Something Ventured, Something Gained

By Daekyung Kim, Thomas J. Johnson and Barbara K. Kaye

WJMCR 24 (September 2010)

Introduction|Literature Review|Research Questions|Method|Results|Discussion

Abstract

Citizen conversation about public affairs has long been considered crucial in the functioning of a democracy. Studies have found that exposure to news media alone does not automatically make people informed. This online survey of 1,366 politically interested respondents tests the importance of interpersonal communication on political activity and the importance of interpersonal communication through blogs on political activity. The study found that reliance on blogs did not affect political activity.

Introduction

Citizen’s conversations about public affairs have long been considered as playing a crucial role in the functioning of a participatory democracy, with a belief that interpersonal discussion may increase political knowledge and facilitate political participation.1 Past research suggests that political discussion leads to various forms of political participation and an increased understanding of political issues.In this process, news media provide citizens with information they can use in discussing with other people, which may in turn help them engage in political participation.3

In particular, Scheufelesystematically suggested a differential gains model of news media effects research, which called for an understanding of the interactive effects between interpersonal discussion and news media on political participation. He found that people who talked about political issues more frequently reported higher levels of political knowledge and political participation than those who did not so.

The growing popularity of the Internet as a political platform drew the interest of political communication researchers to investigate the effects of online interpersonal discussion on political participation. The interactive nature of the Internet offers people an abundance of opportunities for political discussion and participation by empowering them to be active parts of the political process. For example, the ease of participating in online political discussion allows a great diversity of viewpoints and exchange of information. As Berman and Weitzner5 pointed out, the interactive nature of the Internet is “another key attribute that makes it such a unique and effective forum for democratic discourse.”

Therefore, a recent study examined the differential gain effects of interpersonal discussion through online chatting, and found that it moderated the effects of Internet news on political participation.6 However, online chatting alone does not represent various levels of interpersonal discussion that take place on the Internet. Recently blogs have drawn considerable attention from media researchers as a new communication medium and emerged as a main source of political discourse.

The blogosphere, referred to as the universe of blogs, has been touted as ‘a democratic and interactive space’ that fosters a healthy debate among blog users.Especially political blogs became one of the most interesting research topics during the 2004 presidential election.8 Accordingly, by adopting the differential gain model, this study examined not only the effects of traditional interpersonal discussion but also the potential of blogs in promoting political activity through an online survey conducted during the 2004 presidential election.

Literature Review

The Differential Gain Model: Moderating effects of interpersonal discussion

Interpersonal communication has consistently been one of the underlying elements in political communication research. Previous studies found that simple exposure to the news media messages does not automatically transform people into informed citizens.Instead, Robinson and Levy10 found that interpersonal discussion on political news was a more significant predictor of news comprehension than exposure to the political information in the news media.

McLeod and associates11 pointed out that the media could stimulate interpersonal discussion and political interest during the election campaign, and many studies found that news media use was significantly related to political talk, which in turn led to higher levels of political knowledge and political participation.12 Those findings indicate that the political effects of news media were contingent upon interpersonal discussion about political issues with other people.

In this vein, Scheufele13 suggested the differential gains model that interpersonal discussion on politics would moderate the effect of mass media on political participation, based upon a belief that interpersonal discussion about political affairs is key to a healthy and functioning democracy. The primary argument is that political effects of news media messages are contingent upon interactive effects of mass and interpersonal communication. Thus, interpersonal discussion on politics with other people could moderate the effects of news media on citizens’ political knowledge and participation.

Scheufele14 found that hard news media use was significantly related to political talk, and also found significant interaction effects between media use and political discussion on political participation. In other words, people who more frequently discussed political issues covered by news media are more likely to participate in politics than those who discussed politics less frequently.

More specifically, Kwak and associates15 examined the influence of integrative discussion, ‘the degree of convergence between mass and interpersonal communication,’ on political knowledge and political participation. They found that interpersonal discussion about politics with others was significantly related to political knowledge and political participation, respectively, after controlling for demographics and structural variables of discussion. Indeed, the study illustrated “a quite dynamic nature of political discussion” in mobilizing citizens’ political engagement (p. 104).

Not all studies have found support for an information gains model. Hardy16 tested both a communication confusion and a differential gains model and found more support for a communication confusion model.17 Hardy speculated that more people discuss politics after key campaign events and such individuals are less knowledgeable or interested in politics, so that increased political discussion leads to confusion. Rush19 pointed out weaknesses in the differential gain model in that many types of political participations study involve communication; political talk needs to be separated from political participation. Also, political talk needs to be recognized as a multidimensional concept characterized by affect (positive or negative discussions about politics) and openness (whether it is between hetereogeneous or homogeneous groups). This study partially addresses Rush’s concerns by separating out blog reliance as a discussion-based political activity

Differential Gain Model and the Internet

During the past decade, the Internet has dramatically changed the way people gather news and participate in politics. While only 4% of the public went to the Internet for political news in 1996, nearly three in ten of the general population (29%) accessed online political news during the 2004 presidential election.20 Given the growing popularity of the Internet as a political news source, it is important to test the differential gain models on the Internet. The unique nature of the Internet is that it integrates different modes of traditional media use and interpersonal communication. 21

Unlike traditional news media, the Internet has greater potential to promote and facilitate interpersonal discussion with other fellow citizens through email, chat rooms, discussion forums, listservs and bulletin boards.22 Some researchers argued that cyberspace can be conceptualized as a discursive space, in which citizens can freely express their opinions and interact with others.23 Past studies found that people are motivated to use electronic space, such as the Web, electronic mailing lists/bulletin boards, and chat rooms/instant messaging, for accessing information and talking with others.24

More specifically, Hardy and Scheufele25 conducted a follow-up study of the differential gains model to examine whether interpersonal discussion through online chat rooms would also moderate the effects of Internet news on political participatory behaviors. They found that the effects of online news are contingent upon the degree to which individuals engage in political discussion about politics with other people through chat rooms. As mentioned above, interpersonal discussion through chat rooms, however, did not account for all interactive applications of the Internet that would promote interpersonal discussion.

In addition, Kim and Johnson26 examined a broader concept of interpersonal communications on the Internet, defined as “person-to-person communications mediated by interactive application of the Internet in both synchronous and asynchronous ways”, which are made possible through online polls, emailing news articles to friends or other people, responding to reporters or editors and participating in discussion forums. The study found that online news use alone was not related to political participation. Instead, the effects of Internet news use were mediated by both traditional face-to-face and online discussion. Accordingly, it is worth examining the role played by other interactive components of the Internet in facilitating interpersonal discussion online, which would in turn lead to higher levels of political participation.

Blogosphere as a discursive space

The focus of this study is on the political function of blogs as a mean of mobilizing interpersonal discussion about politics online. Blogs have drawn considerable attention from researchers as a new form of a communication medium.27 In particular, blogs were one of the most prominent buzzwords during the 2004 presidential election campaign in terms of the political effects.

Studies found that online political users during the 2004 presidential election were motivated to use blogs for social utility needs, indicating that they were blogging to get certain chances to express their thoughts as well as to communicate with other people.28 Eveland and Dylko29 also found that blog readers were more likely to use alternative forms of political communication like email and discussion forums than non-blog readers, and that blog reading was significantly related to political participation.

In this vein, some suggested that blogs can foster a sense of community and provide forum for public debate about certain political issues.30 Through posting links or “blogrolling.” a list of blogs that bloggers frequently read, bloggers or blog readers can not only visit the sites and post their news analysis and commentary, but also provide links of their own information.31

Matheson32 therefore pointed out that the blogosphere could serve as “a democratic and interactive space” in which people express their voices and interact with others through interactive forms of social or political communication. Furthermore, Scott33 argued that the nature of blogs offers “a useful paradigm for thinking about civic-oriented, participatory, deliberative online news sharing.”

On the other hand, because most political blogs typically are devoted to particular issues from a particular political ideology, they may limit the topic of conversation by only encouraging like-minded people. Kaye and Johnson34 examined the potential for blogs to create a true public sphere online, but they found that blogs are ‘issue-oriented zones’ that only attract people of similar interest who debate a limited number of topics. Similarly, Sunstein35 argued that political polarization is inevitable because of the linking patterns of the blogosphere in which liberals are connected mostly to liberals and conservatives mostly to conservatives. Therefore, despite the wider range of available information and perspectives, most blog readers are more likely to obtain one-side views of political issues, which are consistent with their pre-existing views.

Research Questions and Hypothesis

This study explores the two research question and two hypotheses:

  • Research Question 1: What is the impact of traditional interpersonal discussion on political activity?
  • Research Question 2: What is the impact of interpersonal discussion through blogs on political activity?
  • Hypothesis 1: The impact of online news use on political activity will be moderated by face-to-face discussion about politics.
    Hypothesis 2: The impact of online news use on political activity will be moderated by interpersonal discussion through blogs about politics.

Method

This study examined 1,366 responses to an online survey posted on the Web from October 19 to November 16, 2004, the two weeks before and the two weeks after the 2004 presidential election. The survey was specifically targeted to politically interested Web users. They were informed of the survey URL through announcements sent to media and politically-oriented newsgroups, Web sites, Weblogs, electronic mailing lists, and chat rooms representing a cross section of political ideologies.

Although this study relies on a convenience sample, the method used was appropriate for specifically reaching a narrow group of Internet users: Blog readers. Attempting to generate a random sample of this small, but influential, group would result in a large non-qualification rate because only a small percentage of Internet users access Blogs.36 Therefore, the best way to find Blog readers was to post announcements of the survey on Weblogs and other online resources that Weblog readers tend to access. Unlike telephone and mail surveys where samples can be produced through census lists and random digit dialing, the Internet has no central registry of users and e-mail addresses are so varied that they are virtually impossible to construct randomly.37Therefore, this study employs a convenience sample of Internet users who were directed to the survey through online announcements and hyperlinks.

Surveys were returned via e-mail accompanied by the date and time of receipt as well as the respondent’s Internet server address. Additionally, the survey’s first question asked respondents to enter their e-mail address; 1,244 complied. The Internet server and e-mail addresses were used to delete duplicate surveys and to track the number of surveys received each day.

Dependent Variable: Political activity

The dependent variable, political activity, was measured with a behavioral indicator. Behavioral political activity was assessed through five items taken from the National Election Studies. Respondents indicated whether they “ very often,” “often,” “sometimes,” “ rarely,” or “never,” engage in the following political activities: persuade others to vote for a political candidate or issue, promote a candidate or issue with a lapel button, bumper sticker, or yard sign, go to politically-oriented dinners, rallies or speeches, donate money to a candidate or issue, or do any other type of work to endorse a candidate or issue. The five items were summed to form a political activity index, which has a reliability of .85.

Independent Variables: Interpersonal discussion and blog reliance

The differential gains hypothesis argues that the interaction between interpersonal discussion and media use will increase the effects of the media on political activity. This study will compare the effects of traditional interpersonal communication with blog reliance. Blogs allow a range of involvement between reader and blogger. The reader can simply read the postings and click on links on the site. However, the blog reader can interact with the blogger by both commenting on the posts as well as suggesting news to be added to the blog. Similarly, the blog users can debate other comments made by blog readers. Respondents were asked how much they relied on face-to-face discussion with others as well as Weblogs for their political information. Possible responses were “heavily rely on,” “rely on,” “sometimes rely on,” “rarely rely on” and “don’t rely on at all.”

Online media reliance

This study explored how much interpersonal discussion and blog reliance interacted with Internet media reliance. This study employed three measures of online media reliance. Respondents were asked whether they “heavily rely,” “rely,” “sometimes rely,” “rarely rely,” and “don’t rely” on online newspapers, broadcast television news sites and cable news sites.

Traditional media reliance

This research also included three measures of traditional media use: newspapers, Cable TV news and broadcast news. Like the online news measures, possible responses were: “heavily rely,” “rely,” “sometimes rely,” “rarely rely,” and “don’t rely.”

Predispositions

Six political variables were included as predispositional measures: political interest, election interest, political ideology, party strength, trust in government and political efficacy.

Interest in politics in general and interest in the 2004 presidential election in particular were gauged through 10-point scales running from 0 (no interest) to 10 (extremely interested). Respondents were also asked how strongly they are affiliated with their party of choice (0 –10 scale ranging from no party ties to very strong ties), and their political ideology (very liberal=5, liberal=4, neutral=3, conservative=2, very conservative=1). Respondents were also asked whether they “strongly agree,” “agree,” “neutral,” “disagree,” or “strongly disagree” with three self-efficacy and three trust items taken from the National Election Studies. Self-efficacy was measured by the following three statements: “I consider myself well qualified to participate in politics,” “I think that I am better informed about politics and government than most people,” and “I feel that I have a pretty good understanding of the important political issues facing our country.” The three measures of trust in government are “Most of our leaders are devoted to the service of our country,” “Politicians never tell us what they really think,” and “I don’t think public officials care much about what people like me think.” The polarity was reversed on last two statements of the trust index. The trust and efficacy measures were combined into separate efficacy and trust indexes which ranged from 3-15. The reliability for the efficacy index is .73 and the trust index is .75. 

Demographics

This study also included four demographic measures: gender, age, education and income. Respondents were asked to indicate whether they are male or female, their age as of their last birthday, their highest level of education, and their estimated income for 2004.

Data Analysis

The data were analyzed in two stages. First, frequencies were run on the various demographic and predispositional variables. Second, a bivariate correlation was run to examine the relationships between the explanatory variables, including demographics, political attitudes, blog reliance and face-to-face talk. Finally, a hierarchical regression was conducted to determine whether the interaction between discussion and online media use was stronger than the effects of the two variables separately on the political activity. The political activity index was entered as the dependent variable, and the two discussion measures (face-to-face discussion and blog reliance) were examined separately, creating two different hierarchical equations. To determine the appropriateness of regression analysis, multicollinearity among the explanatory variables was examined on the correlation coefficient matrix. As shown in Table 1, none of the correlation coefficients between the explanatory variables was greater than 0.9. Also, tolerance values and the variance inflation factors (VIFs) were calculated to detect multicollinearity problem in the regression model. The tolerance values of all explanatory variables were greater than 0.20 and VIFs were less than 5.0. As a result, the hierarchical regression was considered appropriate because there was no multicollinearity problem with the explanatory variables.

The independent variables were entered as blocks with demographics first, followed by the predisposition measures (political interest, election interest, political ideology, party strength, trust in government and political efficacy), traditional media reliance (newspaper reliance, broadcast TV news reliance and cable TV news reliance), online media reliance (online newspaper reliance, broadcast TV news reliance and cable TV news reliance), and either face-to-face discussion or blog reliance. The final stage tested the interaction effects between either face-to-face discussions or blog reliance on online broadcast TV, online cable TV and online newspaper reliance. To avoid potential problems with multicollinearity between interaction terms and their components, all main effects variables were standardized prior to the formation of the interaction terms.

Results

Profile of Respondents

The respondents expressed considerable interest in politics in general and the election, Slightly more than 9 out of ten (92.4%) respondents expressed a strong interest in the 2004 presidential campaign and 73.6% agreed that they were very interested in politics in general. As shown in Table 1, the older are more interested in politics (r = .070, p< .05) and the election (r = .089, p< .01). Just less than one-quarter (23.8%) reported high levels of traditional political activity with most (42.6%) claiming low activity.

Four out of ten (41.1%) of these politically interested respondents were moderately trusting of the government with 28.4% reporting high levels of trust. On the other hand, nearly all (97.2%) of the respondents were confident in their power to bring about change and reported high levels of self-efficacy. Men with higher levels of political, election interest and party strength are more likely to report higher levels of political efficacy (Table 1).

Table 1

Bivariate Correlation Coefficients Between Explanatory Variables

Gender Age Edu. Income Political interest Election interest Political Ideology Party Strength Gov. Trust Political Efficacy Blog Reliance
Age .025                    
Edu. -.024 .091**                  
Income -.029 .011** .145**                
Political Interest .051 .070* -.020 .036              
Election Interest .001 .089** -.009 .040 .485**            
Political Ideology .202** .002 .054* -.031 .059* -.007          
Party Strength .080** .055* -.037 .007 .262** .221** -.022        
Gov. Trust -.035 -.043 .119** .035 .087** .170** -.228** .138**      
Political Efficacy -.109** -.003 -.027 .033 .388** .206** .034 .139** .020    
Blog Reliance -.105** -.077** .032 .028 .092** .095** -.138** -.061 .086** .122**  
Face-to-face talk .049 .062* -.102** -.019 .359** .262** .051 .261** -.001 .257** -.006

Note: ** p<.01      * p<.05

Almost half of the respondents (48.6%) claimed they are strongly affiliated with their party of choice: Republican (30.5%), Democrat (35.4%), Libertarian (5.4%), Green (1.2%) independent (15.6%) or other party (1.6%). Further, 38.1% considered themselves conservative/very conservative, 25.6% are moderate, and the remaining 36.3% said they held liberal/very liberal views.

These politically-oriented respondents rely/ heavily rely on Weblogs (63.4%) to get political information followed by political Web sites (56.7%), mailing lists/bulletin boards (35.1%), Web portals (30.0%), candidate sites (29.1%), and chat rooms (8.5%). Respondents spend an average of 8.3 hours per week on political sites, followed by 7.4 hours per week on Weblogs. More than half (54.4%) reported that they often and very often discussed about political issues with others during the election. According to Table 1, respondents with higher levels of political, election interest and political efficacy are more likely to rely on blogs and face-to-face discussion about political issues.

Respondents were overwhelmingly well-educated white males of moderately high incomes. More than 6 in ten (62.6%) of those who completed the survey were males. Nearly all (97.3%) have some college, a college degree or higher and the average age is 43.4 years. Most (91.2%) of the respondents were white, just more than 1 percent (1.3%) Black and the remaining classified themselves as other non-whites. The average annual income was a little more than $64,000. While this demographic profile doesn’t match studies of Internet users in general, it does resemble similar ones that have examined online news users. These studies also suggest those who seek out political information online tend to be well-educated white middle class males38.

Differential Gains Model

Differential gain researchers argue that scholars often underestimate the effects of the media on political attitudes because media use interacts with interpersonal discussion to produce effects. This study found that while face-to-face discussion was significantly linked to the political activity, reliance on blog was not a significant predictor of the political activity. Also, none of the interactions between interpersonal discussion, blog reliance and online media use proved significant.

Face-to-face discussion

Those who talk about politics with others are significantly more likely than those who do not discuss politics to engage in political activities such as persuading others to vote for a political candidate or issue, sporting a lapel button or attending politically-oriented dinners, rallies, or speeches (? = .27, p< .01) (Table 2). Interpersonal discussion (R2 = 6.3%, p< .01) explained more variance in political activity than any other block of variables except predispositions. 
Predispositions were the strongest predictors of the three measures of political activity with 32.3% of the variance in traditional political activity (Table 2).

Table 2

Predicting Political Activity (Online Media x Interpersonal Discussion)

Before Beta Entry Final Beta
Demographics (4)    
Age   .14**
Sex   .10**
Education   -.02
Income                    -.02
Incremental R² (%) 5.7**  
Predispositions (6)    
Political Interest .27** .23**
Election Interest .07** .07**
Political Ideology             .17** .14**
Party Strength .31** .31**
Trust in Government   -.01 -.02
Political Efficacy .10** .09**
Incremental R² (%) 32.3**  
Traditional Media Reliance (3)    
Broadcast TV News .02 -.02
Cable TV News .01 .02
Newspapers News .02 -.01
Incremental R² (%) .1  
Internet Media Reliance (3)    
Online Broadcast TV News .03 .03
Online Cable TV News -.05 -.05
Online Newspapers News .01 .02
Incremental R² (%) .2  
Interpersonal Discussion (1)    
Interpersonal Discussion of Politics .27** .27**
Incremental R² (%) 6.3**  
Interaction (3)    
Online Broadcast TV x Political Discussion .01  
Online Cable TV x Political Discussion .00  
Online Newspapers x Political Discussion .00  
Incremental R² (%)   .0  
Total R² (%)    44.6

Note: ** p<.01      * p<.05

Which predisposition was the strongest predictor varied by political activity measures examined. For instance, for behavioral measures of political activity, those who reported a strong attachment to a political party (? = .31, p< .01) and high political interest (? = .23, p< .01) engaged in more political activities. Ideology (? = .14, p< .01), and election interest (? = .07, p< .01) also proved significant, with conservatives and those highly interested in the election reporting that they participated in more political activities.

Demographics were also a significant predictor of the measures of political activity, ranging from 5.7% of the variance in engaging in political activities. Gender and age were the most consistent predictors of political activity. Women were significantly more likely to engage in political activities (?=.10, p<.01), and older individuals said they participated in more political activities (?=.14, p<.01). Neither traditional media nor online news use had an influence on levels of political activity.

Blog Reliance

Blog users communicate with the creator of the blog by e-mailing his or her comments or items for the blog. Blog readers can also post messages on most blogs, expressing their views on the news or the opinions of their fellow blog users. However, blog reliance was not as strong a predictor of political activity as was interpersonal use, nor did it interact strongly with online media use to predict level of political activity (Table 3).

Table 3

Predicting Political Activity (Online Media x Blogs)

Before Beta Entry Final Beta
Demographics (4)    
Age   .11**
Sex   .12**
Education   -.04
Income                    -.01
Incremental R² (%) 5.7**  
Predispositions (6)    
Political Interest .27** .27**
Election Interest .07** .07**
Political Ideology             .17** .15**
Party Strength .31** .31**
Trust in Government   -.01 -.02
Political Efficacy .10** .11**
Incremental R² (%) 32.3**  
Traditional Media Reliance (3)    
Broadcast TV News .02 .01
Cable TV News .01 .03
Newspapers News .02 .02
Incremental R² (%) .1  
Internet Media Reliance (3)    
Online Broadcast TV News .03 .03
Online Cable TV News -.05 -.05
Online Newspapers News .01 .01
Incremental R² (%) .2  
Blogs (1)    
Reliance on Blogs -.01 -.02
Incremental R² (%) .0  
Interaction (3)    
Online Broadcast TV x Blogs .00  
Online Cable TV x Blogs .03  
Online Newspapers x Blogs -.01  
Incremental R² (%) .0  
Total R² (%)    38.3

Note: ** p<.01      * p<.05

Predispositions were again the strongest predictors of the political activity measures, explaining 32.3% of the variance in traditional political activity, (Table 3). Demographics were second strongest predictor of political activities, with age and gender having the strongest influence on the political activity measures. Similarly, both online and traditional media reliance were not predictors of political activity.

Discussion and Conclusion

The primary goal of this study was to examine if interpersonal discussions about politics plays a moderating role in the political effects of online news on political activity. This study examined the interaction between online news use and political discussion via both face-to-face and blog reliance.

Compared to other differential gain studies39, this study shows somewhat different results and weak support in terms of the moderating effects of interpersonal discussion on political activity. One possible reason could result from the characteristics of respondents. It should therefore be noted that the data come from self-responses of politically interested Internet users who seem to be already politically motivated users and go online for political purposes. The dispositions block, consisting of political variables, indeed accounted for the biggest variation explained in the regression models with other variables having a small portion of variance.
Despite this limitation, this study revealed some interesting results with regard to the political influences of blogs. For politically interested Internet users, there were no main effects of both traditional media and online media on political activity. Instead, a person’s interpersonal discussion on politics had the main effect, indicating that people who more frequently talked to others about politics were more likely to be political participants than those who did not so, supporting past studies that found the important role of political discussion in facilitating political participation.40

This study reveals that blog reliance has no influence on political activity. Despite the growing use of blogs for political purposes, whether blogosphere can be “a democratic discussion forum” still remains to be further investigated. That is, the greater control over information selection and choices on the Internet could result in the fragmentation of online news audiences, indicating that people may only purse news topics that interest them.41 Davis and Owen6 also pointed out that people online tend to meet with like-minded other people on a narrow range of topics, due to the audience fragmentation. Blogs may offer a polarized and fragmented space where people have little contact with others with different viewpoints.42

In this vein, it should be noticed that most blog content represents the bloggers’ political or ideological leaning.43 The “blogrolling” may attract people of the same ideological perspectives and make them follow the same political party lines, instead of a variety of political viewpoints. Therefore, blogs seems to foster a sense of community through interaction with link-minded people and thus make them feel involved in their political community.44

In this regard, Herring and associates45 pointed out that despite the potential of blog for interconnected conversation mainly made possible through hyperlinks, the conversations taking place on blogosphere are partially interconnected and sporadic. They examined the nature and degree of interconnectedness of the blogosphere with a sample of more than 5,000 blogs and concluded:

The blogosphere appears to be selectively interconnected, with dense clusters in parts, and blogs minimally connected in local neighborhoods, or free-floating individually, constituting the majority. Moreover, it seems likely that the much-touted textual conversation that all of the blogosphere is supposed to be engaged in involves a minority of blogs as well, and sporadic activity even among those blogs46

Given that blogs are still in an early stage of adoption, it is still difficult to explicate the political roles or effects of blogs. Indeed, past uses and gratification studies about blogs found that people still access blogs to get information they want because it is easy and convenient, while social utility is a weaker motive.47

As mentioned previously, the result of this study cannot be generalized to the whole population because it depends on an online survey designed to attract politically interest Internet users. It is very difficult to draw a random sample from the Internet population because there is no registry system on the Web. Therefore, this study relied on a purposive sample, which has become an acceptable way to get a specific group of the online population. Future study should be taken with a larger and more comprehensive sample.

In addition, as Scheufele48 put it, social utility needs for using the Internet or its applications may have some effects that moderate the political impact of traditional or online news media on political participation. Therefore, it may be worth examining if social communication motives for the Internet and its interactive components would facilitate a person’s discussion with others about politics, which in turn results in high levels of political participation.

Daekyung Kim as an assistant professor of digital media at Idaho State. Thomas Johnson is the Amos Carter Centennial Professor, 
School of Journalism, University of Texas, Austin.. Barbara Kaye is an associate professor of journalism and electronic media at Tennessee-Knoxville.