Show Less
Open access

E-Political Socialization, the Press and Politics

The Media and Government in the USA, Europe and China

Series:

Edited By Christ´l De Landtsheer, Russell Farnen and Daniel B. German

This book examines the state of print and electronic media in the United States of America, Europe, and China. The latest mass communication advances demonstrate that we live in an increasingly media-centric world. The chapters include theoretical and empirical studies that shed light on the meaning of this development. The trajectory of people’s move to electronic communication is a global phenomenon affecting their daily life. Does this trend aid or impede democracy? Is there an emerging digital divide contributing to an increasing gap between the rich and poor people and nations? The four parts of this book explore various aspects of political socialization and its relationship with different media, including print, broadcasting, and the Internet.
Show Summary Details
Open access

8 The Internet Upholds the Powers That Be

| 157 →

Chapter 8

The Internet Upholds the Powers That Be

Henk Dekker and Arie in ’t Veld

Leiden University, Leiden, the Netherlands

Abstract

Information and Communication Technology (ICT) innovations and the Internet in particular transform society and politics a good deal. ICT and the Internet also change political socialization because political elites and middle-level political actors use the Internet for political information, communication, persuasion, and mobilization. These political Internet messages reach a growing number of citizens who rate the credibility of Internet information quite high. They also consider the Internet a convenient, attractive channel for political information. Whether political websites achieve their desired effects cannot easily be assessed. We invited young people to participate (under clinical laboratory conditions) in a 40-minute political party website surfing/browsing session. Time spent on the sites is considered an important condition to measure the influence of political party sites. Our goal was to explain observed variance in the amount of time our subjects spent on political party websites. The independent variable was the variance in sites’ quality. All 10 quality indicators refer to two motivation stimuli: self-efficacy and curiosity. To avoid distortions in our analyses due to unequal distribution of party preferences, we weighted the data for party preference. The hypothesis that the higher the site quality, the longer the site is used could not be falsified: users respond positively to higher-quality sites.

Political Socialization

Political socialization is the whole of those processes and structures through which people develop particular political behaviors and acquire particular political orientations, including political behavioral intentions, emotions, values, attitudes, opinions, beliefs or perceptions, and knowledge. “Processes” include the ways in (and the conditions under) which people receive, process, and more or less accept political messages. “Structures” include the sources/channels and contents of their informative and affective political messages. “Through” means influence. Political socialization research aims to answer the question: What are the origins of individuals’ political behaviors and orientations (i.e., how and due to the influence of whom or what do people develop and perform political behaviors and acquire their personal political orientations)? Political socialization research helps to explain the establishment, maintenance, or change of polities, policies, and politics. Political systems, policies, and political processes are influenced by and are dependent on the acquisition, continuity, or change in the political behaviors and orientations of their members, subjects, and participants. These are, themselves, mainly the effects of political socialization processes (Dekker, 199la, b).

The so-called intellectual elites may have an independent, critical contribution to political socialization, but they may also help the political elites legitimize their ← 157 | 158 → personal preferences and policies. The various individual agencies and the socializers within them possess a relative autonomy. To understand individuals’ socialization, we must study all possible socialization agencies and socializers active therein (Farnen, 1993; Washburn, 1994; Niemi and Hepburn, 1995; Sigel, 1995; Farnen, et al., 1996, 2000; Conway, 2000).

Theoretically, the most influential messengers for information, feelings, and emotions are the ones who first exert influence on the subject with respect to the object under investigation (e.g., grandparents, parents, peers, television programs), who exert influence for the longest period of time (e.g., parents, best friend, spouse), whose credibility the subject believes to be the highest (e.g., parents, teachers, television news), who have the most power over the subject (e.g., parents, teachers, spouse, employer), who have most power to prevent oppositional influences from other socializers (e.g., parents, elites), and who have the most resources and skills to influence and manipulate the subject (e.g., elites in cooperation with public relations and political advertising and marketing experts).

The new Information and Communication Technologies (ICT) and the Internet in particular have changed politics and society a great deal (Hill and Hughes, 1997; Holmes, 1997; Street, 1997; Hague and Loader, 1999; Barney, 2000). One of the benefits over traditional media is that digital hypermedia allow for a greater volume of information to be transmitted across space at faster speeds; the Internet practically eliminates the barriers of time and place. Another of its benefits is the integration of audio, video, graphic, and textual information, including datasets, interactive graphs, and movies. Another important difference is that the web is a solicited method of communication. The targeting is automatic because those customers interested in political information will visit the site. E-mail allows political elites to transmit a great volume of information across space at faster speeds to multiple recipients with minimal costs. Have ICT and the Internet also changed political socialization? Does the Internet have an influence on users’ political behaviors and orientations? There can only be a political socialization “influence” when there is a political message, when that message reaches the individual, and when that individual is receptive to the message. We explore this topic next.

The Internet and Politics

ICT, the Internet, and the World Wide Web particularly enjoy a growing interest among political elites and middle-level political actors (Barber, et al., 1997; Connell, 1997; Horrocks, et al., 2000). In many countries, the head of state, government, ministries, parliament, individual members of parliament, political parties, interest groups, and (opposition) socio-political movements have one or more websites. All urge citizens to visit them on these sites. Norris (2000) used common search engines to gage the amount of politics occurring on the Internet and monitored the frequency that eight common keywords are found. The terms ← 158 | 159 → “computers,” “sex,” and “television” proved the most popular keywords. “Politics” came next, slightly outweighing “movies,” “religion,” and “investing.” One in l0 sites referred to “politics.” This probably represents a conservative estimate of political sites because many of them are indexed under other terms such as government, parliament, elections, political parties, and interest groups,

Heads of state and governments/administrations use the Internet as an additional agency of political information, public relations, and public diplomacy. Web-enabled government substantively changes public administration across all advanced industrial countries (Prins, 2001). Dunleavy and Margetts (2000) state that informatization of government and public administration makes central control and manipulation of populations potentially much easier (Macpherson, 1998).

Parliaments also distribute many different types of information directly and simultaneously: the daily agenda for parliamentary business, complete versions of official documents (such as the full text of pending legislation and government reports), streaming audio-visual feeds of debates in the legislative chamber, and political education spreads for teachers and students. Members of Parliament have a home page and e-mail address to “keep in touch,” “meet,” and influence their voters and possible future supporters. As of April 1, 2000, parliaments in 101 countries have established their presence on the Web (57% of the total number of parliaments). While Europe leads the way with 87% of its national parliaments operating websites, Africa lags behind with a mere 33%. The overall number of parliaments with websites has nearly tripled within two years (Inter-Parliamentary Council, 2000). Critical evaluations of current practices point to their relatively poor quality (e.g., features for searching, feedback pages, MPs’ personal web pages and e-mail contact addresses are missing) (Norris, 2000; Inter-Parliamentary Union, 2000). Moreover, many MPs foolishly do not respond to constituents’ emails (Meeks, 2000).

Internet voting is being integrated into the electoral process. The goal is increasing voter turnout. People will not have to make separate trips to an election site to cast a ballot. This means easier registration, no traveling costs, and less time. Moreover, Internet voting is expected to spark the interest of people who are not so much attracted by politics and bring them into the electoral process. However, there are some serious technical and social problems that need to be solved to ensure the security and reliability of the voting process (Gibson, 2002). The 2000 Arizona Democratic primary election was the first binding Internet election to ever occur. While turnout was small, it was substantial. It demonstrated an increase of 575% over 1996; Arizona had the highest percentage increase in turnout in Democratic primary elections. Voting methods (preference for Internet voting versus voting through mail in advance or via machine or paper balloting on election day) was moderately associated with education, income, and age; Internet voting is more popular among well-educated voters, those coming from higher income households, and younger voters. Ideology, “race,” gender, and location of residence ← 159 | 160 → failed the test of statistical significance. Logistic regression analysis showed that age and education contributed most to the voting method decision, while income was not a significant predictor (Solop, 2000). There is also some speculation that Internet voting may favor US Democrats because Republicans have always had a very high turnout rate, despite the fact that more people are registered as Democrats.

Political parties use the Internet for their intra-party and inter-party communication and their interaction with members and possible future members/voters. Party elites have an additional access to their followers to convince them of particular views. The Internet allows smaller parties more opportunities to get their messages across. They can compete on an equal footing with the major parties to communicate their message to the wider Internet electorate. New members are recruited from categories that are less reached by the traditional media. Parties can invite non-members to take part in Internet discussions to start building a relationship with these possible future members.

Parties have invested heavily in online campaigns. Although accurate figures are not available, it is estimated that the three parties in the UK (by 2001) have put over a million pounds of their funds into websites, e-mail campaigning, and associated new media innovations (Crabtree, 2001). An important element is image marketing and management. Now, leaders like to show off their computer skills to convey an image of a modem, computer-literate leader. Opponents spread online rumors that can keep the accused away from governing; “dirty” Internet tricks and e-“smear” campaigns hurt leaders’ and candidates’ images. The Internet allows candidates for public office and parties to disseminate quickly retrievable and up-to-date information without interference or “mediation” (e.g., by critical journalists). Instead of merely trying to influence, direct, and manipulate the traditional media, political parties now also go directly to voters with their online efforts.

Parties have new opportunities to raise funds for their campaigns. Partisans are invited to actively participate in the campaign via the Internet. Some parties offer voters the opportunity to make “your own party homepage” (e.g., “your own Gore homepage” in 2000). They provide texts on the party and/or candidate; users can choose from links files what they want to send by e-mail to their own separate distribution lists. Friendly partisans are also invited to order campaign materials.

Political party websites share many elements. All three main parties in the 2001 parliamentary election campaign in the UK had a homepage with identifying features; the party manifesto/program; information about the candidates; daily news stories; full versions of speeches; cartoons, screen savers, and posters to be downloaded; calendars of events; invitations to contact and interact with the party via an online e-mail form or e-mail address to join the party and donate money; and a password-protected area for party members (a members-only “extranet”), including downloadable campaign material, graphics, and advice on how to respond to voters’ questions about the day’s events and news (Coleman and Hall, ← 160 | 161 → 2001). There were also clearly identifiable differences. Each site of the three main parties in the UK during the 2001 parliamentary election campaign had one or more unique elements. These included Frequently Asked Questions (FAQs), games, audio/video clips (of press conferences, interviews), live interviews (for which site users could use e-mail for questions), pop-up boxes (a small screen that “floats” over a site page, drawing one’s attention to particular content or to solicit members), and online petitions (Coleman and Hall, 2001). Party sites also differed in sophistication (Earnshaw, 2001) and level of interactivity (e.g., whether the email is answered, answered late, or specifically answered) (Hansard Society, 2001). Differences in the overall quality of party/candidate campaign sites in the US by 1996 and 1998 positively correlated with campaign resource level (Sadow, 2000; Sadow and James, 2001).

Social-political movements (human rights defenders, environmentalists, anti-globalists, but also terrorists, anti-democrats, racists, anti-Semites, neo-Nazis, sexists, and anti-gays) use Internet opportunities to inform and mobilize fellow citizens on a neighborhood, local, regional, national, and even global scale. The Internet gives today’s activists an information-gathering network that their predecessors lacked. “Bearing witness” webcam reports alarm and mobilize their viewers. A “NetStrike” consists of a massive and simultaneous access to the same website until it can no longer bear the demand and becomes unavailable. “Rogue” (joke or cyber-squatter) sites try to undermine the effectiveness of their original sites. An increasingly important (and probably effective) grassroots tool is the email petition. In this case, the organization writes the text to make it easier for people to contact their representatives in parliament and other public officials. There are online grassroots calendars where any political group can post its calendar in a centralized location for free and that notify activists when events are added to calendars which interest them. A new technique is to use e-mail to propose grassroots lobbying efforts which then employ handwritten and mailed previously-prepared text to ensure the message will be properly weighed in the representatives’ office, where a canned e-mail would not be so highly valued.

ICT and the Internet have also affected the traditional socialization agencies such as mass media and school. Many traditional mass media have developed an additional Internet version (e.g., CNN.com). There are new online news“papers” and weeklies and daily political news and commentary sites. ICT and the Internet have also affected schools and political education (Filzmaier, 2001). Internet political education projects aim to narrow the political digital gap and to improve political competence and interest (for examples of projects, see the politeia.net site of the Politeia Network for Citizenship and Democracy in Europe). NGOs and private organizations provide services to voters and aim to help select a party/candidate in elections, to make informed political choices, and to notify them about relatively simple but important practical matters such as when, where, and how to vote. Online voter guides offer information about the various parties and ← 161 | 162 → candidates (e.g., their biography, qualifications for office, policy statements, speeches, and previous performance on fulfilling election promises). Voter guides offer clear voting advice. After answering questions about the relative importance of issues and about the user’s own issue opinions, the computer links the user’s opinions to party programs and informs the user about which party comes closest to his/her opinions (a voter matching site). In the US in 2000, youth were able to participate in a national (non-binding) online election for president (see the youthevote.com site). In Germany, young people could vote for state and federal parliament parties (see the Juniorwahl.de site). In the Netherlands, young people could vote for parties in the national parliament (see the publiek-politiek.nl site).

Since there can only be political socialization “influences” from the Internet when there are political messages on the Internet, when these messages reach the individual, and when that individual is receptive to the messages, we may conclude that the first requirement is met. Political elites and middle-level political actors use the Internet for political information, communication, persuasion, and mobilization. But now we must ask, do these messages reach many individuals and are they receptive to them?

Political Internet Sites and Citizens

How many people have Internet access? Never before did a new communication medium evidence such a fast growing number of users as has the Internet (Bell and Tang, 1998). However, only 7% of the world’s population (mostly in the West) uses the web (Sussman, 2001). In most EU countries, less than half of the population uses the Internet. In Autumn 2000, on average, 26% of EU citizens used e-mail and/or the Internet. Use varies greatly among the 15 EU member states (from 61% in Sweden to 11% in Greece). Managers are most likely to use the Internet and/or e-mail (61%), followed by people who left full-time education at the age of 20 or older, and people who are still studying (both 57%). It is lowest among the elderly; only 4% of retired people use it (European Commission, 2001). There is a digital divide due to the costs of Internet access and/or becoming computer literate in the linguafranca, English. Not all users are frequent visitors. There are more “sensors” (who prefer the real over the virtual) and “hoppers” (who pop in and out of the Internet) than “assimilators” (who absorb the Internet into their lives). “Sensors” fear that the Internet holds its users back from social interaction, do not believe the Internet is the final authority for information, think it is easier to call or go out to shop, are irritated with the abundance of information available online, and do not like junk mail (Lee and Anderson, 2000).

How many citizens see the Internet as an important source of political information? In the US, low (though growing) percentages of Internet users say they view the Internet as very or somewhat important as a source of information ← 162 | 163 → about elections. In the presidential election of 2000, half of the Americans viewed the Internet as very (19%) or somewhat (32%) important as a source of information about the election. The figures among actual Internet users are higher (very = 23%, somewhat = 41%) (American University, 2000). In the UK, many fewer Internet users say they are “certain” or “likely” to use it to find information about the election (respectively, 2% and 13%). More than 8 in 10 express no interest in using the Internet during the election: 39% are “not likely” to use it and 45% will “definitely” not use the Internet (Crabtree, 2001; Industrial Society, 2001a).

How many citizens have in fact accessed the Internet for information about politics? In the US, low (though growing) percentages of Internet users say they have actually accessed the Internet for information about politics. In the 2000 presidential election, nearly 1 in 5 Americans said they went online for “election news,” compared to fewer than 1 in 10 during the 1996 campaign (Pew, 2001). One-third of the Americans who use the Internet said they obtain information about “politics, candidates, or political campaigns” online. Veteran online users (who have been online for at least three years) were far more likely to get election news (45%) than Internet “newbies” (who began going online in the past six months) (17%) (American University, 2000). Two-thirds of the youth ages 12 through 17 have searched for news online (Lenhart, et al., 2001). Interest in US online campaign news peaked around election day; fully 12% of Americans went online for political news on November 7, 2000. That figure rises to 28% among those who voted on November 7 (Pew, 2000a). Users do not spend much time on political websites: in the presidential election year 2000, more than half of US Internet users who accessed information about politics, candidates, or political campaigns have spent no more than one hour doing so during a “typical” week (53%) (American University, 2000). In the UK, another 1 in 5 respondents who had access to the Internet used it for any 2001 election-related activities. Young people were much more likely to use the Internet to find out about politics than older citizens (Coleman and Hall, 2001).

Why do people use the Internet? The Internet’s main appeal is as a campaign news source. It is its convenience that motivates, rather than a desire to tap new or different information sources because users were not getting all the news they wanted from traditional media (respectively, 56% and 29%) (Pew, 2000a). Why do people not use the Internet for political information? A self-evident primary cause is no access to the Internet. Second, people are largely ignorant of political Internet content. Third, people have negative perceptions: they think political information online is aimed at party activists, not at them. Fourth, people do not feel they have the ability to make use of the Internet and its political content. Fifth, people feel they are already overloaded with information on the election. There is no evidence that gender and socio-economic status have an influence on using or not using it. If voters do search for political information, popular expectations are often reversed; they are likely to find the Internet a useful political tool and prefer it to much offline media (Crabtree, 2001). ← 163 | 164 →

What are people doing on and with the Internet when it comes to politics and public affairs? Most political Internet users are likely to use it to get information about various issues and candidates (respectively, 36% and 35%) rather than to communicate or chat with others about issues (19%), forward voting-related information via e-mail (17%), communicate or chat with a candidate (7%), donate money to a candidate (5%), or volunteer for a campaign (5%) (American University, 2000). Another study made clear that when it comes to politics and public affairs, Internet users turn to e-mail more than the web. They also prefer humor to action; more than half of Internet users say they have sent or received e-mail jokes about the candidates or campaign (George Washington University, 2000). Political website users are partisan information seekers and/or issue-oriented seekers, but also uncertain information seekers who follow a visual attraction strategy or a browsing strategy (Kern, et al., 1999). In the UK, voters who consider using e-mail during the election are also likely to discuss the election with friends and family, but avoid parties and politicians. Whereas 32 % said they might send e-mail to friends or family about the election and 34% said they might send a joke, fewer than 10% said they would e-mail a candidate, party, or politician (Crabtree, 2001).

What kinds of websites are used? Focus group participants in the Just, et al. (1997) study visited many sites. They spent much of their time at such sites. In 2000, the American campaign sites were used much less frequently than other media-sponsored sites (Pew, 2001). In the UK, traditional news media sites (such as the BBC website) also are the overwhelming choice for reliable information about the election. Political party websites are significantly less popular; only a third of those who considered using the Internet to find information about the election thought they would visit a national political party site. However voters who do search for political information online favor information on policy presented by parties online to offline (Crabtree, 2001). It is difficult to obtain data about the numbers of visitors to party and candidate websites. Most of the party and candidate sites have removed their counters (which would indicate how many hits the sites were receiving) or have simply hidden or not installed them. The few remaining counters (and anecdotal data from campaign managers) indicate that there were not too many hits overall (Dulio, et al., 1999). One of the reasons for relatively low party website hits is that low (though growing) percentages of Internet users access the Internet for political information in general. Another reason is that the voters habitually prefer mediated content that is mostly provided by traditional media. A third reason is that political party websites are hard to find; the web is “about as well organized as a bookstore after a hurricane” (Toulouse, 1998, p. 3).

How do users evaluate political websites? In general, users are pleased with their political Internet experiences. The reasons that US focus group participants gave for liking their experience with the Internet were that the Internet provides access to a lot of information, the experience is interesting (“exciting,” ← 164 | 165 → “fascinating,” “appealing,” and “fun”), the web is easy to use, it offers diverse views, and it is accessed quickly. The comments that participants gave about the web were that there was not enough information or that it was not current, reliable, or interesting. Inexperienced users found the Internet slow (Just, et al., 1997). In general, Internet users prefer websites of major news organizations as sources of political information. For election news, CNN.com received the highest percentage approval of any website tested; those users rated that site as very useful. Just over one-third of online users who visited the Bush/Cheney and Gore/Lieberman sites (respectively 7% and 6% of online users) found them very useful (Pew, 2001). In the UK, voters who search for political information online seem impressed with the medium and its usefulness. They are likely to be critical of parties and candidates without an Internet presence, are highly demanding in terms of web content and are extremely critical if their demands are not met. These voters express consistent and definite preferences for styles of web presence and content. They have little tolerance for amateurish websites (Crabtree, 2001).

One might think that the low percentages of Internet users who say they are certain they will use the Internet to get election information has something to do with a possible low rating for the credibility of Internet information. However, available data show the opposite. Credibility (believability) ratings for the online sites of the major national news organizations in the US are substantially higher than ratings for the news organizations themselves. The figures for ABCNews.com and ABC News were 44% and 29% respectively, for CBSNews.com and CBS News 41% and 27%, for USAToday.com and USA Today 37% and 21%, and for FoxNews.com and Fox News Channel 34% and 21%, respectively (Pew, 2000a). A possible explanation for this important phenomenon is that people consider information more credible the more they have personally (experienced or) looked for or found it. In Arizona, more than half of adults supported Internet voting being added as an option to all future Arizona elections. Younger people, higher income earners, and more well-educated people expressed stronger levels of support for Internet voting there (Solop, 2000).

How do users estimate the effects of their using political websites? Most focus group participants in the Just, et al. (1997) study believed they learned something about politics from their Internet experience. After people experience using the Internet for voting (e.g., in Arizona), many more Internet voters said they would be more likely to vote if Internet voting was an option in future elections (Solop, 2000). A growing number of US online election news consumers say online election news affected their voting decisions (1996: 31%; 2000: 43%). This subjective effect of online campaign news has been particularly pronounced among young people. Half of the online election news consumers under age 30 say the information they received made them want to vote for or against a particular candidate (Pew, 2000b). However, only 6% of those UK voters who were online said that the Internet was very or fairly important in providing them with ← 165 | 166 → information that helped them determine their vote. More younger (17%) than older people (5%) reported this subjective Internet influence upon their vote (Coleman and Hall, 2001). UK voters who search for political information online are likely to judge political parties by their Internet presence. They felt that their image of the party/candidate with the “best” site had improved their image of the party/candidate. The reverse was also true: bad sites translate to a bad image. Having a “bad” website can be a dangerous liability for parties (Crabtree, 2001).

In short, (almost) a majority of the population in industrialized countries uses the Internet. A small, though growing, minority regularly accesses Internet information about politics. Political website users are partisan and issue-oriented, but also uncertain, information seekers. They prefer websites of major news organizations as sources of political information, compared to political party and campaign sites. Credibility ratings for the online sites of news organizations are substantially higher than ratings for the news organizations themselves. Voters who search for political information online are likely to be critical of parties and candidates without Internet presence, are highly demanding in terms of web content, and are extremely critical if these demands are not met. We may conclude that political Internet messages reach a small, though growing, number of citizens and probably influence these citizens’ political behaviors and orientations. They rate the credibility of Internet information quite high and consider the Internet a convenient and attractive channel of political communications. It is worth our effort to study the Internet in the context of political socialization (Groper, 1996; Ward and Gibson, 1998). Crucial research questions relate to the intended and unintended effects of political Internet activities. How much influence do political websites have on their users? How can we explain variance in this influence?

Political Party Websites and Their Effectiveness

Our empirical study focuses on political party websites. We selected sites of political parties because parties are integral to the operation of political systems. Moreover, elections are the very heart of democracy and voters vote for a party list in many countries; the list system is the most common method applied in proportional representation systems. The elector votes for a party and its list of candidates, rather than for a single candidate. The research project aimed to answer the question: How can we explain observed variance in effectiveness of political party websites?

Whether political party websites are effective and have their desired effects cannot easily be assessed. Behavioral effects are the least complicated to measure (e.g., the number of e-mails sent by users, orders to send a printed party program, and new membership enlistment). Much more difficult to measure is the sites’ influence on political knowledge, beliefs, opinions attitudes, emotions, and behavioral intentions, including party preferences. The main reason for this is that one ← 166 | 167 → cannot easily isolate the site’s effects from the effects from other political socialization agencies.

Our study is based on the assumption that for any effect and influence, the site needs to be used for at least some time. The longer a site is visited, the more stimuli the user probably receives and the higher is the chance that the desired effects are actually reached (Briggs and Hollis, 1997). A relatively long visit time is a necessary, though not a sufficient, condition for such influence. Time spent on the site is just one (albeit important) indicator of success. Other scholars have used the number of pages read as the dependent variable. However, this number is not easy to observe in a reliable way due to the use of “frames” (two pages within one page), the presence of pop-up windows, and other factors.

How do we explain variance in time spent on political party websites? Why is one website used longer than another? We hypothesize that there is a lot to the quality of the site: a higher quality website is used longer.

What makes for high quality? To answer this question, we apply the Hypermedia Interaction Cycle and the Psychological Model of the Internet-User developed by Fredin (l997) and Fredin and David (1998). Here, the key question is: what intrinsically motivates the user to stay at a particular website? What drives people to continue with the next interaction with a website?

The Hypermedia Interaction Cycle model is an iterative, self-regulatory model that captures the dynamics of hypermedia interaction from a user’s perspective. In many ways, it is based on Bandura’s 1989 general social learning theory regarding motivation and self-efficacy. It has two dimensions: cycle stages and motivational components. A cycle is the period between leaving a main menu (e.g., a menu on a home page) and returning to a main menu after a successful or failed search. There are three cycle stages: preparation, exploration, and consolidation. The preparation phase includes making a choice from a menu of options. The user makes an estimate of a good choice, taking into account his/her goals. The exploration phase starts when the choice has been made and the user is presented with a variety of information. The consolidation phase occurs when the user reached his/her goal or failed to do so and has decided to try another option. Then the cycle starts again. The time spent on a website is longer the more cycles are started and the longer the cycles are.

Motivation has two components: goals and self-efficacy. Goals motivate by providing the potential of satisfaction (or dissatisfaction). Goal foreshadowing occurs during the preparation phase of the interaction cycle, while goal evaluation occurs during the consolidation phase. Goal evaluation includes the degree of success in finding something good and the degree of liking what was found. Self-efficacy is the conviction that one can do what is required to accomplish a particular outcome (Bandura, 1989). In the preparation phase, the self-efficacy factor is confidence in finding something specific to the immediate task at hand (e.g., useful information). In the consolidation phase, the self-efficacy factors are ← 167 | 168 → confidence in finding other interesting information and the degree of surprise with the information found. Self-efficacy is supported by a high quality of the site’s information and technology. Surprise is an emotion, resulting from the violation of expectations; it has immediate effects, including motivational ones. Goal conditions and self-efficacy factors act in a cyclical pattern. Achieving a goal strengthens self-efficacy, while self-efficacy, in turn, affects goal foreshadowing.

The three key words in the Psychological Model of the Internet-User are “self-efficacy,” “curiosity,” and “flow.” Self-efficacy is the sense that one can do what is required to accomplish a particular outcome. Curiosity (i.e., the goal is to know more) is raised by stimuli that are novel, somewhat complex, and surprising or ambiguous. In the presence of such stimuli, people may arrive at serendipitous experiences (i.e., emotionally satisfying discoveries). Moreover, emotional satisfaction derived through curiosity often leads to further curiosity. Self-efficacy and curiosity contribute to arriving at a state of flow. Flow is “a state of often intense concentration” and “the experience of exercising control in a complex, difficult activity.” It gives an intrinsic motivation because in flow, the reward for achieving a goal, is largely intrinsic; performing the action well is largely its own reward (Fredin, 1997, pp. 5-6; Csikszentmihalyi and Csikszentmihalyi, 1988). A person in a state of flow often forgets the time; if a state of flow ends, one is embarrassed about not knowing the real time involved. Self-efficacy, curiosity, and flow result in self-sustaining actions, which, in their turn, motivate the user to spend more time on any site that encourages these feelings of satisfaction. Websites have a higher quality the more they strengthen the user’s feeling of self-efficacy, the more they evoke curiosity, and, as a result, the more they provide a strong sense of satisfaction or the feeling of “flow.”

What site characteristics strengthen the feeling of self-efficacy, evoke curiosity, and provide flow? The relevant literature suggests several different quality indicators (Fredin, 1997; Just, et al., 1997; Day, 1997; Schneiderman, 1997; Buchanan and Lukaszewski, 1997; Fredin and David, 1998; Lu and Yeung, 1998; Nielsen, 2000). We selected 10 of these indicators. The first indicator is the first impression the home page of the site makes upon its users. A positive first impression of the site is expected to evoke curiosity. The second indicator relates to the site’s appearance/looks and includes layout (e.g., symmetry and visual elements such as colors, illustrations, typeface, and background). A visually pleasing (attractive) site is expected to evoke curiosity. The third indicator relates to site content. The content can be more or less informative, relevant/useful, up-to-date and fresh, reliable and objective, easily accessible (thanks to a clear structure, short paragraphs, and summaries), made to measure, and interesting. The higher the quality of the information the site offers and the more understandable the information is, the more the site strengthens one’s feeling of self-efficacy. The fourth indicator relates to the site’s navigation opportunities. A clear and consistent navigation menu throughout the entire site (including frames, “hyperlinks,” and a ← 168 | 169 → search feature) will help the users to find their way and, as a result, will give them a feeling of self-efficacy. Speed of downloading is the fifth indicator. Waiting too long for site content to download annoys the user. A “speedy” site satisfies the user’s feeling of self-efficacy. The sixth indicator relates to interaction opportunities, allowing for two-way communication and vital feedback. Thanks to interaction with the site owner/maker, fellow visitors, and/or with elements of the site itself, the user may make fortunate and surprising discoveries that strengthen his/her curiosity. The next two indicators are the presence of video fragments and audio snatches. Opportunities to watch video or to listen to spoken words or music are expected to evoke curiosity. The ninth indicator is the number of missing elements. The more a site offers the things the users are looking for, the more the site is expected to strengthen the user’s feeling of self-efficacy. The tenth indicator is the number of elements that do not function well. Possible examples are “links” to pages that are absent or “back”’ keys that do not function due to “automatic referral.“ Ill-functioning elements are expected to undermine the users’ self-efficacy.

The Experiment

To explore the relationship between times spent on political party websites and their quality, we invited young people in the Netherlands, aged 18 to 21, to take part in a research study under clinical laboratory conditions in a 40-minute session. This young population had the right to vote in national parliamentary elections for the first time. We also focus on young people because surveys found that the Internet is more likely to be used for political information and discussion among the young and young adults (together with those with higher education degrees and people in higher income categories). We also had to limit our study to party websites from just one country. The Netherlands offers good opportunities for research in this field: its electoral system is a proportional representation system; votes are cast for parties and their lists of candidates rather than for single candidates; there are a large number of political parties; almost all parties have their own website (Voerman, et al., 2002); and the Internet has a rapidly growing number of users.

We recruited participants via a website dedicated to our study and another youth website with a link to our research site. We also publicly displayed posters with a reference to our website. The website and corresponding posters invited youth to participate in “research into the evaluation of websites.” The word “political” was excluded purposely to recruit those respondents with low or no political interest. Computer skills and website surfing experience were not required. We wanted to attract other than “dotcom generation” representatives. We made an appointment with the first 41 youngsters who responded by e-mail. We had to limit the number of participants because of a lack of funds. The number 41 corresponds with the “n” in Kern’s study, which largely inspired and motivated us to design our ← 169 | 170 → own research project (Kern, 1997; Kern, et al., 1999). The personal reward for participation was 15 Dutch guilders (7 Euros).

The respondents formed a highly selective group. Almost all participants were university or higher vocational education students (38 and 2, respectively); one participant attended secondary school classes. Women were over-represented (27). The average age was 20. Nine participants had no website surfing experience. Almost all intended to vote in the upcoming parliamentary elections. A large majority said they had voted for the local council last year and most had a relatively high level of political interest. The various limtations with respect to population and sample size limit the generalizability of our research.

Each session followed a set procedure. Upon arrival at the lab (Leiden University, Department of Political Science), the subjects were given a short briefing. The researcher also provided ample opportunities for participants to clarify their doubts. Then the respondent was invited to write answers for 12 closed-ended questions. These questions tapped, among others: subjective political knowledge, political interest, party preference, party attachment, and voting intention.

After completing the questionnaire, each participant was invited to sit in front of the monitor and keyboard. If necessary, participants were given a demonstration of the basic features and introduced to basic commands. The respondent was told that he/she could make a choice on the websites of six parties (PvdA = Labour Party, CDA = Christian-Democrats, D66 = Liberal-Democrats VVD = Liberal-Conservative Party, SP = Socialist Party, and GL = GreenLeft).

These six sites were the only “bookmarks” available in the browser. Each bookmark referred to the “root” of the site. The bookmark opened either the site’s homepage or introductory page. All participants were presented with the same menu of choices. One could start with any party he/she chose, was free to visit any number of sites, and could revisit a site. One was also free to stay at a particular page as long as the respondent wanted. The only limitation was a maximum allowed computer session time of 40 minutes. In practice, 35 of the 41 participants used all six sites.

The researcher sat behind the research subject; the researcher’s position did make the subject feel he/she was kept under constant surveillance but at the same time, was close enough to help the respondent immediately in case of technical problems and to closely watch the screen. The researcher made notes about the order in which the sites were visited, the site elements that did not function, the respondent’s use of interaction opportunities, and the respondent’s viewing of video fragments and listening to audio snatches. The “Surfing Spy” program made a log file (i.e., an electronic account of activity on the research site) for each respondent. The “cache” included all pages that the respondent downloaded, including all elements such as illustrations. ← 170 | 171 →

Following the computer session, a structured debriefing interview was held with the respondent to determine his/her evaluation of the sites’ characteristics. During this interview, we showed the home- or intro-pages of the party sites that the respondent had visited as a memory aid on the screen. The interviews lasted approximately 25 minutes. The interview questions relate to all 10 quality indicators. For the first impression (appearance, contents, and speed), the respondents were invited to give each site a grade between 0 and 10. Next, they were asked for amplification. Regarding missing elements, the respondents were asked whether they thought particular things, elements, or opportunities were lacking and, if so, what they missed. The answer options for the question about navigation were: very clear, clear, not clear, and not clear at all. The interviews were tape recorded and transcribed. The transcripts (in Dutch) are available for review.

We expected that party website use would positively correlate with party preference and party attachment. Respondents preferring party x were expected to use the site of party x longer than respondents who did not prefer party x. Kern, et al. (1999) demonstrated this correlation in the US. In three of the six cases, the correlation between website use (in seconds) and party preference (a dichotomous variable) was statistically significant (sites 3, 4, and 6), while in the other three cases, this significance was almost reached (sites 1, 2, and 5). The Pearson’s r varied from .20 to .50 (site 1 PvdA: r=.21, p=.10, N=39; site 2 CDA: r=.26, p=.07, N=36; site 3 D66: r=.50, p=.00, N=35; site 4 VVD: r=.29, p=.05, N=35; site 5 SP: r=.20, p=.12, N=38; site 6 GL: r=.40, p=.00, N=40; we considered not visiting a site to be a missing value). We also explored the correlation between party site use and subjective political knowledge, political interest, voting intention, and voting or abstention in the past. No correlation was found, with the exception of the time spent on one of the six sites and subjective political knowledge (site number 1; r = .35, p = .02, N = 33). Almost none of the respondents expressed any party attachment.

Party preference was unequal in the group of respondents (site 1 party PvdA: 10 participants; site 6 party GreenLeft: 10 participants; site 3 party D66: 7 participants; site 4 party VVD: 5 participants; site 2 party CDA: 2 participants; and site 5 party SP: 1 participant; no party preference had 6 participants). To avoid distortions in our analyses aiming at exploring the effects of the sites’ quality, we weighted the data for the time spent on the party sites for party preference. To illustrate this weighting, we use the data about the fifth (SP) site, visited by 38 respondents. Out of these, 32 preferred one of the six parties. Out of these, l preferred the party SP, while the other 31 respondents who visited the SP site, did not prefer the SP. The data from the 1 SP sympathizer counts much more than the data from the other 31 respondents. In case of an equal distribution of party preference (32 respondents with a party preference/6 parties = 5.333), respondents would have had an SP preference and 32.666 another party preference. To reach an ← 171 | 172 → equal distribution, the data of the SP sympathizer was weighted with (5.333/1 =) 5.333, while the data of the others were weighted with ((38- 5.333=)32.667/37=) 0.883.

Table 1: Average time spent on six party websites in minutes and seconds (weighted)

The average time spent on the six party websites differed considerably. The first site received longest attention. The difference between the time spent on this site and the second most popular site was almost one minute. The shortest time was spent on the sixth site. The difference between this site and the site on the next to final place in the rank order was also almost one minute.

We made 2-by-2 comparisons to determine whether the sites statistically significant differ with respect to average time. The same weighting procedure was followed for the comparisons of two sites. To illustrate this, we use the data about the first (PvdA) site and fifth (SP) site. In total, 36 participants visited both sites. Out of these, 3l preferred one of the six parties. In case of an equal distribution of party preference (31 participants/6 parties=), 5.167 participants would have had a PvdA preference, 5.167 participants a SP preference, and so forth. In reality, there were 9 participants with a PvdA preference and 1 with a SP preference. To reach an equal distribution, the data of the PvdA sympathizers were weighted with (5.167/9 =) 0.574, the data on the SP sympathizer with (5.167/1 =) 5.167, and the data of the others with ((36-5.167-5.167=) 25.666/26=) 0.987. The 2-by-2 comparisons show that the differences in five combinations are statistically significant (t-tests, p<.05). Four clusters can be identified: first, the first site to which more time was spent than the fifth and sixth sites. Second, the second, third, and fourth site to which more time was spent than the sixth site. Third, the fifth site to which less time was spent than the first site. Finally, the sixth site to which less time was spent than the first, second, third, and fourth site. ← 172 | 173 →

Table 2: Difference in average time spent on six political party websites in two by two comparisons (weighted)

Table 3: Average scores for 10 characteristics of six party websites (weighted)

We used the 10 indicators mentioned previously to measure respondents’ evaluations of the websites’ general quality. Average grades were counted for respondents’ first impression of the site plus its appearance, contents, and down1oad speed (using a 1-10 scale). The navigation assessment is the average level of clarity (using a 1-4 scale). Interaction, video, and audio are measured as the average number of times respondents used the opportunities. For the missing elements and those not functioning well, we counted the average numbers. We weighted the data about the individual indicators the same as for the average time spent at sites. Some indicators received clearly different average scores for the ← 173 | 174 → various sites (e.g., appearance varied from 6.0 to 7.5). Other indicators received almost the same average scores (e.g., average number of audio fragments the respondent listened to). In general, sites on which respondents spent the longest time also received high quality scores. The low time score for the GreenLeft site is also coupled with low quality scores.

The general quality of a party site is the sum of the average scores for the first seven characteristics, minus the average scores for the last two indicators. We excluded the average numbers of audio fragments the respondents listened to because they were extremely low for all six sites. For example, one respondents’ scores were: (first impression: 8 + appearance: 7 + contents: 6 + speed: 5 + navigation: 2 + interaction: 2 + video: 1 -) 31 - (missing elements: 1 + not well functioning elements: 1-) 2 = 29. Together, the nine indicators formed a reliable scale (Cronbach’s alpha .70; N = 6 party sites). The first two sites received the highest scores; the third, fourth, and fifth sites were in the middle, while the sixth site appeared last.

Table 4: Average general quality scores for six party websites

WebsiteQuality
1 PvdA31.76
2 CDA32.66
3 D6629.42
4 VVD30.39
5 SP30.03
6 GL28.32

The 2-by-2 comparisons showed that the differences in eight combinations were statistically significant. Four out of the five combinations of sites that differ in time they were visited by the participants also differ with respect to the evaluation of the general quality (1-6, 1-5, 2-6, 3-6). The remaining combination of sites (4-6) differs with respect to the time they were visited, but not with respect to the evaluation of the general quality. This may be due to the relatively small number of respondents that answered all relevant questions for these two sites (n = 25). Four other combinations of sites also differ with respect to general quality, but have not differed in the time they were visited by the participants (2-3, 2-4, 2-5, 3-1). The remaining six possible combinations of two websites (which are not mentioned in Table 5) do not differ with respect to time or to general quality (1-2, l-4, 3-4, 3-5, 4-5, 5-6).

There is a strong correlation between the average time spent on a party website and its average score for quality (Pearson’s r .80, p .03*, n = 6). Quality explains 63.7% of the variance in average time spent on the sites. This analysis relates to six cases: the six party sites. There are 12 scores: six scores for the dependent variable (i.e., the average time spent on the six sites) (see Table l, third column) and six ← 174 | 175 → scores for the independent variable (i.e., the average quality score for the six sites) (also see Table 4). Note that we do not use the 41 subjects as the cases to be analyzed because we do not intend to explain differences among these subjects. Rather, we want to explain differences among the six party sites. The conclusion is that the higher the quality of the site, the more time the participant spent on it. There are a few combinations of sites that do not fit in this pattern (e.g., the 1-2 sites) (Figure l). The first site received more time from the subjects than the second site, although the latter received a higher quality score. The second site scored higher than the first site on only two out of the nine indicators: navigation and speed. It may be that some quality indicators are more important than others (e.g., appearance and content compared to navigation and download speed).

Figure l: Correlations between the average times spent (in seconds) on political party sites and average score (range of 27-33) for their quality ← 175 | 176 →

Table 5: Difference in general quality between six political party websites in two by two comparisons

Our comments address three main points, each related both to the limitations of the present work and to directions for future work. First, we had to limit our study to party websites from just one country (the Netherlands). That country’s electoral system is a proportional representation system. Future comparative research should include party sites from other countries with the same proportional representation system and sites from countries with plurality, majority, and mixed electoral systems. Second, we expected a positive correlation between the amount of time spent on party websites and the subject’s party preference and party attachment. We found empirical evidence for the positive correlation between party site use and party preference. Almost none of the respondents expressed any party attachment. No correlation was found between party site use and subjective political knowledge, political interest, voting intention, and voting or abstention in the past (with the exception of the time spent on one of the six sites and subjective political knowledge).

The absence of correlations might be explained by the absence or low level of dispersion of values for these variables. In turn, this is caused by the selective composition of the group of subjects. In future studies, a random sample (of youth and young adults) may allow us to better analyze such correlations between party website use and party attachment, political knowledge, political interest, voting intention, and voting behavior in the past. Third, we gave all quality indicators the same weight in our analysis. There are some indications that users consider appearance and content more important than navigation and speed. In our study, both appearance and content were covered by just one indicator. More specific appearance indicators may relate to typeface, colors, and background. More specific content indicators may be the sites’ levels of comprehensibility and conciseness. We may obtain a higher percentage of explained variance in time spent on party sites by weighting quality indicators differently and using more specific appearance and content quality indicators. ← 176 | 177 →

Epilogue and Conclusions

There are two competitive hypotheses about the overall political influence of ICT on representative democracies. First, the “mobilization thesis” predicts that ICT will facilitate forms of direct democracy, allowing more opportunities for citizen decision-making (initiatives and referenda), grassroots mobilization, and community organization. ICT reduces the need for indirect representation (Macpherson, 1998). ICT will contribute to shrink the distance between politics and citizens. Political Internet activities will strengthen citizens’ political competence and activism. In April 2001, I received an e-mail “View from the Harvard Yard: It Is a Revolution” from Phil Noble, President of PoliticsOnline and Resident Fellow of the Institute of Politics at Harvard University’s John F. Kennedy School of Government. The message was “More than ever, I am convinced that the Internet is causing a revolution in politics and government. We may only be at 10am in the morning on day one of the revolution, but it is here. It is happening now.”

Second the “normalization hypothesis” predicts that the major traditional offline institutions and interests (such as governments, strong political parties, established media, and corporations) will increasingly dominate the Internet (Coleman and Hall, 2001). Core political institutions and middle-level actors will be thereby strengthened (Norris, 2000). The gap between political elites and large groups of citizens will widen due to a growing digital divide. Access to the Internet will be sharply uneven because economic and technological resources/access and computer skills are unequally distribued, creating an information underclass. Offline inequalities are being replicated in cyberspace (Resnick, 1998).

Third, a separate version of the normalization hypothesis says that new computer networks are an essentially conservative technology that strengthens the prevailing liberal and capitalist global order. Computer networks produce greater elite control over citizens, tighten the screws, and make global economic and political elites richer and more powerful. Computer networking is the Trojan horse for democracies; it is the ultimate capitalist tool due to its predominant control function (Barney, 2000).

But “revolution” is probably an incorrect term; the Internet has transformed politics, but it is currently not a medium that destroys or replaces existing political systems and their power distributions. ICT seems to support the maintenance of pre-existent basic power structures (Bovens, 2003). Autocracies are helped by economic growth due to ICT (Taubman, 1998; Hachigian, 2001; Sandschneider, 2001). Representative democracies (which suffer from a serious crisis due to lowering election turnouts) may be helped by Internet voting because it may bring young first-time voters (who have consistently low rates of election turnout) back into the electoral process.

ICT offers more opportunities to political actors who have more power and financial resources (among other power factors) to buy Internet management ← 177 | 178 → expertise, but with an interest in maintaining the status quo. There is also reason to speculate that parliaments will lose influence and power to ministries and public agencies that have more expertise in managing Internet activities. Private companies may also acquire more political influence thanks to their ICT resources and expertise.

At the moment, there seems to be more evidence for the stability thesis than the revolution thesis and more evidence for the normalization thesis than the mobilization thesis. Political socialization in existing political systems plays fundamentally a conservative role; the status quo is maintained. Individuals are led to develop actual behaviors and orientations that are according to, and fit into, the existing political culture and system. The two Internet sites which received the longest participant/subject attention in our research belong to the major Dutch political parties. The shortest time was spent on the oppositional political party site. The first two sites also received the highest quality scores, while the oppositional site arrived in last position. Parties with many members have more financial resources to buy website expertise. ICT and the Internet particularly offer new opportunities for promoting the classic stability and supportive nature of traditional political socialization.

Acknowledgements

The authors thank Montague Kern (Rutgers University) and Rudy Andeweg (Leiden University) who inspired and motivated us to design this research project. We are grateful to Russell F. Farnen (University of Connecticut), Katerina Pouliasi (ERCOMER, Utrecht University), Hsuan Chou (University College Utrecht), and participants in the 2001 IPSA Research Committee on Political Socialization and Political Education Conference, the 2001 Netherlands Institute of Government seminar, the 2001 Annual Conference of the Netherlands Association of Political Science, and the 2001 General Conference of the European Consortium for Political Research for their comments on drafts of this chapter.

References

American University Center for Congressional and Presidential Studies, School of Public Affairs (June 12, 2000). Public Attitudes. Toward Campaigns and Campaign Practices. http.//-www.american.edu/acadenuc.depts/spa/ccps/campaignconduct/publications.html.

Bandura, A. (September 1989). “Human Agency in Social Cognitive Theory,” pp. 1175-1184 in American Psychologist, Vol. 44.

Barber, B. (1984). Strong.Democracy: Participatory Politics for a New Age. Berkeley, CA: University of California Press.

Barber, B., K. Mattson, and J. Peterson (1997). The State of “Electronically Enhanced Democracy”: A Survey of the Internet. New Brunswick, NJ: Walt Whitman Center.

Barney, D. (2000). Prometheus Wired. The Hope for Democracy in the Age of Network Technology. Chicago, IL: The University of Chicago Press. ← 178 | 179 →

Bell, H. and N. Tang (1998). “The Effectiveness of Commercial Internet Web Sites: A User’s Perspective,” pp. 219-228 in Internet Research: Electronic Networking Applications and Policy, Vol. 8, No. 3.

Bovens, M. (2003). De digitale republiek (The Digital Republic). Amsterdam, the Netherlands: Amsterdam University Press.

Briggs, R. and N. Hollis (1997). “Advertising on the Web: Is There Response Before Click- Through?,” pp. 33-45 in Journal of Advertising Research, Vol. 37, No. 2.

Buchanan, R. and C. Lukaszewski (1997). Measuring the Impact of Your Web Site. New York, NY: Wiley.

Coleman, S. and N. Hall (August 10, 2001). “Spinning on the Web: E-campaigning and Beyond,” pp. 7-24 in S. Coleman (ed.) 2001: Cyber Space Odyssey. The Internet in the UK Election. The Hansard Society. http://www.hansardsociety.org.uk.

Connell, M. (1997). “New Ways to Reach Voters and Influence Public Opinion on the Internet,” pp. 64-68 in Campaigns & Elections, Vol. 18, No. 8.

Conway, M. (2000). “Some Alternative Political Socialization Research Designs,” pp. 259- 270 in R. Farnen, H. Dekker, D. German, and R. Meyenberg (eds.) Democracies in Transition: Political Culture and Socialization Transformed in West and East. Oldenburg, Germany: BIS.

Crabtree, J. (June 11, 2001). Whatever Happened to the E-lection. A Survey of Voter Attitudes towards New Technology during the 2001 Election. London, UK: Industrial Society, Futures Department. http://www.indsoc.co.uk/whats_new/press/010601.. -htrn

Csikszentmihalyi, M. and I. Csikszentmihalyi (eds.) (1988). Optimal Experience: Psychological Studies of Flow in Consciousness. New York, NY: Cambridge University Press.

Day, A. (1997). “A Model for Monitoring Web Site Effectiveness,” pp. 109-115 in Internet Research: Electronic Networking Applications and Policy, Vol. 7, No. 2.

Dekker, H. (1991a). “Political Socialization Theory and Research,” pp. 16-58 in H. Dekker and R. Meyenberg (eds.) Politics and the European Younger Generation. Oldenburg, Germany:BIS.

Dekker, H. (1991b). “Political Socialization in Europe,” pp. 318-349 in H. Dekker and R Meyenberg (eds.) Politics and the European Younger Generation. Oldenburg, Germany: BIS.

Dulio, D., D. Goff, and J. Thurber (March 1999). “Untangled Web: Internet Use during the 1998 Election” in Political Science and Politics. Also on APSA’s PS Online: http://www.apsanet.org/PS/march99/thurber.cfin (accessed 16 August 2001).

Dunleavy, P. and H. Margetts (November 9, 2000). “The Advent of Digital Government: Public Bureaucracies and the State in the Internet Age.” Paper prepared for delivery at the 2000 Annual Meeting of the American Political Science Association, Washington, DC. http://pro.harvard.edu/abstracts/024/024002MargettsHe.htrn.

Earnshaw, B. (August 10, 2001). “Winning a Seat in Cyber Space: Candidate Web Sites in the 2001 General Election,” pp. 47-60 in S. Coleman (ed.) 2001: Cyber Space Odyssey. The Internet in the UK Election. The Hansard Society. http://www.hansard-society.org.uk.

European Commission, Directorate-General Press and Communication (2001). Eurobarometer. Public Opinion in the European Union. Report Number 54. Fieldwork: November-December 2000. Brussels, Belgium: European Commission, Directorate-General Press and Communication (http://europa.eu.int/comm/dg 10/epo).

Farnen, R. (ed.) (1993). Reconceptualizing Politics Socialization and Education. Oldenburg, Germany: BIS. ← 179 | 180 →

Farnen, .R., H. Dekker, R. Meyenberg, and D. German (eds.) (1996). Democracy, Socialazation and Conflicting Loyalties in East and West: Cross-national and Comparative Perspectives. New York, NY: St. Martin’s Press and London, UK: Macmillan.

Farnen, R., H. Dekker, D. German, and R. Meyenberg (eds.) (2000). Democracies in Transition. Political Culture and Socialization Transformed in West and East. Oldenburg, Germany: BIS.

Filzmaier, P. (June 11, 2001). “Netzgeneration und Politische Bildung: Wissen Kompetenz, und Qualifikation in der ‘Cyberdemocracy’ als digitale Klassen gesellschaft” (The Net Generation and Political Education). Paper prepared for the Institute of Political Science, University of Innsbruck, Austria. http://wwwpoliteia.net/semmar/Papers/ONLINE-Filzmaier.htrn.

Fredin, E. (1997). Rethinking the News Story for the Internet: Hyperstory Prototypes and a Model of the User. Journalism and Mass Communication Monographs, 163. Columbia, SC: Association for Education in Journalism and Mass Communication.

Fredin, E. and P. David (1998). “Browsing and the Hypermedia Interaction Cycle: A Model of Self-efficacy and Goal Dynamics,” pp. 35-54 in Journalism & Mass Communication Quarterly, Vol. 75, No. 1.

George Washington University, Graduate School of Political Management, Democracy Online Project (2000). Post-Election 2000 Survey on Internet Use for Civics and Politics. http://-democracyonline.org/databank/dec2000survey.shtrnl.

Gibson, R. (2002). “Elections Online: Assessing Internet Voting in Light of the Arizona Democratic Primary,” pp. 561-583 in Political Science Quarterly, Vol. 116, No. 4.

Groper, R. (1996). “Political Participation and the Internet: A Review Essay,” p. 247-249 in Political Communication, Vol. 13, No. 2.

Hachigian, N. (March/April 2001). “China’s Cyber-Strategy,” pp 118-133 in: Foreign Affairs.

Hague, B. and B. Loader (eds.) (1999). Digital Democracy; Discourse and Decision Making in the Information Age. London, UK: Routledge.

Hansard Society (August 10, 2001). “E-mail Test,” pp. 29-30 in S. Coleman (ed.). 2001: Cyber Space Odyssey: The Internet in the UK Election. The Hansard Society. http://www.hansardsociety.org.uk.

Hill, K. and J. Hughes (1997). “Computer-mediated Political Communication: The Usenet and Politica1 Communities,” pp. 3-27 in Political Communication, Vol. 14, No. 1.

Holmes, D. (ed.) (1997). Virtual Politics. Identity and Community in Cyberspace. London, UK: Sage.

Horrocks, I., J. Hoff, and P. Tops (eds.) (2000). Democratic Governance and New Technology. London, UK: Routledge.

Industrial Society (June 11, 2001). iSociety. London, UK: The Industrial Society Futures Department. http://vyww.mdsoc.co.uk/isociety/press release htrn

Industrial Society (June 11, 2001). Whatever Happened to the E-lection. A Survey of Voter Attitudes towards New Technology during the 2001 Election. London, UK: The Industrial Society, Futures Department. http://www.indsoc.co.uk/whats_new/press/010601.htm.

Inter-Parliamentary Council (May 6, 2000). “Guidelines for the Content and Structure of Parliamentary Web Sites.” Approved by the Inter-Parliamentary Council at its 166th session in Amman. http://www.ipu.org/cntr-e/web.pdf.

Just, M., A. Crigler, and M. Kern (1997). “Citizens and the Internet: Diversifying Information Sources in Cyberspace.” Paper presented at the Annual Meeting of the National Speech Communication Association, Chicago, IL. ← 180 | 181 →

Kern, M. (1997). “Social Capital and Citizen Interpretation of Political Ads, News, and Web Site Information in the 1996 Presidential Election,” pp. 1238-1249 in American Behavioral Scientist, Vol. 40, No. 8.

Kern, M., M. Just, A. Crigler, and I. Xie (July 18, 1999). “Patterns of World Wide Web Use in Election Campaigns: Evidence from Interactive Interviews.” Paper presented at the Annual Meeting of the International Society for Political Psychology in Amsterdam, the Netherlands.

Lee, B. and J. Anderson (June 12, 2001). “Current Attitudes of Internet Users.” Paper submitted for presentation at the Association for Education in Journalism and Mass Communication, The Communication Technology and Policy Division, Washington, DC, 2001. Elon, NC: Elon University, School of Communications. http://www.elon.edu/pew/oneweek and http://-www.pewinternet.org/papers/paper.asp?paper=8.

Lenhart, A., L. Rainie, and O. Lewis (2001). Teenage Life Online. The Rise of the Instant-message Generation and the Internet’s Impact on Friendships and Family Relationships. Washington, DC: Pew Internet & American Life Project. http://www.pewinternet.org.

Lu, M. and W. Yeung (1998). “A Framework for Effective Commercial Web Application Development,” pp, 166-173 in Internet Research: Electronic Networking Applications and Policy, Vol. 8, No. 2.

Macpherson, M. (December 7, 1998). “Some Applications of Information Technology to Citizen Participation in Politics.” Online paper. http://www.snafu.de/-mjm/miercurea.html.

Meeks, B. (2000). “Most E-mail to Congress is Ignored; Lawmakers Overwhelmed, Study Says. Flooded by Spam, Politicians Reply by ‘Snail Mail’ If at All.” http://www.msnbc.com/news-/540014.asp (11 June 2001).

Nielsen, J. (2000). Designing Web Usability: The Practice of Simplicity. Indianapolis, IN: New Riders.

Niemi, R. and M. Hepburn (1995). “The Rebirth of Political Socialization,” pp. 7-16 in Perspectives on Political Science, Vol. 24, No. 1.

Norris, P. (November 9, 2000). “Democratic Divide? The Impact of the Internet on Parliaments Worldwide.” Paper prepared for delivery at the 2000 Annual Meeting of the American Political Science Association, Washington, DC. http://pro.harvard.edu/abstracts/038/038010-NorrisPipp.htm.

Pew Research Center For The People & The Press (June 11, 200la). “Internet Sapping Broadcast News Audience.” Section V: Media Credibility. http://www.peoplepress.org/media00sec-5.htm.

Pew Research Center for The People & The Press (June 11, 2001b). “Youth Vote Influenced By Online Information.” http://www.people-press.org/onlineOOrpt.htm.

Pew Internet & American Life Project (June 12, 2001). “Internet Election News.” Report 27. http.//www.pewinternet.org/reports.asp?Report=27&Section=ReportLevel1&Field=LevelIID&ID=9.

Prins, J. (ed.) (2001). Designing E-government; On the Crossroads of Technological Innovation and Institutional Change. The Hague, the Netherlands: Kluwer Law International.

Resnick, D. (1998). “The Normalization of Cyberspace,” pp. 48-68 in C. Toulouse and T. Luke (eds.) The Politics of Cyberspace. New York, NY: Routledge.

Sadow, J. (2000). “A Uses and Gratifications Theory of Internet Campaigning.” Paper prepared. for delivery at the 2000 Annual Meeting of the American Political Science Association, Washington, DC.

Sadow, J. and K. James (June 11, 2001). “A Theory of Internet Political Campaigning: A Revolution that Isn’t, Yet.” Paper presented at the 2000 Annual Meeting of the Southwest Political ← 181 | 182 → Science Association, Galveston, TX. http://www.people.uis.edu/jsado1/uis/cwsp/swpsa00.htm.

Sandschneider, E. (2001). “The Internet and China’s Political System,” pp. 23-26 in European Political Science, Vol. 1, No. 1.

Schneiderman, B. (1997). “Designing lnformation-abundant Websites: Issues and Recommendations,” pp. 5-30 in International Journal of Human-Computer Studies, Vol. 47, No. 1.

Sigel, R. (1995). “New Directions for Political Socialization Research,” pp. 17-22 in Perspectives on Political Science, Vol. 24, No. 1.

Solop, F. (November 9, 2000). “Digital Democracy Comes of Age in Arizona: Participation and Politics in the First Binding Internet Election.” Paper prepared for presentation at the 2000 Annual Meeting of the American Political Science Association, Washington, DC. http://pro.harvard.edu/abstracts/036/036015SolopFrede.htm.

Street, J. (1997). “Remote Control? Politics, Technology and ‘Electronic Democracy,’” pp. 27-42 in European Journal of Communication, Vol. 12, No. 1.

Sussman, L. (2001). “The Internet in Flux,” pp. 1-4 and 41 in Press Freedom Survey 2001. New York, NY: Freedom House. http://www.freedomhouse.org.

Taubman, G. (1998). “A Not-so World Wide Web: The Internet, China, and the Challenges to Nondemocratic Rule,” pp. 255-272 in Political Communication, Vol. 15, No. 2.

Toulouse, C. (1998). “Introduction,” pp. 1-16 in C. Toulouse and T. Luke (eds.) The Politics of Cyberspace. New York, NY: Routledge.

Voerman, G., A. Keyzer, F. den Hollander, and H. Druiven (2002). “Archiving the Web: Political Party Web Sites in The Netherlands,” pp. 68-75 in European Political Science, Vol. 2, No. 1.

Ward, S., and R. Gibson (1998). “The Potential of the Internet: Issues for Research,” pp. 6-8 in ECPR News, Vol. 10, No. 1.

Washburn, P. (1994). “A Life Course Model of Political Socialization,” pp. 1-26 in Politics and the Individual, Vol. 4, No. 2.