Eric Riedel
Center for Applied Research and Educational Improvement
University of Minnesota
Amy Gangl
Department of Political Science
Union College
Alina Oxendine
Department of Political Science
University of Minnesota
Melinda Jackson
Department of Political Science
University of Minnesota
John L. Sullivan
Department of Political Science
University of Minnesota
Eugene Borgida
Department of Psychology
University of Minnesota
This research was supported in part by grants from the Center for Urban and Regional Affairs at the University of Minnesota, NSF Grant no. SBR9619147 and funding from the University of Minnesota's College of Liberal Arts to Eugene Borgida and John L. Sullivan. Direct correspondence to Eric Riedel, Center for Applied Research and Educational Improvement, University of Minnesota, 270 Peik Hall, 159 Pillsbury Drive S.E., Minneapolis, MN 55455-0208, email: riedel@tc.umn.edu.
Abstract
Previous research has consistently shown print media news sources to have an important impact on political knowledge. During the past decade, use of the Internet as a news source has increased, aided in part by the willingness of print news sources to publish online versions of their publications. This study compares the effects of the Internet on political knowledge with other print news sources using a pooled sample from repeated surveys of two communities. Controlling for other sources of political knowledge and selection bias problems, Internet news sources predict national but not local political knowledge and appear to supplement national news media in general. The lack of impact on local political knowledge is attenuated by how broadly news sources are conceived.
Introduction
Where do people gain political knowledge, in an information environment with numerous traditional and technological options? That newspapers serve as the primary medium through which people obtain political information is a longstanding finding in research examining the relationship between Americans' media use and their political knowledge (Clarke & Fredin, 1978; Delli Carpini & Keeter, 1995, Robinson & Levy, 1996; Wagner, 1983). Even after controlling for the dominant demographic predictors -- education and income (Delli Carpini & Keeter, 1996) -- print newspapers are time and again shown to be an important component of political knowledge (Graber, 1987; Robinson & Davis, 1990). Specifically, attention to local newspapers significantly predicts local political knowledge and reading a national newspaper on a regular basis significantly predicts national political knowledge.
Public perceptions of the sources of political knowledge, however, are at odds with these findings. When asked where they get their political information, surveys suggest that most people believe television to be their primary source. Given the increase in the number of evening news programs on the major networks in recent years and the decline in daily readership of newspapers, it is not surprising that many people think that television provides them with most of their political information (Ansolabehere, Behr, & Iyengar, 1993; Robinson & Levy, 1996). Although reliance on television news has surpassed print news media in overall usage, evidence suggests it is not as effective a means of political learning as print media. Whereas time devoted to reading one or more newspapers on a regular basis continues to be a key component of political knowledge, hours logged in front of the television does not, regardless of the content of the programs people watch (Robinson & Davis, 1990; Robinson & Levy, 1996).
The emergence of the Internet and its potentially dramatic effect on the information environment raises questions about the role of newspapers as a political educator. More than 5,400 newspapers around the world are published on the Internet today, according to the American Journalism Review NewsLink. In the U.S., more than 3,400 general circulation newspapers are currently online. Only a very small fraction of the publishers charge online readers subscription fees for access to their daily editions. Almost every major national newspaper (New York Times, Washington Post, Chicago Tribune, Los Angeles Times, etc.) provides free access to their online versions, which are updated on often an hourly basis. During the 2000 presidential election season, the Washington Post and the New York Times provided virtual real-time coverage of the primary and general election campaign. Hourly updates of the candidates' campaign speeches and rallies, as well as the results of new public opinion polls, could often be found first on these national newspapers' Web sites, and, then, in their print versions on the following day.
In addition to complimenting existing media, the Internet also offers direct access to primary sources of information, allowing citizens to bypass print and television media altogether. These sources include "newsgroups" and chat rooms, where the politically engaged offer their own political analysis (Hill & Hughes, 1998). They also offer information on political campaigns (Selknow, 1998) and government agencies' Web sites (Groper, 1996). All of these sources offer an interactive aspect not found in traditional media.
Clearly there are many sources of political news on the Internet, and the demand is equally as great. According to a survey conducted in March 2000 by the Pew Internet and American Life Project, of those Americans with Internet access, 60 percent have gone online to obtain news, and 22 percent do so daily. A more recent poll by the same organization revealed that accessing news online increased in the wake of the Sept. 11, 2001 terrorist attacks with 27 percent of those with Internet access going online for news daily following September 11. In a different study of Internet usage conducted by the Stanford Institute for the Quantitative Study of Society, Nie and Erbring (2002) conclude that people who regularly use the Internet spend less time with print newspapers and magazines. The authors of the study speculate that many are seeking out newspapers on the Internet instead of turning to more traditional media.
Other studies suggest that Internet news sources supplement rather than substitute for traditional media. Althaus and Tewksbury (2000), based on a study of undergraduates, found that Internet news sources supplement print media and they appear to substitute for television entertainment. The authors concluded that students seek out news on the Internet in order to pursue specific news interests more fully. Hill and Hughes (1998) reached similar conclusions based on a 1996 national survey conducted by the Pew Research Center for the People and the Press. They argue that, "Use of the Internet is often stimulated by news stories on TV or in newspapers, prompting activists to look for more detailed information" (p. 43).
The Internet as a news source poses a number of questions regarding its place as political educator alongside more traditional news sources. By comparing the public's faith in television news with mixed evidence of its effectiveness in facilitating learning about politics, we should not assume that sheer use of the Internet as a news source automatically leads to increased political knowledge.
In the study presented here, we test the proposition that news sources on the Internet are linked to learning about politics, as indicated by increased levels of political knowledge. Specifically, we pose three hypotheses describing the relationship of the Internet to political knowledge. Hypothesis 1, consistent with Delli Carpini and Keeter's (1996) finding that national and local political knowledge cover different domains, national and local sources of news will also fall into different patterns of usage. That is, users of one local news source will also tend to be users of another local news source and vice versa for national news sources, but the relationship between local and national news source use will be much weaker. Hypothesis 2, Internet news sources will tend to be associated with more national news sources than with local news sources. Hypothesis 3, controlling for relevant demographic and political characteristics, use of Internet news sources will predict gains in national political knowledge but not local political knowledge.
The second and third hypotheses, that Internet news sources are primarily national in orientation and predict national political knowledge but not local political knowledge, are rooted in both practical and psychological considerations regarding Internet use. Practically, the availability of local news online pales in comparison to the amount of national news that is on the Internet. Newspapers on the Internet, even metro-based dailies, are much more likely to focus and prioritize issues of national importance on their Web sites than stories about the local city council or community school district.
From a psychological perspective, we believe that most people actively and intentionally make choices about the kinds of media they use to find news and information about different topics. Although early research on media use assumed that audiences were passive consumers of information, more recent work indicates that people are active and selective in their choice of media (Levy & Windahl, 1984; Perse, 1990). Williams, Rice and Rogers (1988) find that people make decisions about the appropriate media channel to use in terms of advantages relative to other forms of media. Similarly, Katz (1974) finds that specific needs and purposes determine the media choices people make.
We argue that when people go online to access Internet news, they are more likely to both seek out and find information that is more national or international in nature than what they could find in their local newspapers. As access and familiarity with the Internet have grown over the last five years, Internet users have become aware of the competing advantages and disadvantages of this medium relative to print news. This argument is consistent with the uses and gratification perspective in mass communication research (Levy & Windahl, 1984, Perse, 1990), which finds that the more familiar people become with different media channels, the more likely they are to weigh the advantages and disadvantages of each.
Our third hypothesis suggests that people who regularly seek out information on the Internet are likely to be aware of the greater availability of national news on the Internet as well as the multiple Internet sources they can access any time with little or no cost. Additionally, there is more in-depth coverage of national news on the Internet than can often be found in print newspapers or magazines. Hypertext links to in-depth and related information items are embedded within Internet news stories, potentially enhancing both the amount of information attainable and the psychological engagement people develop with an issue, political controversy, candidate, etc. In fact, daily print newspapers increasingly reference their Web sites for readers who want more information on the stories that are covered. Moreover, online news sources provide archives for past stories that are often referenced in contemporary articles, or can be accessed through a generalized search of the site for interested readers.
Methods
In order to investigate the potential of the Internet to affect local and national political knowledge, we drew upon survey data collected in 1997 and 1999 from respondents in two Minnesota communities. The surveys were administered as part of an ongoing longitudinal study of the social and political effects of new communications technologies funded by the National Science Foundation (Borgida et al., 2002; Riedel et al., 1998; Sullivan et al., 2002a; Sullivan et al., 2002b). The two communities of Grand Rapids and Detroit Lakes, in Itasca and Becker counties respectively, were selected for demographic similarities following a cluster analysis of all Minnesota counties.
The first survey was conducted in the fall of 1997 with residents from Grand Rapids and Detroit Lakes. One thousand surveys were mailed to residents in each community. The sample for each community was randomly selected from phone book listings (40%) and voter registration records (60%) to balance socioeconomic biases associated with each source. The response rate was approximately 40 percent for each community (Grand Rapids n = 404, Detroit Lakes n = 401).
A second set of surveys was administered to an expanded sample in 1999. This sample was drawn in a similar fashion to the first (40% phone book, 60% voter registration records). The sample was also expanded to include residents outside of Grand Rapids and Detroit Lakes but within Itasca and Becker Counties respectively. Mail-in surveys were sent to 500 respondents each from Grand Rapids, Itasca County, Detroit Lakes, and Becker County. The response rate ranged from 52 percent in Itasca County to 60 percent from Detroit Lakes (n = 269 Grand Rapids; n = 261 Itasca County; n = 301 Detroit Lakes; n = 272 Becker County). Respondents from the 1997 sample who were selected for the 1999 sample were removed and replaced prior to administering the survey. The total sample was 1,870 respondents from both surveys.
The surveys asked respondents about technology ownership and use, attitudes towards computer use, attitudes towards their community, social and political behaviors, and basic demographic questions (see Appendix A for question wording). National political knowledge was measured by a five-item scale recommended by Delli Carpini and Keeter (1996) as a proxy for more in-depth measures of political knowledge (Cronbach 
reliability = .73). Local political knowledge was originally assessed by three items which asked respondents to name the school superintendent, one county commissioner, and the length of a city council member's term ( = .52). The low reliability score on local political knowledge increases likelihood of a Type II error when used as a dependent variable. Thus, the analysis of the predictors of local political knowledge may be overly conservative in its assessment of the impact (or lack thereof) of Internet news sources. However, further analysis into specific uses of the Internet may attenuate concerns about this limitation.
The two counties were initially selected for their demographic and geographic similarities. Appendix B compares the six sub-samples (by year and community).
The differences present were small and none of the differences persisted in both years among the same sub-samples. The following analyses were originally separated out between 1997 and 1999 data, but the results were nearly identical across years. Based on the initial cluster analysis and the relative lack of strong differences observed on the survey between the two communities and two survey administrations, we feel confident in combining the samples to increase the number of respondents in the analyses presented below.
Using the pooled data, the average respondent had some college education, had a modest annual income (median = $15,000 - $34,999), and most were female (57%). Ownership of a computer modem or Internet access was modest (Grand Rapids: 20%; Detroit Lakes 25%) but jumped significantly by 1999 (Grand Rapids: 32%; Detroit Lakes 37%; Itasca County 41%; Becker County 38%). The difference was significant based on a chi-square comparing rates of access between 1997 and 1999 and pooling the two communities each year (x2 = 49.77, p < .001).
Results
The surveys asked respondents to report whether they regularly used a number of print news sources and/or the Internet as a news source. We first assessed how many respondents indicated that they regularly used the Internet as a news source compared with traditional print newspapers and magazines. As Table 1 shows, the Internet remains one of the least utilized news sources, with 13.7 percent of respondents reporting regular Internet use for news in the pooled sample. Only use of national newspapers, generally not available on a daily basis in the two communities, ranks lower (10.7%). By contrast, large majorities of respondents reported regular use of local (91.4%) and regional newspapers (64.7%) in each community. National news sources in the form of newspapers issuing from the Twin Cities (28.0%) or national news magazines (22.5%) rank substantially lower but still were mentioned by more respondents than Internet news sources.
| News Source | Percent Indicating Use (n) |
| Local newspapers | 91.4% (1710) |
| Regional newspapers | 64.7 (1209) |
| Twin Cities newspapers | 28.0 (524) |
| National newspapers | 10.7 (200) |
| National news magazines | 22.5 (420) |
| Internet | 13.7 (256) |
The news sources, as presented on the survey, ranged in regional coverage from local to national scope. This diversity provides a ready opportunity to test Hypothesis 1 concerning whether respondents make a distinction between local and national political news media. In order to see whether our data reflect a similar distinction, a principal components analysis with oblique rotation was performed on the six media items for each sample. The results, displayed in Table 2, show that two components of local and national news media capture approximately 43 percent of the variance among the six news sources as is predicted by the first hypothesis. Twin Cities newspapers, national newspapers, national news magazines, and Internet news sources each fall along the first component of national news media. Local and regional newspapers load on the second component of local news media. Use of the Internet as a news medium loads distinctly on the national news media component, consistent with Hypothesis 2.
(oblique rotation)
| First Component (eigenvalue=1.440) | Second Component (eigenvalue=1.137) |
|
| Local newspaper | -.066 | .699 |
| Regional newspaper | .092 | .690 |
| Twin Cities newspaper | .642 | -.030 |
| National newspaper | .645 | -.021 |
| National news magazine | .618 | .244 |
| Internet news | .466 | -.335 |
Given that the Internet appears to be related to other national news sources, we explored the extent to which individuals relied on it exclusively for national news. Did the use of the Internet substitute or supplement other sources of national news? Table 3 shows the patterns of media use for respondents in each sample, comparing those who indicated using the Internet as a news source with those who did not. Local news media includes local and regional newspapers while national news media includes Twin Cities newspapers, national newspapers, and national news magazines. Those respondents who reported using the Internet as a news source used local news sources slightly less frequently (91% used local news sources) than those who did not use the Internet as a news source (96% used local news sources). This difference is statistically significant based on binomial test. The difference in use of national news sources is more pronounced with 59% of those using Internet news sources reporting also using national news sources versus 42% of those using Internet news sources. These patterns suggest that those who use the Internet for news are also more likely to utilize other print news sources for national news, supporting a view of the Internet as a national news supplement.
Sources by Use of Internet as a News Source
| No | Yes | |
| Local News Sources | 96% ** | 91% ** |
| National News Sources | 42% ** | 59% ** |
| Local + National News Sources | 41% ** | 56% ** |
* p < .05, ** p < .01 based on binomial test.
Turning to examine Hypothesis 3 and the impact that the Internet as a news source may have on political knowledge, we face two potential problems. First, some of the same indicators that predict political knowledge, such as income and gender, are also related to computer use, though for somewhat different reasons. This may result in a selection bias problem in modeling political knowledge. The potential danger from this problem is biased estimates of treatment effects, which is the Internet as a news source in this case (Achen, 1986). The second problem is that of comparing the Internet as a news source to other print news sources. With the exceptions of extreme poverty and illiteracy (generally mitigated in a mail-in survey), there are few barriers to newspaper readership aside from those of interest and motivation. The same is not true for the use of the Internet as a news source. Individuals must not only possess the interest and motivation to pay attention to the news, as with any other news source, but they must have regular and convenient access to the Internet.
To alleviate both of these potential problems in investigating the effects of the Internet compared with other news sources on political knowledge, we model political knowledge using a sub-sample of respondents who reported owning a computer modem or Internet access. This limits our sample to 582 respondents of which 36.3% reported using the Internet as a news source.
Tables 4 and 5 present the results of multivariate regression models predicting local and national political knowledge. In each model we control for individual characteristics found to consistently predict political knowledge by Delli Carpini and Keeter (1996). These include respondent's age, gender, formal education, annual income, sense of political efficacy, and report of whether he or she talks about politics. In addition, six dummy variables representing the six news sources examined above are included.
Local Political Knowledge
| Unstandardized Coefficient (Standard Error) |  | Observed Power | |
| Constant | -.468 (.373) | .003 | .241 |
| Age | .014 (.004) ** | .030 | .975 |
| Annual Income | .105 (.059) | .006 | .423 |
| Education Level | .116 (.043) ** | .014 | .761 |
| Political Efficacy Scale | .027 (.034) | .001 | .122 |
| Talk About Politics? (1=Yes) | .131 (.156) | .001 | .133 |
| Gender (2=Male) | -.169 (.090) | .007 | .467 |
| Local News | .415 (.138) ** | .017 | .849 |
| Regional News | .147 (.090) | .005 | .368 |
| Twin City News | .026 (.094) | .000 | .059 |
| National Newspaper | -.149 (.134) | .002 | .200 |
| National News Magazine | .120 (.104) | .003 | .211 |
| Internet News | .060 (.092) | .001 | .100 |
| Adjusted R Square | .081 | ------ | ------ |
| F (12, 506) | 4.827 (.000) | ------ | ------ |
* p < .05, ** p < .01
National Political Knowledge
| Unstandardized Coefficient (Standard Error) | | Observed Power | |
| Constant | .567 (.416) | .004 | .275 |
| Age | .020 (.004) ** | .050 | .999 |
| Annual Income | .089 (.066) | .004 | .269 |
| Education Level | .191 (.048) ** | .030 | .977 |
| Political Efficacy Scale | .052 (.038) | .004 | .277 |
| Talk About Politics? (1=Yes) | .803 (.174) ** | .040 | .996 |
| Gender (2=Male) | .132 (.100) | .003 | .260 |
| Local News | -.026 (.154) | .000 | .053 |
| Regional News | .183 (.101) | .006 | .440 |
| Twin City News | -.014 (.104) | .000 | .052 |
| National Newspaper | -.077 (.149) | .001 | .081 |
| National News Magazine | -.005 (.115) | .000 | .050 |
| Internet News | .264 (.102) * | .013 | .735 |
| Adjusted R Square | .157 | -------- | -------- |
| F (12, 506) | 9.051 (.000) | -------- | -------- |
* p < .05, ** p < .01
Both regression models had modest predictive power with adjusted R squares of .08 and .16 for local and national political knowledge respectively. In both models, age and educational levels were statistically significant predictors (p < .01), both accounting for somewhat more variance in the national political knowledge than local. In addition, talking with others about politics is a significant predictor of national but not local political knowledge. The contributions of the various news sources also confirm our expectations. Use of local newspapers predicts local political knowledge (B = .415) but has no effect on national political knowledge. Use of the Internet as a news source predicts national political knowledge (B = .264) but has no effect on local political knowledge (B = .060). This latter coefficient has a low level of power (.10) which is most likely the result of low reliability of the dependent variable, making detecting a significant effect on local political knowledge unlikely. Given this, we are more tentative in retaining the null hypothesis that Internet news sources have no impact on local political knowledge. These findings were robust. To check assumptions about multicollinearity we replaced the Internet news use variable with a dummy variable representing respondents who used the Internet as a news source but not any other national news source (84 respondents). We also ran the regression models with a dummy variable added representing county. The results in both cases were nearly identical to the regression models presented here with the impact of various news sources replicated and the presence of county providing no additional predictive power.
Earlier we suggested that although there are numerous online versions of print media, the Internet may also expand access to national political information by giving opportunities to learn new information directly from candidates' campaigns and government agency Web sites. Given our tentativeness in accepting the null hypothesis of no effects of Internet news sources on local political knowledge, we sought to explore whether the online sources of local political knowledge should be conceived more broadly. In addition to replicating the 1997 survey questions, the 1999 survey contained a battery asking respondents how often (never, once or twice, sometimes, often) they went online to seek out various topics of information. Table 6 shows the correlations between local and national political knowledge and how often respondents went online to access various types of information. Again, the sample is restricted to respondents who own modems or have Internet access in order to alleviate selection bias problems.
The findings depicted in Table 6 suggest that there are other sources of local political knowledge available online when those sources are understood in a broader context. Many community organizations and institutions maintain web pages, and these correlations suggest that some may contribute to local political knowledge. In particular, seeking out information about schools, through school district Web sites, is modestly associated with increased levels of local political knowledge (r = .20). The results also suggest that Internet users who go after information about national politics have a particular interest in elections as shown by the correlation between national political knowledge and seeking information about campaigns (r = .13).
Accessing Information Online (1999 Sample Only)
| How often have you used a computer to find information about . . . | Local Political Knowledge | National Political Knowledge |
| Political Campaigns | .052 | .130 ** |
| The Community | .113 ** | .021 |
| Distance Learning | .078 | .054 |
| Employment | -.122 ** | -.053 |
| Government | .109 ** | .088 * |
| Health | -.036 | .042 |
| Schools | .200 ** | .068 |
| Social Services | .132 ** | .062 |
* p < .05 two-tailed, ** p < .01 two-tailed
Discussion and Conclusion
This study's results suggest that the long-standing finding that newspapers are the medium that contribute the most to political knowledge may be changing. Among those who have Internet access in these two Minnesota communities, we find that Internet news is a significant predictor of national political knowledge, even while controlling for common correlates of political knowledge and adjusting for potential problems with selection bias. This finding has significant implications for the study of political knowledge and underscores the need for researchers to take into consideration Internet news sources.
Our results reveal that, at of the time of our study, most people use the Internet in conjunction with other national and local news sources. However, we suspect that over time it may become more of a substitute for traditional newspapers and magazines than a supplement. The Internet has grown significantly over the past few years, and the number of people who access and utilize it is on the rise.
Our initial investigation of the predictors of local political knowledge revealed that attention to local print media still predicted levels of political knowledge, whereas Internet sources appeared not to have an effect. Subsequent investigation of the impact of online sources on local and national political knowledge suggested that our measure of Internet news use may conceive of the Internet as a political educator in overly narrow terms. Web sites by local government institutions and community organizations may also serve to educate the public and our results lend modest support to this idea. Thus, our earlier hypothesis that the Internet is a source of national but not local news holds true only when we conceive of Internet news sources in ways that mirror print sources (e.g. online versions of newspapers and magazines).
During the past decade there have been a number of efforts at the local level to implement community electronic networks, often with the explicit goal of enhancing citizen engagement in local affairs (Bryan, Tsagarousianou, & Tambini, 1998; Guthrie & Dutton, 1998; Kavanaugh, Cohill, & Patterson, 2002; Schuler, 1994). Rather than being the purview of a single agency or organization, such networks typically aim to bring together a number of different local entities such as city or county government offices, schools, social service organizations, business organizations, and civic groups. They often attempt to facilitate communication among citizens through sponsoring public access to the network and the Internet as well as structuring online communication. In the process they may accentuate existing ties between citizens and community institutions and even create new ones as citizens learn about others in their community.
This possibility contrasts with a longstanding criticism of the Internet as in many ways, a fragmenting medium. On the Internet, users can actively select stories they want to explore and, with the help of Web links, find extensive information from the past and present about these stories. In this way, people have the opportunity to evaluate topics of choice in depth, but this also makes it easier to circumvent other sections of the news. During the entrance of the Internet as a news medium, Abramson, Arterton, and Orren (1987) speculated on the consequences of the specializing quality of new communications technologies:
. . . Let us simply acknowledge that the national media--broadcast radio and television, the wire services, national newspapers--have been one important source of our common political culture, one that promotes a shared political vocabulary and agenda and in some respects a homogenization of opinion. To the extent that the new media segment the audience geographically and functionally more than it is divided today, that pool of common knowledge and common perspective will be reduced. (p. 113)
Future research needs to investigate further this issue of whether use of the Internet as a source of public information tends to narrow or broaden the scope of an individual's knowledge, or whether both effects may occur under different conditions.
Our findings should be viewed in light of current concerns about the "digital divide," the recurring finding that access to new information technology is differentially distributed by income, education, race, age, and gender. Our research strategy took this explicitly into account by selecting only those respondents who own a modem or otherwise have personal Internet access, recognizing that we may otherwise have a selection bias problem. Our findings suggest a new aspect to the digital divide. Not only do we confirm a well-known finding that political knowledge is stratified by socioeconomic resources, but that an additional avenue to political knowledge, the Internet, stands to compound this gap. Specifically, those with higher income and levels of education are more likely to own computers and have Internet access than those of lower income and education strata (Anderson et al., 1995; McConnaughey & Lader, 1998; Wresch, 1996). Of course, age stands as a singular and strong example of where these trends run contrary to one another. Age is positively linked to political knowledge while negatively linked to Internet use (Laguna & Babcok, 1997). Indeed, the more frequent use of the Internet as a news source by young adults may represent a significant generational effect (Bennett & Rademacher, 1997). It will be important to explore the extent to which people who lack Internet access will be disadvantaged in terms of political knowledge.
Our findings suggest that, in light of the changing nature of technology, it is necessary to rethink the roles of all personal and media sources of political knowledge, and the long-held dominance of newspapers in this process. We find that Internet news sources best explain political knowledge among those who use them and suggest that Internet news may surpass newspapers as the select political information source of the future.
Abramson, J. B., Arterton, C.F., & Orren, G. R. (1987). The electronic commonwealth: The impact of new media technologies on democratic politics. New York: Basic Books.
Achen, C. (1986). The statistical analysis of quasi-experiments. Chicago: University of Chicago Press.
Althaus, S., & Tewksbury, D. (2000). Patterns of Internet and traditional news media use in a networked community. Political Communication, 17, 21-45.
Anderson, R. H., Bikson, T. K., Law, S. A., & Mitchell, B. M. (1995). Universal access to e-mail: Feasibility and societal implications. (Rand Report MR-650-MF). Santa Monica, CA: RAND Corporation.
Ansolabehere, S., Behr, R., & Iyengar, S. (1993). The Media Game: American Politics in the Television Age. New York: Macmillan Publishing Company.
Bennett, S. E., & Rademacher, E. W. (1997). The 'age of indifference' revisited': Patterns of political interest, media exposure, and knowledge among Generation X. In S. C. Craig & Stephen Earl Bennett (Eds.), After the boom: The politics of Generation X (pp. 21-42). Lanham, MD: Rowman & Littlefield Publishers, Inc.
Borgida, E., Sullivan, J. L., Oxendine, A., Jackson, M. S., Riedel, E., & Gangl, A. (2002). Civic culture meets the digital divide: The role of community electronic networks. Journal of Social Issues, 58, 125-142.
Bryan, C., Tsagarousianou, R., & Tambini, D. (1998). Electronic democracy and the civic networking movement in context. In R. Tsagarousianou, D. Tambini, & C. Bryan (Eds.), Cyberdemocracy: Technology, Cities, and Civic Networks (pp. 1-17). London: Routledge.
Clarke, P., & Fredin, E. (1978). Newspapers, television, and political reasoning. Public Opinion Quarterly, 42, 143-160.
Delli Carpini, M. X., & Keeter, S. (1995). What Americans know about politics and why it matters. New Haven, Conn: Yale University Press.
Graber, D. (1987). Processing the news (2nd ed.) New York: Longman.
Groper, R. (1996). Electronic mail and the reinvigoration of electronic democracy. Social Science Computer Review, 14, 157-168.
Guthrie, K. K., & Dutton, W. H. (1992). The politics of citizen access technology: The development of public information utilities in four cities. Policy Studies Journal, 20, 574-597.
Hill, K. A., & Hughes, J. E. (1998). Cyberpolitics: Citizen activism in the age of the Internet. Lanham, MD: Rowman & Littlefield Publishers, Inc.
Katz, E., Blumber, J.G., and Gurevitch, M. (1974). Utilization of mass communication by the individual. In J. G. Blumler & E. Katz (Eds.), The uses of mass communication (pp. 19-32). Beverly Hills, CA: Sage.
Kavanaugh, A. L., Cohill, A. M., & Patterson, S. (2002, 30 September). Use and impact of community networking in Blacksburg Electronic Village 1994-1997. Virginia Polytechnic Institute and State University. Retrieved 30 September 2002 from http://www.bev.net/project/research/index.html#highlights.
Laguna, K., & Babcock, R. L. (1997). Computer anxiety in young and older adults: Implications for human-computer interactions in older populations. Computers in Human Behavior, 13, 317-326.
Levy, M., & Windahl, S. (1984). Audience activity and gratifications: A conceptual clarification and exploration. Communication Research, 11, 51-78.
McConnaughey, J. W. & Lader, W. (2002, 30 September). Falling through the net II: New data on the digital divide. National Telecommunications and Information Administration. Retrieved 30 September 20002 from http://www.ntia.doc.gov/ntiahome/net2/falling.html.
Nie, Norman H., & Erbring, L. (2002, 30 September). Internet and society: A preliminary report. Stanford Institute for the Quantitative Study of Society. Retrieved 30 September 2002 from http://www.stanford.edu.group/siqss.
Perse, E. (1990). Audience selectivity and involvement in the newer media environment. Communication Research, 17, 675-697.
Pew Internet and American Life Project. (2002, 30 September). Tracking online life: How women use the Internet to cultivate relationships with family and friends. Retrieved 30 September 2002 from http://www.pewinternet.org/reports
Pew Internet and American Life Project. (2002). The commons of the tragedy: How the Internet was used by millions after the terror attacks to grieve, console, share news, and debate the country's response. Retrieved 30 September 2002 from http://www.pewinternet.org/reports.
Riedel, E., Dresel, L., Wagoner, M. J., Sullivan, J. L., & Borgida, E. (1998). Electronic communities: Assessing equality of access in a rural Minnesota community. Social Science Computer Review, 16, 370-390.
Robinson, J., & Davis, D. K. (1990). Television news and the informed public: An information processing approach. Journal of Communication, 40, 106-119.
Robinson, J., & Levy, M. (1986). The main source: Learning from television news. Beverly Hills, CA: Sage.
Robinson, J. P., & Levy, M. R. (1996). News media use and the informed public: A 1990s update. Journal of Communication, 42, 129-135.
Schuler, D. (1994). Community networks: Building a new participatory medium. Communications of the ACM, 37, 39-51.
Selknow, G. W. (1998). Electronic whistle-stops: The impact of the Internet on American politics. Westport, Conn.: Praeger.
Sullivan, J. L., Borgida, E., Jackson, M. S., Riedel, E., Oxendine, A., & Gangl, A. (2002). Social capital and community electronic networks. American Behavioral Scientist, 45, 868-886.
Sullivan, J. L., Borgida, E., Jackson, M. J., Riedel, E., & Oxendine, A. R. (2002). A tale of two towns: Assessing the role of political resources in a community electronic network. Political Behavior, 24, 55-84.
Wagner, J. (1983). Media do make the difference: The differential impact of mass media in the 1976 presidential race. American Journal of Political Science, 27, 407-430.
Williams, F., Rice, R. E., & Rogers, E. M. (1998). Research methods and the new media. NY: Free Press.
Wresch, W. (1996). Disconnected: Haves and have-nots in the information age. New Brunswick, NJ: Rutgers University Press.
Political Efficacy Scale ( = .58)
1=None, 2=Very Little, 3=Some, 4=A lot
- How much influence do you think someone like you can have over local government decisions?
- How much influence do you think someone like you can have over national government decisions?
Please check each of the types of news sources you read regularly.
- Local newspapers (Grand Rapids: e.g. Itasca Herald Review; Detroit Lakes: e.g. Detroit Lake Tribune or Becker County Record)
- Regional newspapers (Grand Rapids: Duluth News-Tribune; Detroit Lakes: e.g. Fargo Forum)
- Twin Cities newspapers (e.g. Minneapolis Star-Tribune)
- National newspapers (e.g. USA Today)
- National news magazines (e.g. Newsweek)
- Internet (e.g. Web page news sources)
= .73)
- Which party currently has the most members in the House of Representatives in Congress?
- Who is the current vice-president of the United States?
- Which political party is more conservative at the national level?
- How much of a majority is required for the U.S. Senate and House to override a presidential veto?
- Whose responsibility is it to determine if a law is constitutional or not . . . is it the President's, Congress's, or the Supreme Court's?
= .52)
- What is the name of the superintendent of schools in Grand Rapids (Detroit Lakes)?
- How many years is the term of a Grand Rapids (Detroit Lakes) city council member?
- Please name one current Itasca (Becker) County commissioner.
How often have you used a computer to find information about . . . political campaigns, the community, distance learning, employment, government, health, schools, social services. (Coded 1=Never, 2=Once or Twice, 3=Sometimes, 4=Often)
Demographic Characteristics
What is your approximate annual household income? (Coded 1 = Less than $5,000; 2 = $5,000 - $14,999; 3 = $15,000 - $34,999; 4 = $35,000 - $74,999; 5 = $75,000 or more.)
Please indicate your highest level of education: (Coded 1 = grade school; 2 = some high school; 3 = high school graduate; 4 = some college; 5 = college graduate; 6 = post- college.)
| Grand Rapids 1997 | Detroit Lakes 1997 | Grand Rapids 1999 | Detroit Lakes 1999 | Itasca County 1999 | Becker County 1999 |
|
| Political efficacy scale | 4.44 | 4.54 | 4.51 | 4.53 | 4.57 | 4.58 |
| Age * | 53 | 55 | 60 | 57 | 54 | 53 |
| Income * | 3.57 | 3.49 | 3.45 | 3.42 | 3.60 | 3.35 |
| Education * | 4.17 | 4.11 | 4.00 | 3.95 | 3.96 | 3.87 |
| Female * | 62% | 61% | 54% | 54% | 54% | 49% |
| Talk about politics * | 83% | 83% | 88% | 90% | 93% | 89% |
| National political knowledge scale | 3.67 | 3.85 | 3.82 | 3.69 | 3.79 | 3.63 |
| Local political knowledge scale * | 1.55 | 1.80 | 1.46 | 1.43 | 1.29 | 1.26 |
| Modem or internet ownership | 20% | 25% | 32% | 37% | 41% | 38% |
| Local newspaper * | 92% | 93% | 93% | 92% | 88% | 90% |
| Regional newspaper | 65% | 67% | 68% | 65% | 58% | 64% |
| Twin Cities newspaper | 30% | 29% | 23% | 30% | 31% | 24% |
| National newspaper | 10% | 14% | 7% | 10% | 9% | 13% |
| National news magazine | 26% | 25% | 22% | 21% | 22% | 17% |
| Internet news source * | 11% | 10% | 15% | 14% | 20% | 16% |
* Indicates a statistically significant difference in proportions among sub-samples for dichotomous indicators (chi-square) or difference in means among sub-samples for other indicators (ANOVA).
In citing this document please use the following style:
Riedel, E., Gangl, A., Oxendine, A., Jackson, M., Sullivan, J. L., & Borgida, E. (2003). The role of the Internet in national and local news media use. Journal of Online Behavior, 1 (3). Retrieved
