in 2018. Moreover, the standardized coefficients online posting about politics increased between 2014 (β = 0.159, SE = 0.047, p < .01) and 2018 (β = 0.167, SE = 0.038, p < .001). The same was true of online discussions about politics between 2014 (β = 0.180, SE = 0.039, p < .001) and 2018 (β = 0.194, SE = 0.032, p < .001). For online content creation, however, the coefficients decreased slightly between 2014 (β = 0.180, SE = 0.040, p < .001) and 2018 (β = 0.135, SE = 0.038, p < .001), and online sharing, which was significant in 2014, was no longer significant in 2018. There is thus overall but not overwhelming support for H2b.

Finally, when considering the third hypothesis, that increased participatory media use in the form of creative media activities would be positively associated with higher levels of political system efficacy, there was no evidence to support the prediction (see Table 3). The analysis showed a negative relationship between online discussions and political system efficacy in 2014 (β = −0.120, SE = 0.030, p < .05) and lack of significance altogether in 2018. When looking at consumptive activities, a few significant associations materialized, including television viewing in 2018 (β = 0.066, SE = 0.015, p < .05) and paying attention online in both 2014 (β = 0.186, SE = 0.026, p < .001) and 2018 (β = 0.091, SE = 0.013, p < .05). But, when looking more closely at the change across time, not only were none of the creative activities significant in 2018, the R-squared change of the entire block was non-significant.

	2014		2018
Variables	B	(SE)	B	(SE)
Newspaper	−.025	(.013)	.049	(.016)
TV	.054	(.015)	.066	(.015)*
Radio	.058	(.014)	.018	(.016)
News Aggregator	−.025	(.014)	.025	(.016)
Social Media	.008	(.008)	−.006	(.016)
Online Pay Attention; Online Search	.186; .078	(.026)***; (.030)	.091; .052	(.013)*; (.026)
Online Share	.081	(.029)	.080	(.030)
Online “Like”	−.075	(.028)	−.068	(.025)
Online Post	.070	(.036)	−.010	(.031)
Online Discuss	−.120	(.030)*	.080	(.027)
Online Create	−.034	(.030)	−.062	(.028)
N	1142		1169
Adjusted R^2	.105		.055

Table 3.

Regression analyses for political system efficacy, 2014 compared to 2018.

^*

p ≤ .05.

^**

p ≤ .01.

^***

p ≤ .001.

Note: Coefficients reported are standardized and results summarize the final media blocks in a hierarchical regression model where R² change was significant for demographic and consumptive blocks.

5. Conclusion

From a critical point of view, consumptive media practices flow from a detached view of citizenship where responsibility for political change is vested in elites and other civic actors. As consumers, citizens are content to monitor the information environment and may decide to vote when formally called upon to take part in elections but little else—and even then, when it comes to midterm elections especially, most do not even vote (see [27]). In his political writings, Lippmann [28, 29] described such an arrangement as desirable, given the growing complexity of governing, diplomacy, and modern society: Why rely on the involvement of a disengaged “phantom public” whose grasp of the issues was incomplete at best? In Lippmann’s view, the onus of civic vitality rightly rested with a cadre of educated and technocratic elites, who should remain circulating and accessible to the masses but who should make the critical decisions for society without much interference. Such forms of elite pluralism, however, are inevitably considered thin versions of democracy.

A more engaged view of citizenship is supported by creative media activities, where the conception of citizen-as-spectator shifts to citizen-as-participant with the pursuit of various mediated behaviors including content generation and sharing during elections as well as interactively responding to content provided by others. Granted, it does not take much effort to share posts on media platforms and occasionally comment on an issue or candidate. Nevertheless, generating messages and engaging in debate do imply a distinctly stronger form of civic involvement than passive reception of political information. Such forms of media participation are important because they facilitate the right of all to participate in democratic discussion, which theorists have noted is an important norm governing the behavior of political elites ([25, 30], p. 152).

Our findings are consistent with this view. Creative media activities were not only consistent and positive predictors of campaign participation, they also generated the largest coefficients across both elections studied. Although we found that the influence of individual creative activities changes over time and with varying election contexts, as a class of media behaviors they consistently contribute more to active citizenship than consumptive activities, where the net effect is either stagnant or declining over time. Creating media online, discussing politics online, and posting about politics online also all had consistently stronger positive relationships with crossover political talk, that is, engaging in discussions with people whose political views respondents disagreed with, compared to consumptive activities. Thus, across two very unique elections, participatory media engagement contributes positively to active campaign participation and to a hallmark quality of democratic citizenship: engaging in political conversation with non-like-minded others.

Interestingly, and unlike relationships we have documented in presidential election years (see [12]), participatory media use had almost no effect on estimates of political system efficacy—the notion that the political system is open, responsive, and works effectively on behalf of citizens. In fact, the only significant coefficient, for online discussions, was negative in 2014. However, consumptive media use, namely, paying attention to politics online (both years) and television news use (in 2018) was positively associated with assessments of system efficacy, although not strongly. The weak effects for both election years, and both forms of media engagement (creative and consumptive) suggest the presence of an underlying third variable that may be moderating the relationship. Further modeling should assess whether this is the result of the midterm election context or perhaps underlying partisan dynamics.

As with any empirical study, this research is not without limitations. As noted, some of the relationships documented here are significant but not strong. Additional studies might identify and model individual differences between users to more precisely identify how interactive media use enhances perceptions of system responsiveness. Technological sophistication is likely to play an important role, as are media repertoires or patterns of media use and engagement across platforms and content genres. With the identification of moderators, path modeling might better identify the sequence of causality. Additionally, motivations for media use, which in the media participation literature overlap with motivations for political involvement, likely drive the effects that creative media use can have. If a user’s goal is merely to be entertained, we would expect less civic impact from either creative or consumptive media use. If, however, the goal is to be informed and socially engaged, then the door opens to a broader panoply of expected outcomes.

As research advances, examining the correlates of creative and consumptive media influence across citizens with varying levels of technological and political sophistication, and in ever-more precisely defined networked environments, becomes a compelling proposition. How does media expertise interact with political sophistication to affect democratically important outcomes? Technologically, are so-called “power users” more satisfied with system functioning than “end users” who gravitate towards consumptive media—even online? If so, what factors moderate these relationships? Towards this end, future research should consider the contingent conditions under which different audience segments navigate the political sphere, integrating a wider array of political outlooks and behaviors as outcome measures. It remains to be determined, with more precise measures and modeling, how trends in creative and consumptive media use relate to other critical considerations such as political cynicism and voting.