This item is only available as the following downloads:
1 USES AND GRATIFICATIONS OF ONLINE INFORMATION SOURCES : P OLITICAL INFORMATION EFFICACY AND THE EFFECTS OF INTERACTIVITY By DAVID LYNN PAINTER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Da vid Lynn Painter
3 To all who nurtured my intellectual curiosity and love of learning, you made this milestone possible
4 ACKNOWLEDGMENTS First, I would like to thank Dr. Lynda Lee Kaid for her generosity and support througho ut my doctoral program. She is a great mentor and an even better friend. I would also like to thank Dr. Spiro Kiousis, Dr. Johanna Cleary, and Dr. Michael Martinez for their guidance and encouragement I take great pride in the qualit y of my committee and am most appreciative of all their efforts assistance and most especially, their patience Second, I would like to thank my family for making it possible for me to pursue and achieve a life long dream. Thank you to m y mother, Sara V ance, for giving me life and restoring me to life, and to my sister, Julee Vance, for enriching my life. I would also like to thank my brother, Drew Vance, for constantly if unwittingly, reminding me how little I actually know. Third, I would like to than k my classmates officemates, and collaborators: Juliana Fernandes, Maridith Miles Dunton, K. West Bowers, Eisa Al Nas h mi, and Ji Young Kim. Thanks to all of you for making this arduous journey less onerous I appreciate each of you value our friendship s, and treasure your good humor O ur time together in G034 of Weimer Hall was truly priceless Last, I would like to thank Dr. Debbie Treise the Associate Dean, and the graduate staff in the College of Journalism and Communications. Dr. Treise, I will alway s remember meeting you at the airport in Keystone Heights and am most appreciat ive of all your efforts on my behalf I would also like to thank Jody Hedge for her assistance with m y issues related to registration, deadlines, and scheduling. I am quite sure the graduate program would not function as smoothly as it does without Jody and Kim Holloway
5 for t he i r assistance with my admission and funding throughout the last three years. You four members of the graduate staff make it possible for all of us to succeed.
6 TABL E OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Problem Statement ................................ ................................ ................................ 12 Background of Study ................................ ................................ .............................. 15 Focus and Significance of Research ................................ ................................ ...... 17 2 THEORETICAL BACKGROUND ................................ ................................ ............ 22 Rational Voting Behavior and Po litical Efficacy ................................ ....................... 22 Political Information Efficacy ................................ ................................ ................... 25 Market Penetration of the Internet ................................ ................................ .......... 27 Uses and Gratificat ions Theoretical Framework ................................ ..................... 29 Online Information Formats Uses and Gratifications ................................ .............. 33 Online News ................................ ................................ ................................ ..... 38 Social Network Sites ................................ ................................ ........................ 39 Interactivity ................................ ................................ ................................ .............. 41 Expression Effects ................................ ................................ ................................ .. 48 Hypotheses and Research Questions ................................ ................................ .... 51 3 METHOD ................................ ................................ ................................ ................ 57 Participants and Procedure ................................ ................................ ..................... 57 Stimuli ................................ ................................ ................................ ..................... 58 News Websites ................................ ................................ ................................ 58 S ocial Networking Sites ................................ ................................ .................... 58 Variables ................................ ................................ ................................ ................. 59 Levels of Interactivity ................................ ................................ ........................ 59 Political Information Efficacy ................................ ................................ ............. 60 Political Interest ................................ ................................ ................................ 60 Political Knowledge ................................ ................................ .......................... 60 Political Party Affiliation ................................ ................................ .................... 61 Civic Participation ................................ ................................ ............................. 61 Political Information Sources ................................ ................................ ............ 62 Candidate Evaluation ................................ ................................ ....................... 62
7 Saliency of Election ................................ ................................ .......................... 62 Learning ................................ ................................ ................................ ........... 63 4 RESULTS ................................ ................................ ................................ ............... 64 Demographics and Manipulation Check ................................ ................................ 64 Predictive Power of Political Information Efficacy ................................ ................... 65 Onl ine Format and Interactivity Effects ................................ ................................ ... 69 Partisanship, Candidate Preference, Saliency of the Election ................................ 75 5 DISCUSSION ................................ ................................ ................................ ......... 81 Predictive Power of Political Information Efficacy ................................ ................... 81 Cognitive Effects of Online Format and Interactivity Levels ................................ .... 83 Online Format and Interactivity Effects on Information Efficacy .............................. 86 Candidate Evaluations and the Saliency of the Election ................................ ......... 88 Theoretical Implications ................................ ................................ .......................... 90 Limitations ................................ ................................ ................................ ............... 96 APPE NDIX A INTERACTIVE ACTIVITIES PRETEST ................................ ................................ .. 9 9 B PARTICIPANT INSTRUCTIONS ................................ ................................ .......... 103 C STIMULI ................................ ................................ ................................ ................ 107 LIST OF REFERENCES ................................ ................................ ............................. 108 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 122
8 LIST OF TABLES Table page 4 1 Differences in political characteristics by information efficacy levels .................. 77 4 2 Predictors of traditio nal media use for vote guidance ................................ ......... 77 4 3 Predictors of I nternet usage for vote guidance ................................ ................... 77 4 4 Learning by format, type, and candidate ................................ ............................ 78 4 5 Leaning by interactive condition, type and candidate ................................ ......... 78 4 6 Exposure effects on political Information efficacy by format and condition ......... 78 4 7 Online format effects on candidate evaluat io n ................................ .................... 78 5 1 Summary of findings ................................ ................................ ........................... 98
9 LIST OF FIGURES Figure page 4 1 ................................ 79 4 2 .................. 79 4 3 Political information efficacy gains by format and condition ................................ 80
10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy USES AND GRATIFICATIONS OF ONLINE INFORM ATION SOURCES : POLITICAL INFORMATION EFFICACY AND INTERNET CHANNEL EFFECTS AMONG DIGITAL NATIVES By David Lynn Painter May 201 1 Chair: Spiro Kiousis Major: Mass Communication There are two distinct purposes to this investigation: to develop the predictive power of the theory of political information efficacy and to determine how exposure to political information formats on the Internet under distinctly different interactive conditions affects cogn itive and affective processes Examined i n the first section of this inquiry, then are the relation ships among information efficacy and political interest, knowledge, participation, and part y affiliation Secon d the differential effects of exposure to news web sites or social network sites and three levels of interactivity on gubernatorial campaign in Florida are explored Grounded in the uses and gratifications theoretical framework, the results suggest that political information efficacy predicts levels of political interest, knowledge and party affiliation. Further exposure to online political information sources also political information efficacy, issue and image learning, and candidate evaluations. Specifically, exposure to both online political information sources results in significantly higher levels of political information efficacy; exposure to news web sites results in
11 differentially significant issue learning; exposure to social network sites results in both differentially significant candidate image learning and evaluations. The implications of social orientation toward news consumption and preference for using social network sites as portals for political information seeking are also explored.
12 CHAPTER 1 INTRODUCTION Problem Statement Since the 26 th amendment to the U.S. Constitution lowered the voting age to18 in 197 1 young citizens have lagged far behind their older cohorts in nearly all measures of political participation Scholars point to the low rates of voter turnout, dismal levels of political knowledge, and sense of civic disconnection among yo ung citizens as evidence of the marginalization of young Americans in our political system (Bennett, 1998; Iyengar & Jackman, 2003; Plutzer, 2002). Even w hen participation rates among the young have increased, the y have been labeled anomalies or credited to short terms forces such as candidate image appeal, issue voting, closely contested elect ions, or the mobilization efforts of get out the vote organizations (Kirby & Kawashima Ginsburg, 2009). This lack of civic participation particularly among the young, is cause for concern in terms of both normative democratic theory and because the continued legitimacy of our government depend s upon an engaged citizenry (Hyland, 1995). While young voters were the only age group to achieve statistically significant increases in turnout in both the 2004 and 2008 presidential elections, the fact remains that less than 60% of citizens between the ages of 18 and 24 were registered and less than half actually voted in either election ( U.S. Census Bureau 2009). When examining midterm elections, voting rates fall far below their high point in 2008. In 2006, for examp le, more young voters turned out at the polls than in any midterm election since 1992, but less than half reported that they were registered to vote, and only 22% reported they actually did so (U.S. Census Bureau, 2009). The celebration of a 22% voting rate as evidence of increasing civic engagement among citizens between
13 the ages of 18 and 29 speaks volumes about the low standards used for judging the political engagement of this age group. Further, in closely conte sted, relatively salient off year gubernatorial races in November 2009, young voters failed to maintain voting rates anywhere near their 22% level in 2006 ; there was a drop off in young voter turnout of two thirds in New Jersey and three quarters in Virginia (ISS, 2009; Barone, 2009). a chronic problem Explanations for this predicament include the lack of political socialization, in effect ive civics education, and low levels of political interest (Delli Carpini, 2000; Gans, 2004). Due to their focus on life stage issues such as getting an education, finding a job, and starting a family politics simply see m irrelevant to many daily lives and immediate concerns (Wolfinger & Rosenstone, 1980). S hould the ha bit of voting and participating in politics not be developed early in life however, s cholars contend that disengaged young citizens are very likely to become disengaged older citizens ( Putnam, 2000 ) I n the first couple of decades after modern voter participation rates peaked in 1960 the issue of young voter apathy was treated as an enigma awai ting an innovative solution or explanation M ore current analyses of the declining rates of civic participation however, are phrased in nearly strident terms ( Brody, 1978; N ASS, 1999 ; Te i xeira, 1992) One of the major reasons for this change in tone is th e discovery that th e trend ing in political apathy continues to be concentrat ed among the young, making the future look even bleaker ( Niemi & Weissberg, 2001). Although increasing levels of education and affluence a long with l ess stringent registration requ irements should be pushing rates of civic participation upward, it is apparent that there are overwhelming and countervailing forces depressing turnout ( Cho, 1999 ; Texeira, 1992 )
14 T he most common reason given by young citizens in focus groups at universit ies across the country for abstaining from politics is the ir lack of confidence in their ability to participate meaningfully in the process (Kaid, McKinney, & Tedesco, 2007 ). Those newly eligible to vote are not only unfamiliar with registration and pollin g practices, but they more frequently express concerns over their low levels of information about the candidates and issues being contested in the upcoming election (Delli Carpini 2000) This sense of doubt about their ability to cast an informed vote has been measured using the construct of political information efficacy The underlying theory driving this research is that individual levels of information efficacy are closely related to other political cognitions, attitudes, and behaviors (Kenski & Stroud 2006). Thus, scholars contend, analyses of both media usage patterns among those with high levels of information efficacy and the effects of different media channels and content on individual levels of efficacy may provide suggestions for increasing poli tical participation ( Sweetser & Kaid, 2008 ). In addition to the lack of political information efficacy among the young, o ther factors cited as contribut ing to the declining rates of civic participation are the rise in influence of the electronic media and the weakening of the political parties ( Wattenberg 1996) Indeed, since the television became the major source of political information for Americans in 1963, voter participation rates have continued their steady decline (Patterson, 1994; 2002) With the development of digital media technology many early analyses of the effects of the Internet on political cognitions, attitudes, and behaviors have also been distinctly negative (e.g., Margolis & Resnick, 2000; Sunstein, 2001 ).
15 Background of Study The extent to which citizens in contemporary western democracies experience politic s through the mass media has led to its characterization as not merely mediated, (Stromback & Esser, 2009). While interpersonal communication continues to play an i ntegral role in shaping political cognitions and attitudes, the majority of the political information consumed by citizens today comes from ubiquitous media sources in the form of both news coverage and strategic campaign communicatio ns. Although most research on media effects ha s focused on the ways in which exposure affects receivers, investigations into the ways in which citizens use specific media channels and content may be required before approp riate questions about their effects can be formulated Such analys e s of media usage among young citizens may not only offer insights into the causes of differing levels of political interest, knowledge, and participation among age groups, but they may also provide opportunities for engaging the young in the political process. Recent research, for instance, reveals that large segments of young citizens are turning to the Internet for their political information as well as participating in polit ics through this medium more so than are older citizens (Smith, 2009). political information sources are clear examples of how young citizens differ from older citizens in their political attitudes and media orientations. An appropriate theoretical framework for investigating the patterns of media usage among any group of consumers especially when focusing on emerging technologies and platforms i s the uses and model in studies of not
16 do the media do to people? made clear his belief that answers to questions about motivations behin d patterns of media usage must be answered before the effects of these channels can be fully understood. That is, in order to unde rstand how media affect people, we must first understand how people use media. The development and widespread adoption of new, interactive media platforms provides scholars with an opportunity to analyze usage as well as effects. For instance, emerging media channels heavily relied upon by the younger generation may also possess interactive features lessening the costs of information gathering required abi lity to cast an informed vote. By reducing the costs of political information gathering, one of the most frequently cited reasons among young citizens for abstaining from the politic process might also be addressed Indeed, recent elect ions offer grounds for hope that features of online digital platforms may not only appeal to the young, but may also activate their political interest. Since online political information and interactive media platforms have become more accessible in recent years, more young voters have been turning out to vote. In the 2004, 2006, and 2008 elections young citizens turned out at the polls in increasingly larger numbers. The future, however, is not so promising and t here remains much to learn about online behaviors in terms of political information seeking and processing Further the emergence of the Internet as a major source of political information provides a rich source for investigating the effects of this uniquely interac tive features on political cognitions, attitudes and behaviors during an election campaign.
17 Focus and Significance of Research A great deal of research has examined the effects of information on the political engagement of different age groups during U.S. presidential elections but far fewer studies have focused on the patterns of media effects and rates of civic participation in midterm contests. Still smaller numbers of examinations have investigated the effects of exposure to different online media platforms on confidence in being able to cast an informed vote or political information efficacy, in statewide midterm elections Although the last two midterm elections witnessed increases in participation among young voters, the outlook for the 2010 cycle is uncertain at best and dismal at worst On the other hand, the ascendance of the Internet as an information source and the adoption of online platforms allowing users to interact with this medium have continued their unprecedented arc. Thus, there is an urgent need for research that may answer question s about the cognitive and affective effects of exposure to different online political information sources and content Further, there is hope that the results of such investigations may offer some suggestions for activating young citizens politically. Whil e the mobilization efforts of get out the vote organizations have been credited with a great deal of the success in increasing young voter turnout in recent presidential elections, such efforts during midterm elections have for the most part, failed to yi eld similar results. Further, compared to the amount of research on voting patterns during presidential elections, relatively few analyses ha ve focused on the effects of information on young voters in midterm elections. Presidential elections, by their ver y nature, are highly salient events that capture a great deal of media and public attention. Midterm elections, on the other hand, are limited to a specific geographic location and receive limited national media coverage. Clearly, rates and patterns of par ticipation in midterm
18 elections are different from and generally much l ower than those in presidential elections ( Wolfinger, Rosenstone, & McIntosh, 1981 ) A major reason for this dramatic drop off in midterm participation is t he increased opportunity cost s required to cast an informed vote (Jackson, 1993) One purpose of the theory of political information efficacy is to explain how young citizens process political information differently from members of older age groups ( Kaid, McKinney, & Tedesco, 2007 ). Another purpose of the theory of political information efficacy is to explain how those who lack confidence in their information levels about political issues and actors become politically dis engaged ( p 1101 ) Th is theoretical construct has also been use d as a dependent variable in experimental investigations into the effects of exposure to various information sources during political campaigns (for summary, see Tedesco, 2007) By adopting a broader uses and gratifications approach, t his investigation, on the other hand, test s the predictive value of the theory of political information efficacy in terms of its relationships with other political cognitions and attitudes as well as patterns of media usage. Following u nprecedented use of Internet re sources by the historic 2008 presidential campaigns the 2010 gubernatorial election in Florida provide s a compelling context for investigating the use of these online resources during subsequent ele ctions. N ot only did this closely contested election receive a great deal of attention from the regional news media but it was also characterized as the election in which social media came of age to play a breakthrough role ( Sac hs, 2010 ). From the beginn ing, however, there were clear signals that this race would be especially compelling. Indeed some
19 analysts asserted, w ere it not for the Florida Senate race, the Florida Governor's race RCP, 2011 p. 1 ). Although the three way 2010 senate campaign in Florida may have been more novel it was not nearly as close as the gubernatorial contest Insurgent Tea Party favorite, Marco Rubio (R) beat the sitting governor, Charlie Crist ( who became an Independent when it became clear he could not win the Republican primary ) by over one million votes and the Democratic nominee, Congressman Kendrick Meek by over 1.5 million votes to win the Senate seat vacated by Mel Martinez (R) ( Stacy & Sedensky, 201 0 ) The difference in votes between the winner and the loser in the gubernatorial race on the other hand was less than 70,000 votes ( p. 1 ) With the sitting governor campaigning for the Senate however, these two races became inexorably linked more than a year before the campaign ing began On May 12, 2009, Governor Charlie Crist (R) announced that he would not run for re election but that he planned to run for the Senate seat held by Mel Martinez ( R), who also did not plan to seek re election and subseq uently announced his resignation three months later With the sitting governor not seeking re election, the resulting open race was hotly contested in the Republican primary between the Florida Republican Party establishment favorite and state Attorney Ge neral Bill McCollum, and the insurgent Tea Party favorite and former CEO of Columbia/HCA, Rick Scott. The tone of this race became distinctly negative early making it one of the nastiest nomination fight s of 2010 ( Mills, 2010 ) McCollum repeatedly hammer ed Scott for being forced to resign after his hospital corporation was found guilty of committing 14 felonies and forced to repay the federal government nearly $2 billion in the largest Medicare fraud case in American
20 history ( De s latte, 2010 ) Scott, on th e other hand, used his vast wealth to funnel over $30 million of his own money into the most expensive primary campaign in the state of ( p. 1 ) He branded McCollum a repeatedly accused him of misusing taxpayer doll ars with frivolous plane trips and self serving public service announcements ( p. 2 ) Although polls showed McCollum leading Scott heading into primary E lection D ay, low voter turnout and early voting results allowed Rick Scott to nomination for governor of Florida by a three point margin ( FDOE, 2010) Apparently rewarded for his negative tactics, t he Republican nominee and eventual election winner, Rick Scott, continued to pour money into attack advertising, outspen ding his D emoc ratic opponent state Chief Fiscal Officer Alex Sink, by a margin of 10 to 1 (Mills & Hale, 2010 ). Scott also continued to portray himself as a political outsider a successful businessperson trying to reform the system He spent over $ 7 0 million of his own mone y and ma de this gubernatorial campaign both the most expensive in Florida history and second only in terms of spending during the 2010 cycle ( p. 1 ). His opponent, Alex Sink, was not a particularly strong candidate, commi tt ing frequent and costly missteps during her campaign For example, s he repeatedly referred to the University of Central Florida ( in Orlando ) as USF (University of South Florida in Tampa ) during a commencement address (ABC News, 2011) Perhaps even more c ostly, exactly one week prior to Election Day Sink accepted a smart phone from her make up artist to read a text message from an aid e during a commercial in a CNN televised debate w ith Scott This communication with her campaign staff violated the debate agree ment that stipulated the candidates would not
21 use notes nor receive any coaching during commercial breaks in the encounter. Upon returning to the broadcast, Scott immediately labeled Sink a cheater First Alex, you say you always follow the rules," h e said. rule was no one was supposed to give us messages during the break, and your campaign did CNN, 2010 ). In sum, both the particularly weak nature of his opposition and the anti Democratic sentiment sweeping the country played in Unlike Meg Whitman and other self funded, big spending, insurgent Republican candidacies in 2010, however, was a success. In addition to the heavy use of negative television advertising, both Scott and his Democratic opponent, Alex Si nk (D), also maintained a significant presence on social network sites ( Sachs 2010 ) Since the variety and sophistication of online platforms as well as their formats proliferated during the run up to th e 2010 gubernatorial race in Florida, this campaign provides an opportunity for exploring a variety of informational uses and their effects. Specifically, this investigation attempts to : (1) test the predictive power of the theory of political informati o n efficacy ; (2) explor e the manner in which the purposes, sources, and uses of Internet sources shape the cognitive and affective processes of those exposed to the information ; and ( 3 ) analyz e the effects of interactive features distinctive of Internet communications on political cognitions and attitudes Grounded in the uses and gratifications approach, the results of this in vestigation may not only inform analyses of political communication in our rapidly evolving media environment, but they may also offer insights into practices offering the potential for increasing political participation among the younger generation.
22 C HAPTER 2 THEORETICAL BACKGROU ND Rational Voting Behavior and Political Efficacy It is important to remember that the act of voting requires making two choices: first, whether or not to participate ; and s econd, which candidate s or proposition s to supp ort. When analyzing the first part of this equation, scholars have met with difficulty when using purely mathematical or rational analytical approach es Indeed, at tempts to explain why some people vote ( and others do not ) have embroiled political sci entists in a dispute over the legitimacy of rational choice theories for decades. Anthony Downs man decides to vote just as he makes all other decisions; if the returns outweigh the is that the costs of voting ( in terms of the time and effort required for information gathering registering, and completing a ballot ) are greater than the potential benefits (i.e., victory by preferred candidate ) because the probability that one vote will chang e the electoral outcome is minuscule. Since an individual vote holds little chance of determining the electoral outcome, then, many rational behavior theorists con c l ude that voting is largely an irrational act. If the act of voting were not susceptible to the rules of reason, however, academic analys e s would be moot and there would be no observable systematic patterns, explanations, or theories of voting behavior In their endeavor to make rational behavior models more descriptive of actual voting behavior, Riker and Ordeshook (1970) revised the calculus of voting to include a wider range of potential benefits for individuals who participate in th e political process. While granting that the prerequisite information and opportunity costs f or participating in
23 elections are considerable, these authors argue the rewards of participation may include more than potentially changing the outcome of the election. For example, satisfaction of affirmin may also be one of the benefits of voting (p. 28). Since democratic theory holds that the citizens are responsible for their government al institutions, actors, and policies, voting may be the only opportuni ty most people have for fulfilling th ere civic obligations and responsibilities. A ffirming the belief that one plays a vital role in the political process may not only be considered a reward for voting behavior however but this sense of political efficacy may also influence individual level decisions whether to participate in the proces s or the feeling that individual political action does have, or can have, an impact upon the political process has been studied since the 1950s, scholars came to realize that there were two distinct components o f this construct in the late 1970s (Nieme, Craig, & Mattei, 1991). This political attitude has responsiveness of governmental authorities and institutions to citi zen demands and ( 1408). The value of these constructs is found in their relationships with other p olitical cognitions attitudes and behaviors For example i f one believes that the government is responsive to citizen demands (high level of external political efficacy) then one will be more likely to participate in the political process than a perso n who does not believe his or her actions have the potential for influencing the government (low external political efficacy)
24 Likewise, if one feels confident in his or her understanding of the candidates and issues being contested in the electio n physically participate (high internal political efficacy) then one will also be more likely to vote than a person who feels less competence (less internal political efficacy) Thus, levels of both internal and and level of civic participation over the past several decades (Aberbach 1969; Abramson & Aldrich 1982 ; Clark & Acock, 1989; Miller & Miller, 1975 ; Morrell, 20 03; Pollock, 1983 ; Verba & Nie, 1972 ). Scholars have also noted t he reciprocal nature of the relationship between efficacy and participation and their recursive effects The results of various investigations show that efficacy influences decisions whether to vot e which in turn, affect one political competence and probability of participating in politics (Finkel, 1985, 1987; Norris, 2004 ). Alternately, o ne of the major diffe rences in research results between these two constructs is their relationship with the calculus of voting equation in which electoral victory by the preferred candidate has clear implications Unlike studies of external political efficacy, levels of inter nal political efficacy have not been shown to be affected by election outcomes. That is, while the re is a documented tendency for levels of external efficacy levels to rise when candidate is elected a similar relationship between internal political efficacy and electoral victory has yet to be established ( Karp & Banducci, 2008 ) In fact, many voters who support losing candidacies credit their loss es to a lack of information among the public ( Kuklinski & Quirk, 2000 ). Further, scholars have shown that misinformation may have even greater effects on electoral outcomes than a lack of information ( Bartels
25 1996) many scholars have focused on media coverage and campaign communi cations in their investigations of information effects on voting decisions ( for summary, see Kaid, 2004) While media coverage may provide information about precinct locations and instructions for ballot completion, the more important question s about the e ffects of information on voting decisions are also one of the most common explanations given for fail ing to participate : H ow does exposure to mediated information affect an individual competence in being able to cast an informed vote ? Although low levels of political knowledge are a problem common to all generations in the United States, the shockingly dismal levels among the young have been of particular concern for a number of years (Delli Carpini & Keeter, 1996). Further, the disparity in pol itical knowledge levels between those ages 18 to 29 and those over the age of 30 has been growing significantly in recent years (Pew, 2007). While the lack of political knowledge in general is considered a sign of an unhealthy democracy, it also appears to be a cause of civic disengagement among the young (Bennett, 1997). Specifically, young citizens even more so than older citizens, have frequently cited the lack of political knowl edge as their reason for not voting (Delli Carpini, 2000). In order to understand how low levels of political knowledge among those between the ages of 18 and 29 may cause them not to participate in the political process, scholars have drawn concept ual links between internal political efficacy and political information efficacy. Political Information Efficacy understand and participate in the political system in a general sense, political
26 (Kaid, McKinney, & Tedesco, 2007, p. 1096). Thus, while it ma y be appropriate to measure internal political efficacy at any stage in the processes of the polity, measuring political information efficacy is largely appropriate only during election cycles Methodologically, a clear distinction between these two constructs is found in the differences between the measures. Three of the four items used to measure internal political efficacy and political information efficacy are identical The items used to measure both constructs are : (1) I consider myself well qualified to participate in politics; (2) I feel that I have a pretty good understanding of the important political issues facing our country; and (3) I think that I am better informed about politics and government than most people The fourth item fo r measuring internal political efficacy competence in being able to perform in public office in public office as most other people For political information efficacy, on the other hand, the goal is to measure sense of competence in being able to participate meaningfully in the political process Thus, the fourth item in this construct is worded: would have enough information to help my friend figure out who to vote for While s cale analyses o f both these measures have repeatedly shown high levels of reliability verifying their internal consistency in measuring a single construct this differenc e in the fourth measure makes a clear distinction between internal and information efficacy. Broadly, the concept of self complete specific tasks ( Bandura 1986 ). When looking at the difference betwe en
27 internal efficacy and information efficacy, it is clear that these two constructs are designed to measure individual confidence levels in ability to perform two distinctly different tasks. Indeed, as many teachers, coaches, parents, and professors are very aware, belief in one specific tasks is much different from those same role specific functions. B oth internal and political information efficacy have been used as dependent variables in investigations into the effects of exposure to mediated information. Unlike political information efficacy, however, i nternal political efficacy has also been used as an independent variable to predict other politi cal cognitions, attitudes, and behaviors Thus, research into the predictive power of the theory of political information efficacy would strengthen the theoretical underpinnings of this analytic framework (Chaffee & Berger, 1987 ). Further, s ince research also suggests that different information sources affect political cognitions and attitudes in ways distinguishable by age group the increasing use of the Internet as a source for political information invites investigation of Market Penetration of the Internet It is no exaggeration to say that the Intern et is the fastest growing technology invented by man; that its growth rate far exceeds that of any previous technology B y way of comparison with other technological innovations Dessauer (2004 ) expla ined that t took electricity 50 years to reach 50 million users in the United States, whereas it took radio 38 years, it took computers 16 years, it took television 13 years, and it took the The geometric and freque ntly exponential growth curve of the Internet has been measured according to several criteria: numbers of users, size
28 of bandwidth, processing power, and functionality. In terms of providing information upon which political cognitions, attitudes, and behaviors are f o rmed, the rapid market penetration of the Internet has also been unprecedented. The growing importance of the Internet as a source of political information is illustrated by a comparison of sur vey results from October 2004, when 10% of the public mentioned the Internet as either their first or second most important source of political news to October 2008, when 33% of the public did so (Pew 2 008 a ). In fact, 2008 is the year that and t he Internet surpassed newspapers as an information source during presidential campaigns (Smith, 2009, p. 3) When looking more specifically at younger adults, survey results show that this age group is much more reliant on the Internet for political information than are older citizens (Pew, 2008 b ). While 2008 was her alded as the first year that a majority of Americans went online replaced newspapers as the third most popular news platform, behind local television news and national tele vision news Smith p. 1). In fact, at present, the Internet rivals all forms of television news as the main information source for younger adults Due to the nature of this new technology, however this shift in media reliance has not only allowed people to diversify their political information sources but it has also re shaped the relationship between the news media and its consumers. No longer required to assume a passive role when consuming information, at present Smith p. 3). That is, political information is sought when an event is
29 conveni ent. Thus, when online, media exposure and information gathering are not only active processes, but they are also interactive in that feedback and channels allowing for user content creation are much more immediately and widely available than among traditional media. Unlike the unidirectional, linear process of consuming news from the television or radio, the acquisition of political information from the Internet is ch aracterized as a social process With t he rising market p enetration and interactive nature of the Internet scholars today are called upon to re examine and adapt long standing theoretical frameworks to allow for more holistic analyses of media effects. Uses and Gratifications Theore tical Framework Origins of the uses and gratifications approach to media studies date back to (1948) functional approach to communications research. Around the same time as Lasswell, Lazarsfeld and Stanton (1942) also adopted this analytic framew ork in the ir re search on the motivations driving the spectacular demand for media content such as daily radio soap operas and newspapers. After the Columbia studies in Erie County Ohio during the 1940 presidential race and in Elmira, New York during the 19 4 8 race faile d to yield evidence of direct media effects in electoral decisions the Two Step flow model was developed to explain how interpersonal influence mediated the effects of exposure to the media (Katz & Lazarsfeld, 1955) Klapper (1963) f uel ed th i s shift away from the traditional persuasion model of communications research toward a m ore limited effects paradigm when he asserted that "what people do and can be brought to do with mass communication may largely determine what mass communication does a nd can be brought to do to people" ( p. 523 ). C redited with ushering in the cognitive revolution Klapper also argued that a more functional approach to media studies would
30 restore the audience member to ng him in the passive, almost inert role to which many older studies had relegated him 527). Thus, the uses and gratifications analytic framework views an active audience as a mediating factor in studies of media effects Emphasizing the functional uses" approach of media studies, Katz (1959) argued that scholars should ask "what do people do with the media?" rather than "what do the meaningless if no one is ex posed to or attends it, he advocated a "uses" approach to media studies that could better inform analyses of specific media channel c ontent and information consumption. Katz (1959) assumed that individuals' values, interests, and associations or so cial roles were the factors determining media. In this uses and gratification tradition, audiences are active: they are not merely receiving messages provided by the media, but they are also actively searching for information to meet specific needs ( Kaye & Johnson, 2002; Rubin, 1993 ). The uses and gratifications approach to analyzing the mass media focuses research on the psychosocial influences shaping ption gratifications theory is to explain the psychological motives shaping media usage in a be havioral attempt to fulfill intrinsic needs ( Rubin 1984 ). According to Katz, Blumler, and Gurevitch (1974), the basic assumptions underlying this theory are that media users are goal directed, active, and aware of their needs such that t hey purposefully select media channels and content to satisfy them. The concept of an active audience
31 has strengthened the basis for adopting a uses and gratifications analytical framework in research on Internet communicatio ns Online communication is distinct from traditional media communications in two ways: first, because it requires greater user activity to access specific content from a relatively unlimited menu of options ; and second, because each subsequent choice is highly contingent upon a series of earlier responses (Fredin & David, 1998). That is, even at its most basic level, accessing information on the Internet requires performing an active search and the user mu st a series of hyperlinks using the attendant technology. Since technological innovations and media fragmentation make it difficult for any one framework to explain audience formation across all channels, uses and gratifications scholars argue it may be premature to assess the impact of new media platforms prior a robust understanding of their usage Perse and Dunn (1998), for instance, suggest that using the uses and gratifications approach is appropriate for explain i ng changes in media use patterns upon introduction of new technology that displace s old media platforms. In contemporary settings, then, the scope of uses and gratifications analys e s of communications activities may also need to be narrowed from xaggerated emphasis on using mass media to meet social deficits, to the function it fulfills ). Additionally, such research may be best suited to analysis of specific channels of mediated communications, such as the Internet ( Newhagen & Rafaeli, 1996 ). While most uses and gratifications research is centered on analyses of the motivations shaping overall media usage ( Vincent & Basil, 1997 ), analyses of the needs driving consumption of specific content such as news (Levy,
32 1977 ) and politics ( Blumler & McQuail, 1968 ) have been identified as distinctly appropriate within this tradition. Typically research in the uses and gratifications tradition has focused on social and psychological origins of (2) needs which generate (3 ) expectations of (4) the mass media or other sources which lead to (5) differential exposure (or engaging in ( Katz Blumler & Gurevitch 1974 p. 510 ). Finn (1997 ), for instance, developed a five factor model of personality traits (extroversion, neuroticism, openness to experience, agreeableness, and conscientiousness) which motivated engagement in various communications activities (p. 508). While the results from the regression analysis in this study indicate d that examinations across all five personality trait dimensions and all media usage may yield less than robust results, this framework allowed for comparisons between patterns of engagement i n both mediated and non mediated or interpersonal communications activities. At present, however, uses and gratifications is animated by explor ations into the motivations for consumption of specific media content on specific media channels in an attempt t o narrow the scope of the analytic framework. While past uses and gratifications research has been oriented toward explaining the use of media as an alternative to interpersonal communication, the widespread adoption of computer technology and growth in use of interactive Internet sites has blurred the distinction between computer mediated and interpersonal communication (and) interactive links between uses or motivations, gratifications or consequences, and media and content need to be explored ( Rubin,
33 1981, p. 15 9). By examining affective orientations toward mediated communications, then, we may also provide insight into relationships between people and media channels. Finally the development and popularity of new media pl atforms such as social network sites ha ve re invigorated the uses and gratifications approach to media analysis as these Internet formats facilitate interactions not associated with traditional media usage ( Ruggiero, 2000 ). Online Informati on Formats Uses a nd Gratifications The nature of the Internet allows its users to do more than passively consume news in a one way flow of information. Large numbers of U.S. citizens are now using this media platform to become part of the online political participatory class. Not surprisingly, survey results reveal that the y oung are much more engaged in this Online political users under the age of 30 are much more likely than other age groups to: get customized politica l or campaign news; post their own original content online; (and) take part in political activities on social the age of 30 were significantly more likely than their olde users; watch online political videos; engage politically on a social networking site; post 14). The uses and gratifications approac h to media studies holds that pe rsonal goals, at least in part, shape individual patterns of media usage. Viewed as active communicators, m edia consumers are aware of their motivat io ns evaluate available media content, and then choose a channel from a wid e array of options to satisfy specific needs. Research in this tradition has shown that wide varieties of demographic
34 attitudinal, and behavioral characteristics are associated with patterns of media usage (Johnstone, 1974; Rubin & Rubin, 198 2 ; McLeod & B eck er, 19 81 ). These media usage patterns however, do not develop in a vacuum P re existing knowledge of media channel s, contents or format s affects perceptions and thus cognitions and attitudes in a measurable manner ity in satisfying specific needs is an important aspect of uses and gratifications research because p revious experience with media channels may shape perceptions of how well they fulfill various needs ( Becker & Schoenbach, 1989 ). Specifically research has shown that c ontinued use of specific media platforms over time to gratify specific needs may develop into expectations of the purpose of the information content (Perse & Courtright, 1993 ) I n line with schema theory these expectations may affect information processing because source knowledge influences how inf ormation is organized, categorized, and delimited Based on the perceived purpose, motivation, and past use in terms of need gratification the content, format, and preconceived purposes of different media channels may exert distinct eff ects (Axelrod, 1973; Graber, 1984). For the purposes of this study, the differential cognitive, attitudinal, and behavioral effects between those exposed to online news and social network sites is of particular concern. By limiting the focus of this inves tigation to two distinct online information formats, many of the criticisms of the uses and gratifications approach may be addressed. One of the criticisms leveled at uses and gratifications research on all types of media consumption is the requirement tha t the audience actively evaluate all available media content s and format s prior to selecting a specific channel. Some scholars argue
35 that this requirement places too heavy a burden upon individuals who may simply select specific media channels based on for ce of habit, with little thought given to all the available alternatives ( Donohew, Nair, & Finn, 1984 ). Since many rational behavior theori sts also maintain that humans are cognitive misers that use shortcuts or heuristics in decision making the explanati ons for patterns of media consumption by uses and gratifications research may neglect to account for the strength of habitual media usage (Severin & Tankard, 2001) T his criticism however, does not apply to the present investigation for two reasons. First experimental manipulation of media usage eliminates concerns over habitual behavior because participants are randomly assigned to treatment conditions. Secon d this potential shortcoming actually becomes a strength when analyzing the media usage patterns of citizens between the ages of 18 and 29 In fact, scholars ha ve long recognized the importance of studying the media habits of the young because it is during the young adult life stage that most habits in terms of media consumption, politic al information seeking decision making, and participatory behaviors are formed ( Henke, 1985; Faber, Brown, & McLeod, 1979; Rubin, 1985 ). Further, by limiting the focus of this research to analysis of the effects o f media exposure on young citizens during an election campaign, the results allow for comparison of gratifications expected in terms of the perceived purposes of two distinct online information formats and gratifications received in terms of influence on political cognitions and attitudes Other criticism s of uses and gratifications research lie at its functional core: a lack of predictive power based on theoretical principles Many scholars argue research using this analytic framework does not have the ability to do more than provide description s of
36 media use patterns among different groups of people making it difficult to extrapolate research results beyond the sample ( Baran & Davis, 2009 ). Similarly, some also claim that uses and gratifications research is purely descriptive, that it is limite d to providing typologies of media usage among specific demographic or ideological groups ( Swanson, 1977, 1979) As the analysis in this study is derived from an experimental investigation however, this criticism is not as compelling as it would be had th e analysis be en based on survey results. By adopting a functional approach in an experimental setting by exercising control over the exp osure t o the stimuli in the research laboratory, and by using random principles for participant assignmen t to one of the six experimental conditions most of these concerns are addressed. Finally, the larger purpose of experimental stud ies is theory testing thus the applicability of research results to larger populations is a secondary concern ( Shrout & Bolg er, 2002 ) Probably the most compelling criticism of uses and gratifications research is that its explanatory apparatus is frequently compartmentalized, individualistic, and lacks clarity and consistency in the use of central constructs such as motivation s, gratifications, and needs ( Elliott, 1974; Lometti, Reeves, and Bybee 1977 ) Indeed, it becomes apparent upon a review of the literature that there is a great deal of variation in the operationalization of the motivations shaping media usage T he most c ommon early classifications of motivations for media usage include d surveillance, correlation, enter tainment, and cultural transmis sion ( Lasswell, 1948; Wright, 1960 ) Depending on the scope and purpose of the study, however over time researcher s have dev eloped typologies that included d iversion, personal relationships, personal identity, and surveillance ( McQuail, Blumler, & Brown, 1972 ). More recently, communications
37 scholars in the uses and gratifications tradition have employed classifications of motiv ations for media us age that included surveillance, entertainment, escape, boredom, and vote guidance (Diddi & La Rose, 2006; Kaye & Johnson, 200 3 ; Rubin, 1983; Vincent & Basil, 1997). While these differences in classifications of media usage motivations m ay make meta analyses more challenging this study employs the uses and gratifications approach in three distinct ly specific ways. First, the theory of political information efficacy is placed in the larger uses and gratifications tradition by testing its predictive power in terms of other political cognitions, attitudes, and media use patterns. Second, the distinct purposes of different online information formats create the theoretical dichotomy between online news and social network sites so that the cogn itive and affective effects of these sources can be compared meaningfully Third, the criticism that uses and gratifications research neglects to examine gratifications obtained by usage of specific media channels is circumvented by an experimental design that both functionally defines interactivity and the measures the effects of exposure to the stimuli. Fourth, this analysis is focused on how exposure to different Internet formats impacts cognitive, affective, and behavioral variables in the context of a political campaign. Therefore, concerns about accounting for various motivations for media usage beyon d those that satisfy vote guidance needs are moot. Although few would dispute the purpose, function, and motivation for much media usage is entertainment, diversion, or emotional release, the more compelling question in this investigation is how exposure to different online information platforms impacts learning and attitudes in regard to participating in the political system.
38 Online News Research using the u ses and gratifications approach has a long tradition of examining the functions fulfilled by different news sources. Due to the physical variations am ong media channels, profound differences in content, and wide ranging disparities in effects, there has been a great deal of research into the factors driving individual usage of print and television news sources ( Althaus & Tewksbury, 2000 ; Becker & Dunw oody, 1982; Culber tson, Evarts, Richard, Ka ri n, & Stempel, 1994 ; Holbert, 2005 ). While past research has shown that surveillance needs appear to be driving most news media use, print media usage has been more closely associated with vote guidance needs and television news has been linked more closely with entertainment and escape needs ( ). In a time of media convergence, however, the distinction between print and electronic news has become much less meaningful especi ally considering the market penetration of computer technology and the Internet (Pew, 1998; 2006) Survey r esearch specifically focused on the motivations driving college students the news media has shown that th is group report s much less urgent su rveillance and vote guidance needs than other age groups ( Pew, 2010 a ) Additionally, research has shown that the y oung are also much more likely to consult Internet sources than traditional media sources both overall and in relative to those in other age g roups (Diddi & LaRose, 2006). Indeed, many college students today are digital natives; they are unable to recall a period in their lives when the Internet was not accessible and soon many will not be able to remember when they could not access the Intern et through a handheld device. These d igital natives may not be particularly interested in reading an entire news story. The young today are characterized as news grazers, scanning
39 headlines and activating links to stories about which they find personally compelling not necessarily those stories the news media place at the top of their agenda. This changing media landscape in addition to the wide spread use of interactive features have revitalized the uses and gratifications approach to new s media studies F unctional similarities between the television and the Internet have led some researchers to conclude that they may satisfy similar needs ( Kaye & Medoff, 2001 ) While o nline news w ebsites present stories in a more traditional media format with a narrative directing a uni directional flow of information, most also allow for some form of immediate feedback such as the posting of comments Many online news sites may also be linked to social network and bookmarking sites Thus, similarities in content notwithstanding, structural and technological differences between television and the Internet indicate that they may also satisfy distinctly different needs ( Kaye & Johnson, 200 3 ). In addition to satisfying surveillance and vote guidance needs in t he form of online news sites, there are also innumerable Internet sites dedicated to satisfying needs that are more personal Social Network Sites Social network sites have been defined as based services that allow individuals to (1 ) construct a p ublic or semi public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system Boyd & Ell i s on, 2008 p. 21 1 ). While the technology and terminology may differ across the various s ites, social network sites generally require users to register and answer a series of demographic and interest questions so that the answers can be used to build profile. Typical ly, u sers are also prompted to identify other members with whom
40 they may share a connection or relationship so that a list of friends, fans, or contacts can be developed These lists of online relationship s are usually visible to others who have access to ; t hus, social network sites are unique because they allow users to articulate and make visible their social networks (p. 213) Most sites also allow both public and private Beyond building a profile, locating friends, posting comments and sending private messages, the capabilities of the various social network sites may vary in terms of permitting photograph / video sharing and instant messaging. The history of social network sites extends back to the development and launch of SixDegrees.com in 1997. Considered ahead of its time, this site managed to attract millions of users, but the service shut d own in 200 0 A variety of other social network sites developed between 1997 and 2003, but most wer e geared toward niche users in terms of geographic location, common e thnicit ies and a focus on building business or dating networks ( Chafkin, 2007; Cohen, 2003) Although Friendster enjoyed a brief period of popularity in 2002 and 2003 technical difficul ties with site maintenance and a perception that the company and users were at cross purposes led to a mass exodus. This crossover occurred at about the same time that teenagers fueled the meteoric growth of MySpace in 2004 Facebook initially launched t o serve college networks beginning with Harvard in 2004 opened up its membership to the general public in 2005 and today about 40% of total Internet users log on to this site on a daily basis ( Alexa.com 2011 ). T he video sharing social network site, YouTu be was also launched in 2005, and experienced greater growth than MySpace over the same period of time. Currently You T ube draws about 26 % of global Internet users to the site on a daily basis
41 ( Alexa.com, 2011 ). At present there are literally thousands of social network site including Twitter, which launched in 2006 and draws about 10% of the daily Internet users (Alexa.com, 2011). Awash in a sea of social network sites, the two that stand out in terms of market penetration and potential for interactivi ty are Facebook and Yo uT ube Further, these two sites have consistently ranked among the most visited on the worldwide Internet worldwide for the past several years and are extraordinarily popular among our target population of college students ( Ale xa.com, 2011 ). Analyses orientations toward online information sites reveal both a preference for social network sites and a perception that these sites are more interactive than other sites ( Beer, 2008 ; Stromer Galley, 2000; Sweetser, & Weaver Lariscy, 2007 ; Trammel, Williams, Postelnicu, & Landreville, 2005 ). This preference for ( Raa c ke & Bonds Raa c ke, 2008 ). S ocial network sites allow the user to forag e for interesting topics and investigate those perceived to be the mo st personally appealing ; to watch videos; to activate links to other sites for different points of view and original sources. Further, the computer mediated communication between and among users facilitated by social network sites allows for testing the influence of different types of Interactivity The study of how people interact with online content extends back about a decade when scholars began systematically analyzing the mechanisms by which the Internet allowed for high speed, non hierarchical and horizontal communication; increased user input, feedback, or interactivity; and relati ve low cost as a digital reception and production medium ( Rheingold, 1995 ; Selnow, 1998). The nature of interactivity and its
42 use as a variable in empirical investigations, however, has led to considerable debate as to its definition and the manner in whic h it is operationalized see Kiousis, 200 2 ). S ome scholars have argued that interactivity can be analyzed in all types of communication settings and that it is measured by the extent to which sender and receiver roles are reciprocal in a co mmunication transaction ( Pavlick, 1983) These scholars view interactivity as a process related variable based on the contingency of messages relating to previous messages (Rafaeli, 1988; Rafaeli & Ariel, 2007). O thers claim interactivity is a quality of m ediated communication that can be measured on a by the medium ( Newhagen, 1998 ) Still others claim that interactivity is an inherent characteristic of new media technology that requires inclusion in studies of Internet Han s sen, Jan kowski, & Etienne, 1996 ; Williams Rice, & Rogers, 1988) Numerous sc holars have defined interactivity but three major approaches to this construct have developed over the past couple decades. First, some early studies adopted a perceptual view of interactivity and provided valuable information about the manner in which u sers evaluated their experiences with new media technologies and contents based on their perceptions of the transactions ( Wu, 1999 ). C ontinu ing to base creates difficulties in operationalizing experime ntal conditions based on differing levels of interactivity Although the perceptual approach to analyzing the effects of interactivity may continue to produce valuable results among heterogeneous populations who perceive objectively similar stimuli in much different ways, this current investigation uses a more
43 homogenous and technologically sophisticated sample as well stimuli familiar to the participants. Second, p roponents of the contingency view of interactivity hold it is a process related variable det ermined by the extent to which subsequent messages are contingent upon previous messages These scholars argue that interactivity may be measured in all types of communication settings ; that it is not a function of any specific technology (Rafaeli, 1988; R afaeli & Ariel, 2007). Indeed, Rafaeli ( 1988) provided one of the earliest and most widely cited definitions of interactivity. He argued that interactivity was n expression of the extent that, in a given series of communication exchanges, any third (or later) transmission (or message) is related to the degree to which previous (p. 11). Although defining interactivity as a process related variable has produced some very fruitful content analyses, the value of adopting a contingency view of this variable in an experimental investigation is much less informative relative to a more functional view of interactivity. Similar to the case when examining the perceptual view of interactivity, the classification of in teractivity levels in an experiment using the contingency view would require post hoc analysis to determine which interactivity level participants individually assigned themselves. Although this determination could be made based on the posttest questionnai re in a perceptual based experiment, a content analysis would be required to determine the level of interactivity in a contingency based experimental investigation. Last, the functional view of interactivity was adopted by s cholars who contend ed that inter activity wa s dependent upon the extent to which technology usage allows for two way communication These researchers argue interactivity is a medium specific
44 variable which affects cognitive, attitudinal, and behavioral variables Scholars taking the funct ional view question the value of determining the extent to which messages are contingent upon one another They argue that the process related approach to interactivity serves more as a basis for developing typologies; that it is not a theoretical proposit ion concerned with media effects ( Bucy & Ta o, 2007 ). Similarly, the perceptual view of interactivity has been credited with providing valuable information in the early phases of research into the effects of computer mediated communications, but such a view does not allow for experimental manipulation. Thus, t his study adopted a functional view of interactivity with three levels manipulated by the assignment of specific tasks to complete dependent upon interactivity level. 1998, Stromer Galley (2000) made a distinction between computer mediated human interaction and media interaction. In later work, Stromer Galley (2004) clarified this dis as process, which entails a as product, which entails a in interpersonal interactivity and content interactivity, other scholars have used a finer gradient in their definitions of interactivity Szuprowicz (1995) for example, identified user to user, user to documents, and user to computer activities as the distinguishing features of the three levels of interactivity. Similarly, Chung and Yoo (2009) delineated three levels of interactivity on a continuum: medium, human/medium, and human interactivity. Others have identified comparable and more fully explicated constructs
45 defining interactivity levels at a minimum of three levels (Barker & Tuck er, 1990; Haeckel, 1998; Jensen, 1998) Kayany, Wotring, and Forrest (1996) for instance, suggested that within these three types of interactivity users exert three types of control: relational (or interpersonal), content (or document based) and process/sequence (or interface based) controls. Based on her synthesis of these frameworks for analyzing different types of interactivity, McMillan (2002) developed a typology of interactivity that allows for distinctions based on the use of a es allowing for user to system, user to document, and user to user interactions. McMillan describes user to system interactivity as the interface between people and the computer itself She argues that the interaction between a si ngle human and a single computer is the most elemental form of interactivity. Research on user to system rface with the computer directly ; illumine s the processes by which people develop relationships with computers when utilizing them in a variety of ways, including searching for information. User to document interactivity describes the interaction between people and the content of a website. Research on this mid level for interactions with online content as well as the creation of content displayed on a web page. Feedback on webpage content in the form of comments and relationships among users and the information t hey consume are explored at the user to document level of interactivity. User to user interactivity is defined as the computer mediated communication (CMC) between people over the Internet. Research in the CMC tradition describes the interaction among huma ns as the highest level of interactivity and assumes that the computer and its networks serve primarily as a conduit or channel for
46 communication flow back and forth among any number of participants over time. Relationships between CMC and interpersonal co mmunication are explored at the user to user level of interactivity. researchers have described the evolution of these political information websites over time in terms of t he level of interactivity facilitated by their communications platforms. For instance, activities such as activating hyperlinks to access information and then consuming that information are considered forms of user to system interactivity. Examples of acti vities falling under the user to document interactivity classification Facebook pages. Last, forwarding news stories to others by email and re posting Facebook page are considered types of user to user interactivity. w eb sites as virtual billboards or online broch ure ware These websites provided little more than one information and issue stance s, information that was also available offline (Schneider use of blogs in its web Internet The Dean for America campaign was the first to utilize this medium features that provided the potential fo r human interaction as well as links to external informational web sites ( Stromer Galley & Baker, 2006). successful use of its web site for the mobilization of volunteers and solicitation of
47 donations has been documented in numerous scholarly works and set the standard for political candidate web s ites in the years to follow (Hindman, 2009; Jenkins, 2006). Indeed, the successful uses of the mobilization and donation functions on the 2008 web sit in 2004. In a broader meta analysis of 38 studies examining the relationship between evidence to support the a rgument that Internet use is contributing to urged future research to focus on political interest and the direction of causality in terms of Internet use and political engagement. In his investigation of the effects of w eb exposure and interactive features on the political information efficacy of youn g adults during the 2004 presidential election, (2007) results also support the prediction that interactive features on political information websites positively impact informed vote, or p olitical information efficacy, thus increa si ng their likelihood of participating in the political process. Further, the Internet, researchers have also found that the more interactive the website, the more favora ble the impression made by the sponsoring candidate ( Sundar, Kalyanaraman, & Brown, 2003 ).
48 Expression Effects R esults support ing the contention there is a positive associati on between interactivity and effects resulting from exposure to political information on the Internet suggest tha t the inclusion of expression effects in an alytical models may better inform a nalyses of computer mediated communications S tudies testing a bidirectional model of communication effects may also advance addressing the more interactive, de massified, and asynchronous media environment in which we presently find ourselves (Ru g giero, 2000) Moreover many scholars argue tha t replicable e xperimental investigations comparing the relative effects resulting from exposure to various online stimuli under conditions with differing levels of interactivity are the best method for test ing these effects ( C ho, et al., 2009 ; Kim, 2006 ) E fforts to explicate t he theoretical mechanism s underlying these effects however, ha ve only recently begun to be fully developed. Scholars at least since Soc rates have recognized that dialogic teaching methods compelling students to verbalize their reasoning processes are more effective than lectures involving solely a one way flow of information Thus, educators have long believed that greater cogn itive effects result from more interactive instructional t echniques More recently one of the earliest findings among media scholars was that interpersonal communication had a much greater impact on cognitions and attitudes than did exposure to mediated information from traditional media with a linear flow of information ( Lazarsfeld, Berelson, & Gaudet, 1948 ; Katz & Lazarsfeld, 1955; Klapper, 1960). With the development of digital media platforms, however, comparing the results of expos ure to different online information platforms may not capture the effects of interactivity in the transaction Since online media facilitate two way c ommunication and
49 user content creation researchers must control the manner in which participants interact with online information or they run the risk of deriving confounded results and spurious conclusions Ping ree (2007 ) argues that the expectation of future expression may be manipulated as an independent variable with participants randomly assigned to either control or experimental exposure conditions. H e predicts that those who expect to express themselves dur ing the exposure will pay closer attention to and process messages more elaborately tha n will those who do not expect they will have to express themselves Similar to the theory underlying the ela boration likelihood model, the cognitive and affective effects of information would be greater among those who expect to express themselves than among those who do not have such an expectation ve independent of the nature of the expected expression (p. 446). In addition to the expectation of future expression effects, model also includes compositio n effects and release effects. According to self perception theory, composition (or encoding) effects occur regardless whether a message is actually expressed because the simple act of composing the message may lead one to mak e inferences about his or her attitudes, traits, or feelings. Message release effects, on the other hand, may occur independently of composition effects dependent upon the type of message sent For example, r elatively anonymous comments posted on a webpa ge in the user to document condition may not involve be perceived as making a personal commitment and the re posting of previously composed
50 material on a webpage would require no original composition. Thus, while the release of the message may have consequences, the nature of the message may affect the extent to which social commitment, ego involvement, or feeling heard is engaged scholars have also proposed a theoretical framework for analyzing the effects of relationships between e lite driven campaign c ommunications and political outcomes ( Huckfeldt, Sprague, & Levine, 2000; Just et al., 1996; Page, 1996; Pan, Shen, Paek, & Sun, 2006 ). By re conceptualizing citizens as active processors of information, this line of research suggests that communication among citizens may significantly mediate the effects of news consumption on political participation, particularly when the communication occurs over the Internet ( Shah, Cho, Eveland, & Kwak, 2005) These researchers theorize that comput er mediated communications may encourage greater elaboration during interactions with political information pl atforms on the Internet and mediate cognitive and behavioral effects. T he O S O R model is the theoretical basis supporting scholars contentions that computer mediated communication s may mediate the effects of information exposure on resulting behaviors Borrowed from social psycholog ists Markus and Zajonc (1985 ), t he first O in th is model stands for O rganismic ( or pre O rientations ) and represents an attributes that affect information processing during exposure to a stimulus the S in the model. The second O stands for post Orientation s or the beliefs and attitudes formed after exposure to the stimulu s These post Orientations influence the resulting behavioral response s the R in the model. Recently,
51 many p olitical communication scholars have amended various elements of this model to better specify the manner in which intervening variables may influence post exposure orientations and thus, the resulting beliefs, attitudes, and behaviors. Researchers have suggested, for example, that the direct effects approach to the analysis of political campaign communications may be too simplistic. S uch an approach it is a rgued, neglect s to account for the manner in which these communications may stimulate information seeking and expressi ve behavior, influencing the ir response s to the information ( Cho, et al, 2009). When investigating the effects of exposure to p olitical information over the Internet, a functional approach to interactivity allows experimental researchers to measure the main effects of this variable. That is the quantity and quality of interactions with the c ommunication s technology, the content displayed and communication with other individuals may be controlled during exposure to the stimulus by randomly assigning participants to specific experimental conditions. By manipulating interactivity as an independent variable, this appro ach eliminates the need to analyze these behaviors in terms of their mediating. Thus using a functional approach allows for examination of main effects based on th e O S O R communication mediation model since both frameworks hold that increasing interactiv ity through expression should magnify cognitive and attitudinal exposure effects. Hypotheses and Research Questions I nformation costs are clearly greater in midterm elections than in presidential races, but the ben efits , are cited by rational behavior scholars as an incentive for participati ng in the political p rocess. Additionally use of features l essen ing the costs of
52 gathering information on new media platforms such as the Internet are worthy endeavors in and of themselves But first this research attempts to strengthen the theory of political information efficacy by empirically testing its power to predict relationships among other cognitive and affective political variables as well as media usage patterns Considered broadly, the concept of efficacy, or be accomplish a task, has been conceptualized as one of the self esteem needs in ( Bandura, 1977; 1982 ) In regard to political attitudes and behaviors, efficacy has also been linked to the need for personal control over environment when political events are salient in his or her life (Renshon, 1974). Thus, levels of political information efficacy may be treated as a personal ity characteristic that is predictive of o ther cognitive affective, and behavioral traits Formally stated, these predictions are : H1: Those with high levels of political information efficacy will express more interest in politics than will those with low levels o f political information efficacy. H2: Those with high levels of political information efficacy will have higher levels of political knowledge than will those with low levels of political information efficacy. H3: Those with high levels of political i nformation efficacy will report stronger poli tical party affiliations than will t hose with low levels of political information efficacy. H4: Those with high levels of political information will report more civic participation than will those with low levels of political information efficacy. When placing the theory of political information efficacy in the br oader uses and gratifications tradition, its predictive power in terms of media usage may also be measured. Intuitively one woul d expect t hose high in political information efficacy to use media for vote guidance information in a more goal directed manner than those with
53 low levels of political information efficacy. Specifically, this need for efficacy shoul d motivate media usage such that political information efficacy levels predict both overall and specific Internet usage in terms of fulfilling the vote guidance f unction. This postulation leads to the last prediction associated with political information e fficacy as well as a research question. H5: Those with high levels of political information efficacy will report using all media sources for vote guidance more often than will those with low information efficacy. RQ1: How well does political information efficacy predict usage of traditional media and Internet sources to fulfill vote guidance needs? By adopting a functional approach to the manipulation of interactivity this research is also in line with the uses and gratifications analytical framework. Although traditional uses and gratifications research has probed for relationships between individual characteristics and the amount of time spent with different media channels and content, communicat ions over interactive, digital media platforms require more specific examinations of what people actually do with the media. Not only is the amount of time behaviors du ring that time may also affect resulting responses to the exposure. While both uses and gratifications and the O S O R model allow scholars to predict that increasing levels of interactivity will result in increasing cognitive and affective effects, they a lso allow for the examination of the main effects of the online format on cognitions and attitudes Therefore, the following hypotheses are posed based on a pertinent review of the literature. H6: The main effect of format will result in t hose exposed to news websites repor t ing network sites.
54 H7: There will be a positive association between interactivity level and amount of issue learning. H8: The interaction between format and interactivity will result in those exposed to news websites in the user to user condition report ing the greatest issue learning. H 9 : The main effect of format will result in t hose exposed to social network sites report ing greater image learning than those exposed to news websites. H10: There will be a positive association between interactivity level and amount of image learning. H 11 : The interaction between format and interactivity will result in those exposed social network sites in the user to user cond ition reporting the greatest image learning. H 12 : The main effect of format will result in those exposed to news websites reporting greater gains in political information efficacy than those exposed to social network sites. H 13 : There will be a positive association between interactivity level and gains in political information efficacy. H1 4 : The interaction between format and interactivity will result in those exposed to news websites in the user to user condition reporting the greatest gains in political information efficacy. The experimental design in this investigation allows for comparison of the relative cognitive and affective effects of exposure to online news and social network sites under three distinct interactive conditions in the context of the 2010 gubernatorial campaign in Florida. One of the benefits of placing this investigation in the context of this election was to allow for exploration of sources to satisfy vote guidance needs. In an attempt to gauge the manner in which exposure to these stimu li may affect voting behavior evaluations of the candidate evaluations and perceptions of the saliency of the upcoming election may also be measured across both online informa tion formats as well as interactivity levels. At present, there is no prior research comparing the relative effects of exposure to these
55 sources on affect toward candidates or perceptions of the election to make predictions thus research ques tions are more appropriate When looking beyond the campaign to those who actually participate in midterm elections, some data is available that may better inform analyses of the electoral outcome. Results from political behavior research suggest that thos e who participate in midterm elections typically form a subset of presidential election voters and they are distinctly different from those who do not vote or participate in politics (Wolfinger, Rosenstone, & McIntosh, 1981). When focusing specifically on electoral contexts similar to 2010, researchers have also found that midterm elections held two years after the election of a Democratic president are frequently more conservative than those who do not participate in the process ( Kernel 1977; Abramowitz & Saunders, 1998). Thus, examining not only the effects of exposure to the stimuli, but also the effects of political party affiliation on affect toward candidates and perceptions of the importance of the upcoming election may provide some valuable informat ion. Further, early research on the effects of exposure to mediated information on voting intentions are a guide suggest that the results of such an investigation into individual characteristics such as political party identification may also better explai n the electoral outcome ( Campbell, Co nverse, Miller, & Stokes, 1960 ). Therefore, in addition to testing for main and interaction effects among online format and interactivity levels, this investigation also attempts to gauge the effects of individual char acteristics such as political party affiliation and candidate preference on affect toward the election itself by asking four more research questions. RQ2: How do online format and interactivity level affect candidate evaluations? RQ3: How does political party affiliation affect candidate evaluations?
56 RQ4: How do online format and interactivity level affect the saliency of the election? RQ5 : How do political party affiliation and candidate preference affect t he saliency o f the election?
57 CHAPTER 3 METHOD Participants and Procedure Between October 11 14, 2010, the hypotheses were tested and the research questions were answered using a between participants pretest posttest two (news websites a nd social media sites) by three (user to system; user to document; and user to user interactivity level) experimental design Th ese four days were selected for experimentation as they fell in the last weeks of the 2010 midterm campaign during the hot phase of 2010 gubernatorial election when media coverage and strategic campaign communications efforts began to peak Tw o hundred and ninety one participants were recruited from undergraduate mass communications courses fulfilling general education requirements at a large southeastern research institution. The participants were offered extra credit for par ticipation ; were randomly assigned to one of the six conditions ; and complete d the project in the research laboratory. All participants complete d a n online pretest questionnaire that include d measures of political information efficacy, political knowledge, political interest, strength of political party affiliation civic participation ; m edia usage ; and demographic and candidate evaluation items. Upon completion of the pretest, participants were direct ed to one of the six conditions, exposed to the stimulus, and instructed to engage in activities appropriate for the level of interactivity to which they were assigned. After completing the tasks in the assigned condition, participants were directed to the online posttest que stionnaire that included items reassess ing political information efficacy and candidate evaluation s, measur ing issue and candidate image learning, and perceptions of the saliency of the upcoming election
58 All participants were instructed that they were required to remain in the research laboratory for a minimum of 60 minutes and a maximum of 75 minutes, regardless of how long they required to complet e the project Participants were also instructed that should they fail to follow all experimental direction s carefully the MediaLab software would automatically delete their responses and they would not receive extra credit for participation. Two monitors remained in the back of the research laboratory at all times to supervise participants and to answer any questions about the directions. Following completion of the project, participants were debriefed and dismissed Stimuli News Websites Four types of n ews websites were selected for presentation to participants: an In t news aggregator, a newspaper, a television station, and an online only nonpartisan news source. After a search for objective and even handed stories on each of these site types was conducted, a Yahoo News story; a St. Petersburg Times story; a story from the NBC affiliate in Miami; and a Grist story were selected (see Appendix A ) P articipants were instructed to spend a minimum of 10 minutes interacting w ith each news source by completing tasks appropriate for their randomly assigned condition or level of interactivity T he order of the online news stimuli presented to participants was randomized by Qualtrics software Social Networking Sites Facebook YouTube channel home page were selected for presentation to participants (see App endix A ) Participants were instructed to spend a minimum of 10 minutes interacting with each source by completing tasks appropriate for their randomly assigned level of interactivity
59 T he order of social ne twork site stimuli presented to participants was randomized by Qualtrics software. Variables Levels of Interactivity Theoretical analyses have categorized Internet communications activities i nto one of the three levels of interactivity : user to system user to document, and user to user. In order to assure the functional validity of these determinations this investigation also conducted a survey administered to a sample of 100 college student cohorts to determine which activities are classified as falling into each of the three interactivity levels. Based on th e results of this pretest survey participants in each experimental condition were instructed to engage in activities identified as meeting the criteria for each level of interactivity both theoretically and functionally. These results of this pre experiment survey and assignment of activities appropriate to each interactivity level are provided in Appendix B Participants in the all of the interactivity conditions were instructed to spend a minimum of 1 0 minutes interacting with each of the four randomly assigned and ordered websites Each of these websites provide d information about the two major party candidates in the 2010 Florida gubernatorial contest In order to keep track of the time spent on each website, participants noted the time they entered each site and the time exited each site (from the numerical display on the bottom right hand corner of the computer screen) on the document provided and located beside each computer keyboard
60 Political Information Efficacy was measured in both the pretest and posttest questionnaires. As in previous studies ( i .e., Kaid, et al, 2007 ; Tedesco, 2007), four items from the American National Election Study (ANES, 2009) survey were used to construct the political information efficacy index The items include d : (a) I consider myself well qualified to participate in politics, (b) I think I am better informed about politics and government than most people, (c) I feel that I have a pretty good understanding of the important political issues facing our country today, and (d) If a midterm elections, I feel I would have enough information to help my friend figure out who to vote for. Participants rate d their level of agreement with each statement on a five point Likert scale. This scale received a 0 .86 in the pretest and 0.87 in the posttest. Political Interest was measured in the pretest questionnaire Items used to measure political interest included : (a) If I had more free time, getting involved in some political activity would be high on my list; (b) I would rather do something that is detached from p olitics than something that requires me to engage in politics (reverse score); (c) The idea of participating in some political activity appeals to me; (d) I want to know the details about current political events. Participants were asked to rate their leve l of agreement each items on a five point Likert scale (Shani, 2010) This scale received a Cronbach P olitical Knowledge was measured in the pretest questionnaire administered using eight questions. The first five questions were drawn
61 from the American National Election Studies (ANES, 2 009) questionnaire and Delli Carpini and Keeter (1993) : (a) Do you happen to know what job or political office is now held by Joe Biden? (b) What are the first ten amendmen ts to the U.S. Constitution called? (c) How much of a majority is required for the U.S. Senate and House to override a presidential veto? (d) Which one of the parties is more conservative than the other at the national level? (e) Whose responsibility is it to nominate judges to the Federal Courts ? Since college students would be expected to score higher on these questions than the general public, however, t hree more items were added to this scale: (a) Do you happen to know who is Majority Leader in the U.S. Senate? (b) With which party is Alex Sink affiliated ? (c) With which party is Rick Scott affiliated ? Political Party Affiliation In the pretest, p articipants were asked to identify their political party affiliation as : (a) Strong Republican; (b) Weak Republican; (c) Independent who leans Republican; (d) Strong Independent; (e) Independent who leans Democratic; (f) Weak Democrat; (g) Strong Democrat; or (h) Other (Converse & Pierce, 1985) Civic Participation Eighteen i tems asking participants whether they had ever engaged in specific activities tion in the pretest questionnaire. These items included to : voter registration; frequency of voting in past elections; intentions to vote in the upcoming midterm election; past and prospective campaign volunteer work; donations of money or time to organizat ions with political goals; membership in political and social groups or nongovernment organizations ; attendance at political functions or meetings ; political campaign rally attendance; and participation in organized boycotts or protests.
62 Political Information Sources P all mediated political information sources to fulfill vote g uidance needs were measured in the pretest questionnaire. Participants were asked to indicate how frequently they use 16 media sources to gather information about political candidates : local TV news, national TV news, TV talk shows, morning TV news shows, and satirical news shows; how frequently they read the newspaper and news magazines listen ed to radio news or political radio talk shows discuss ed politics with others and consult ed news websites, blogs, Twitter Facebook, YouTube and other online sources. Candidate Evaluation Participants complete d candidate evaluation items in both the pretest and posttest quest ionnaires. These items include d a feeling thermometer ranging from 0 to 100 Participants will be asked to indicate how unfavorable cold/favorable warm they felt toward the can didates. This thermometer is similar to the one traditionally used by the National Election Studies to measure attitudes toward the candidates (Rosenstone Kinder, & Miller, 1997). Saliency of Election Participants were asked to state how salient they perceived the 2010 Florida governor elect ion using six items previously shown to be valid measures of involvement (Zaichowsky, 1985). The items included: the upcoming election has prominent value in society; the upcoming election has significant valu e in society; the upcoming election has important value in society; the upcoming election is well known in society; the
63 upcoming election has fundamental value in society; and I am concerned about the upcoming election. ha score of 0 .81. Learning Participant reported amount of learning about the (a) candidates and (b) were measured using a five point Likert scale in the posttest questionnaire.
64 CHAPTER 4 RESULTS Demographics and Manipulation Check Before proceeding to the data analysis testing the hypotheses and answering the research questions, some descriptions of the participant s ample and a check on the manipulation of interactivity are in order. T wo hundred and ninety one undergraduate students participated in this experiment but t hirteen entries were deleted due to missing data so 278 responses were analyzed. Sixty six percent of participants were female and all were between the ages of 18 and 2 9. Sixty percent of participants were non Hispanic Whites; 1 7 % were Hispanic; 9 % were African American ; 6% were Asian American; 3% were multi T hese demographic characteristics albeit marginally higher for females are very similar to those reported by the UF Student Health Center in their 2010 Healthy Gators survey report when removing graduate and professional students from that sample. In terms of pol itical party affiliation s 30% ident ified as Republican; 33% as Independent; and 37% as Democratic Both the ethnic characteristics and political party affiliation s of the participants in this investigation and those reported by the Pew Research Center (2010 a ) in their analysis of the millen nial generation are also nearly identical. While some of these demographic characteristics were shown to affect the dependent variables in the analys e s to follow however, there were no significant differences among experimental conditions in the distribut ion of gender x 2 (2, 278) = 1.0, P > .05; ethnicity, x 2 ( 1 2, 278) = .356, p > .05; or political party affiliation, x 2 ( 4 278) = .692 p > .05. Thus, the principles of random assignment should account for any influence exerted by demographic variables ac ross the six conditions.
65 The manipulation of interactivity was measured in terms of self reports of the number of activities performed by participants assigned to each condition The differences be tween the three conditions were significant: u ser to system ( M = 37.41, SD = 3.21) ; user to document ( M = 38.26, SD = 2.36) ; and user to user ( M = 40.95, SD = 3.55), F = 23.526, df = 2, p < .01. Thus, t he manipulation of the three levels of interactivit y level was satisfactory Predictive Power of Political Information Efficacy The first hypothesis predicted that those with high levels of political information efficacy would express more interest in politics than would those with low levels of polit ical information efficacy. The results of a bi v ariate correlation analysis reveal ed there wa s a moderate ly strong, positive, and significant association between levels of political information efficacy from the pretest questionnair e ( M = 12.14, SD = 3.46) and political interest ( M = 12.21, SD = 3.32), r (276) = .453, p < .01. Based on this result, further investigation was conducted by creating a dichotomous variable for political information efficacy using a median split technique : 123 participants with scores between 4 and 11 were designated as having low and 155 participants with scores between 12 and 20 were designated as having high levels of political information efficacy This median split technique was also u sed to test predictions regarding differences between those with high and low levels of political information efficacy in hypotheses two through five As shown in the first row of Table 4 1, the results of an analysis of variance re veal ed that those with high levels of political information efficacy expressed significantly more interest in politics ( M = 13.10, SD = 3.14) than did those with low levels of political information efficacy ( M = 11.18, SD = 3.20), F (1, 276) = 27.94, p < .0 1. Thus, hypothesis one was strongly supported.
66 The second hypothesis asserted that those with high levels of political information efficacy would have higher levels of political knowledge than would those with low levels of political informa tion efficacy. The correlation between political information efficacy and political knowledge ( M = 5.87, SD = 1.39) wa s moderate, positive, and significant, r (276) = .30, p < .01. Additionally, a s shown in row two of Table 4 1, t he results of an analysis of variance reveal ed that the difference in political knowledge between those expressing high levels of political information efficacy ( M = 6.10, SD = 1.39) and thos e expressing low levels of political information efficacy ( M = 5.56, SD = 1.33) wa s significant, F (276) = 10.67, p < .01. These results provide d strong support for hypothesis two The third hypothesis stated that those with hi gh levels of political information efficacy would report closer political party affiliations t han would those with low levels of political information efficacy. To test th is prediction, the political party affiliation variable was re coded to create a strong partisanship variable. T hose who identified as strong Democrat s strong Republican s or strongly Independent were coded 3 ; those who identified themselv es as weakly affiliated with the Democratic or Republican parties were coded 2 ; and those who identified themselves as Independent leaners toward either party were coded as 1 The correlation between poli tical information efficacy and partisanship s trength was moderate positive, and significant, r (276) = .263, p < .01. Further, a s shown in the third row of Table 4 1, t he results of an analysis of variance reveal ed that those with high levels of political information efficacy identifie d their political party affiliation ( M = 2.05, SD = .82) significantly more strong ly than did those
67 with low levels of political information efficacy ( M = 1.85, SD = .68), F ( 1,276) = 4.24, p < .05. Thus, hypothesis three was strongly supported. The fourth h ypothesis predicted that those with high levels of political information efficacy would report more civic participati o n than would those with low levels of political information efficacy. The c orrelation between political information efficacy and civic participation was moderately strong, positive, and significant r ( 276) = 480, p < .01. Further a s shown in the fourth row of Table 4 1, t he results of an analysis of variance reveal ed that t hose with high levels of political information efficacy report ed significantly more civic participat ion ( M = 17.35 SD = 1 7 0 ) than did those with low levels of political information efficacy ( M = 1 5.30 SD = 1 6 5 ), F (1, 276) = 5 1 74 p < .0 1. Th ese result s provide d strong support for hypothesis four The fifth h ypothesis predicted that those with high levels of political information efficacy would report using the media for vote guidance more often than would those with low levels of political information efficacy. The correlation between political information efficacy and media us ag e for vote guidance was moderately s trong, positive, and significant, r (276) = .314, p < .01. Further, the results of an analysis of variance reveal ed that those with high levels of political information efficacy ( M = 46.59, SD = 11.74) reported using all media sources to g ather political information significantly more often than did those with low levels of political information efficacy ( M = 40.35, SD = 11.13), F (1, 276) = 20.56, p < .01 Thus, hypothesis five was s trongly s upported. T o identify more specific ally which media platforms were used the first r esearch q uestion asked how well political information efficacy levels predicted use of traditional media and Internet source s to s atisfy vote guidance needs. To answer this question,
68 two dependent variables were created The traditional media variable was calculated by summing responses to items asking how often : (1) local television news ; (2) national broadcast news ; (3) cable news ; and (4) newspapers were consulted for vote guidance The Internet sources variable was created by summing responses to items asking how often : (1) news we bsites (2) social network sites ; (3) campaign w ebsites and (4) other online sources were consulted for vote guidance. Using traditional media usage as a dependent variable, a multiple regression model was constructed with tradi tional media usage as the dependent variable. P olitical interest, knowledge, participation information efficacy ; partisan strength ; and two demographic variables were entered as inde pendent variables. The demographic variables accounted for ethnicity and gender differences through the creation of a dichotomous dummy variabl e for Caucasian ( with non Hispanic Whites coded 1 and all others coded 0 ) and Female (with females coded 1 and males coded 0 ) As shown in Table 4 2, t he results of th e regression analysis revealed that the model albeit significant, did not account for much of the variance in the us ag e of traditional media sources, R 2 adj = .0 7 F (7, 268 ) = 3. 94 p < .01. Further, political information efficacy was not a significant predictor of traditional media use for vote guidance = .10, t (268) = 1.4, p > .05 Alternately, a regression model was constructed using Internet source s as the dependent variable with the same independent variables used in the previous analysis As displayed in Table 4 3 t he results of th e regression analysis show ed that this model although not particularly strong, still accounted for more of the variance than did the traditional media model R 2 adj = 21 F (7, 2 68 ) = 9.35 p < .01 Further, not only was
69 political information efficacy a significant, albeit moderate, predictor of Internet use, but it also had the largest marginal impact of all the independent variables in the model in terms of the frequency that Internet source s were consulted for vote guidance, 2 4 t (268) = 3 70 p < 0 1 Thus, political information efficacy predicts Internet use moderately well but it is not a good predict or of traditional media use to satisfy vote guidance needs. Online Forma t and Interactivity Effects The sixth h ypothesis predicted that t hose exposed to news websites would report greater issue learning than would those exposed to social network sites. The results of this analysis are displayed in the first three rows of Table 4 3. First, an overall issue learning variable was created by summing score s to questions asking how much participants learned about the issues important to Rick Scott and Alex Sink. The results of an analysis of variance show ed that those exposed to news websites ( M = 8.30, SD = 1.49) reported that they learned signi ficantly more about the issues important to both of the candidates than did those exposed to social network sites ( M = 7.22, SD = 1.91), F (1, 276) = 27.10, p < .01. Second the amount of issue learning by each candidate and online format was examined in dividually T he results of an analysis of variance show ed that those exposed to news websites ( M = 4.07, SD = .88) reported learning significantly more about the issues important to Rick Scott than did those exposed to social network sites ( M = 3.62, SD = 1.05) F (1, 276) = 15.18, p < .01. Similarly, those exposed to news websites ( M = 4.23, SD = .78) reported learning significantly more about the issues important to Alex Sink than did those exposed to
70 social network sites ( M = 3.60, SD = 1.17), F (1, 276) = 28.91, p < .01. These results provide d strong support for hypothesis six. The seventh h ypothesis asserted that there would be a positive relationship between interactivi ty level and issue learning. The results of this analysis are displayed in the first three rows of Table 4 4. W hen looking at total issue learning, t he results of an analysis of variance reveal ed that the differences among those in the user to use r ( M = 8.0 5 SD = 1.65), user to document ( M = 7.83, SD = 1.86), and user to system ( M = 7.34, SD = 1.93) interact ivity levels were significant, F ( 2, 276) = 3.83, p < .05. The differences among interactivity levels in learning about the issues important to Rick Scott, however, not significant, F (2, 276) = .71, p > .05. I n regard to learning about th e issues important to Alex Sink on the other hand, the differences among those in the user to user ( M = 4.14, SD = .87); user to document ( M = 3.96, SD = 1.09); and user to system ( M = 3.61, SD = 1.11) conditions were both significant and in the predicte d direction, F (2, 276) = 6.82, p < .01. T hese results provide d moderately strong support for hypothesis seven. The eighth h ypothesis declared that those exposed to news websites in the user to user condition would report t he greatest issue learning. The results of an analysis of variance however, reveal ed that the interaction between format and interactivity level on total issue learning wa s not significant, F ( 2, 276) = 1.56, p > .05. Similar to the results from the previous hypothesis, w hen looking at each candidate, however, a mixed picture emerge d The interaction between format and interactivity on learning abou t the issues important to Rick Scott wa s not significant, F (2, 276) = .91, p > .05. In regard to learning about the issues important to Alex Sink, on the other hand the results of an
71 analysis of vari ance reveal ed that the interaction between format and interactivity wa s significant, F ( 2, 276) = 4.23, p < .05. As shown in Figure 4 1, however, closer ins pection revealed that the differences in learning about the issues important to Alex Sink were only significant a mong those exposed to social network sites F(2, 276) = 8.24, p < .01. Therefore, these results failed to provide support for hypothesis eight because the interaction effect occurred among those exposed to social network sites not online news in the user to user condition The ninth h ypothesis stated that t hose exposed to social network sites would report greater image learning tha n would those exposed to news website s. The results of this analysis are displayed in the last three rows of Table 4 3. When examining total image learning, t he results of a n analysis of variance show ed that those exposed to social network sites ( M = 7.63, SD = 1.87) reported learning marginally more about the did those exposed to news websites ( M = 7.3 4 SD = 1.86), but that this difference was not significant, F (1, 276) = 1.79, p > .05. When looking at each candidate, however, a familiar pattern emerge d In regards to Rick Scott the difference in image learning betwee n those exposed to social network sit es ( M = 3.69, SD = 1.10) and those exposed to news websites ( M = 3.36 SD = .99) was not significant F (1, 276) = .165, p > .05. The difference in learning about Alex Sink image qualities, on the other hand between those exposed to social network sites ( M = 3.94, SD = 1.03) and those exposed to news websites ( M = 3.71, SD = 1.0) was significant, F ( 1, 276) = 3.97, p < .05. These results provided partial support for hypothesis nine.
72 The tent h hypothesis predicted that there would be a positive relationship between interactivity level and amount of candidate image learning. As shown in rows four through six of Table 4 4 however, t he result s of a n analysis of variance test revealed that the differences in learning about the image qualities of both candidates were not significant, F ( 2, 276) = 1.5 3 p > .05. W hen examining learning about the image characteristics of each candidate individually, the differences between those in the social network sites and online news conditions in learning about the image qualities of Rick Scott, F (2, 276) = 2.63, p > .05, and Alex Sink, F (2, 276) = 2.0, p > .05, were also not significant. Thus, hyp othesis ten was not supported. The eleventh hypothesis declared that those exposed to social network sites in the user to user condition would r eport the greates t image learning. The results of an analysis of variance showed that the differences in total image learning, F (2, 276) = .55 p > .05 and F (2, 276) = .30, p > .05, however, were not significant. As shown in Figure 4 3, on the other hand, when examining the amount of learning about the image qualities of Alex Sink, there was a significant interaction effect in the predicted direction F ( 2, 276) = 3.39, p < .05. Thus, t hese results provide d partial support for hypothesis eleven. The twelfth hypothesis asserted that those exposed to news websites would report greater gains in political information efficacy than would those exposed to social network sites. First, as shown in Table 4 5, the results of a paired samples t test reveal ed that the difference s between pretest political inf ormation efficacy ( M = 12.14, SD = 3.46) and posttest political information efficacy scores ( M = 13.36, SD = 3.23) w ere significant across all conditions t (277) = 9.36, p < .01 Second, a PIE change
73 variable was created by subtracting pretest political information efficacy scores from pos t test scores. Third, an analysis of variance was conducted comparing the mean PIE change values between the two conditions, but the results reveal ed that the difference between those exposed to news websites ( M = 1.10, SD = 2.37) and those exposed to social network sites ( M = 1.33, SD = 1.99) was not significant. Thus, hypothesis twelve was not supported. The thirteenth hypothesis stated that there would be a pos itive relationship between level of interactivity and gains in political information efficacy. The results of a n analysis of variance, however, revealed that differences between those in the user to user ( M = 1.48, SD = 2.04); user to document ( M = 1.13, SD = 2.57); and user to system ( M = 1.07, SD = 1.82) were not significant, F( 2, 276) = .964, p > .05. Thus, hypothesis thirteen was not supported. The fourteenth hypothesis predicted that th os e exposed to news web sites in the user to user con dition would report the greatest gains in political information efficacy. Graphically displayed in Figure 4 3 the results of an analysis of variance test reveal ed that there wa s a significant interaction between format and interactivity on political infor mation efficacy gains, F (2, 276) = 3.79, p < .05. Specifically among participants exposed to news websites this analysis re veal ed that the differences in political information efficacy among those in the us er to user ( M = .93, SD = 2.00); user to document ( M = 1.48, SD = 3.11) ; and user to system ( M = .91 SD = 1.81) conditions were not significant. Among participants exposed to social network sites, on the other hand the differences in political information efficacy gains among those in the user to user ( M = 2.05, SD = 1.94) ; user to document ( M = 1.22 SD = 1.84 ) ; and user to
74 sy stem ( M = 0.85 SD = 2.01 ) conditions were significant, F ( 2, 141) = 4.73, p = .01. Thus, while the interaction between online information format and interactivity level was significant, hypothesis fourteen was not supported because the effect on political information efficacy gains was observed among those exposed to social network sites not online news sites. The second research question asked how online format s and interactivity levels affected candidate evaluations. First, as shown in Table 4 6 the differences between between online formats were examined. T he results of paired samples t thermometer scores were significantly higher in the postte st ( M = 58.47, SD = 26.18) than in the pretest ( M = 42.68, SD = 20.81) across formats and interactive conditions, t (277) = 11.33, p < .01. actually decreased from the pretest ( M = 43.28, SD = 20. 94) to the posttest ( M = 40.39, SD = 27.29), and this difference was also significant, t (277) = 2.17 p < .05. Second, an evaluation change variable was created for each candidate by subtracting pretest feeling thermometer scores from posttest thermomete r scores. Then all negative evaluation change values were converted to positive numbers to calculate the mean evaluation change scores by candidate accurately A total evaluation change variable was then c alculated by summing the candidate individual fe eling thermometer change scores. W hen examining the changes in evaluations of Rick Scott, t he results of an analysis of variance test showed that the main effect of format F (1, 276) = .042 p > .05 ; the main effect of interactivi ty, F ( 2 276) = .2 4 4, p > .05 ; and the interaction effect between the two F (2, 276) = .307, p > .05 were not significant Similarly, in
75 regard to changes in evaluations of Alex Sink, the main effect of format F ( 1 276) = .318 p > .05 ; the main effect of interactivity F (2, 276) = .4 7 9 p > .05 ; and the interaction between the two F( 2, 276) = .276, p > .05 were not significant. Finally, in regard to total changes in candidate evalua tions, the main effect of format, F ( 1 276) = .63 3, p > .05; the main effect of interactivity, F (2, 276) = .544, p > .05; and the interaction between the two F (2, 276) = .049, p > .05, were also not significant. Therefore, changes in both individual and aggregate candidate evaluations were statistically equivalent across online format s and interactivity level s ; resulting in positive outcomes for Alex Sink and negative outcomes for Rick Scott. Partisanship, Candidate Preference, Saliency of the Election The third research question asked how political par ty affiliation affected changes in candidate evaluations. Similar to the analysis in the previous research question, the differences between each candidate were examined. In this case, however, the net gain s in candidate evaluations, instead of the absolute changes in evaluations, were examined by participant s political party affiliation. The results of an analysis of variance test showed that the net gains in Rick were si gnificantly different dependent upon whether the participant identified as a Democrat ( M = 12.79, SD = 20.0), an Independent ( M = 2.76, SD = 21.36), or a Republican ( M = 10.02, SD = 18.47), F (2, 268) = 29.07, p < .01. Along the same lines, the net gains in Alex Sink were also significantly different dependent upon whether the participant identified as a Democrat ( M = 25.31, SD = 20.38); an Independent ( M = 17.08, SD = 23.93); or a Republican ( M = 3.21, SD = 19.60), F (2, 268) = 24.07, p < .01. Thus, while participants political party affiliation exerted a significant effect on net gains in feeling
76 thermometer scores Alex Sink scores increased across all three part y categories while and Indepe mean evaluations of Rick Scott decreased The fourth research question asked how online format and interactivity levels affected perceptions of th e saliency of the election. The results of an analysis of variance showed that the effects of format, F (2, 276) = 1.23 p > .05; interactivity level, F (2, 276) = 1 75, p > .05; and the interaction between the two, F (2, 276) = 1.52, p > .05, were not signif icant. Thus, it would appear that these variables exerted no influence on perceptions of the saliency of the election. The fifth research question asked how political party affiliation affected perceptions of the saliency of the election. The results of an analysis of variance revealed that the Republicans ( M = 32.74 SD = 4.95); Independents ( M = 30.32, SD = 6.49); and Democrats ( M = 31.18, SD = 5.03) were significant, F (2, 268) = 4.14, p < .05. T he sixth research question asked how candidate preference affected perceptions of the saliency of the election. First, when asked for who m would they vote if the election were held today, across all experimental conditions, 32% indicated they would vote for Rick Scott ; 59% for Alex Sink; and 9% for neither. T he results of an analysis of variance on the other hand, revealed that those who preferred Rick Scott ( M = 32.89, SD = 4.90) perceived the election to be significantly more salient than did those who voted for Alex Sink ( M = 30.80, SD = 5.08) and those who voted for n either ( M = 28.27, SD = 7.05), F (2, 276) = 8.16, p < .01. Thus, those who perceived the election to be the most salient were those who s tated they would vote fo r Rick Scott
77 Table 4 1. Differences in political c haracteristics by information efficacy levels Low PIE (n = 123) High PIE (n = 155) F value df p Political interest 11.08 13.10 27.94 1 .000 Political knowledge 5.56 6.10 10.67 1 .001 Par tisan strength 1.85 2.04 4.28 1 .039 Civic participation 15.30 17.35 51.74 1 .000 Table 4 2 Predictors of traditional media use for vote guidance B SE t p (Constant) Political Interest 15.93 0.20 3.4 0.09 0.16 4.64 2.28 .00 .02 Political Knowledge 0.15 0.19 0.05 0.79 .43 Strong Partisans 0.64 0.53 0.07 1.01 .23 Civic Participation 0.26 0.11 0.16 2.32 .02 Political Information Efficacy 0.13 0.09 0.10 1.40 .16 Caucasian 0.11 0.51 0.01 0.21 .83 Female 0.25 0.53 0.03 0.47 .64 (n = 278; Model R 2 adj = 0.07 ) Table 4 3 Predictors of Internet usage for vote guidance B SE t p (Constant) Political Interest 8.03 0.08 2.31 0 6 0 0 08 3.47 1. 29 .00 20 Political Knowledge 0 22 0 13 0 10 0 .69 0 8 Strong Partisans 0.05 0 36 0 08 1 41 16 Civic Participation 0.21 0 08 0 .8 2. 71 01 Political Information Efficacy 0 19 0 06 0 2 4 3. 7 0 .0 0 Caucasian 0 20 0 35 0 03 0 3 2 57 Female 0 44 0 36 0 .0 7 0 6 9 22 (n = 278; Model R 2 adj = 0.21 )
78 Table 4 4. Learning by format, type, and candidate Social Network Sites (n = 144) News Websites (n = 134) F value df p Issue Learning Scott Sink 7.22 3.62 3.60 8.30 4.07 4.23 27.10 15.18 28.91 1 1 1 .00 .00 .00 Image Le arning Scott Sink 7.63 3.69 3.94 7.34 3.36 3.71 1.79 0 .165 3.97 1 1 1 .1 6 .59 .05 Table 4 5 Leaning by interactive condition, type and candidate S yst em (n =90) Document (n = 98) User (n = 90) F v alue df p Issue Learning 7.34 7.83 8 .05 3.63 2 .0 3 Scott 3.73 3.87 3.90 .792 2 .49 Sink 3.61 3.96 4.14 6.26 2 .00 Image Learning 7.26 7.72 7.47 1.53 2 .2 2 Scott 3.73 3.87 3.90 .712 2 .49 Sink 3.6 7 3. 87 3.94 1.81 2 .16 Table 4 6 Exposure effects on political Information efficacy by format and condition Social Network Sites ( n = 144) News Websites ( n = 134) Pre test Post test Pre test Post test User to System( n = 90) 11.56 12.78* 11.95 12.86* User to Document ( n = 98) User to User ( n = 90) All Conditions ( n = 278) 12.70 11.82 12.07 13.56* 13.86* 13.40* 11.93 12.72 12.21 13.41* 13.65* 13.31* *p < .01 Table 4 7 Online format effects on candidate evaluation Sink Scott Pre test Post test Pre test Post test Overall ( n =278 ) 42.68 58. 47 43.28 40.39* SNS ( n = 144) 41.9 7 55.73 42.55 4 1 .03 News ( n = 134) 43.45 61.42 44.06 38.63* *difference between pre test and posttest feeling thermometer is significant at p < .01.
79 Figure 4 1. es by format and condition Figure 4 2. Learning
80 Figure 4 3. Political information efficacy gains by format and condition
81 CHAPTER 5 DISCUSSION Predictive Power of Political Information Efficacy In prior research, the theory of political information efficacy has been used to explain differences in political information processing among age groups and to gauge the effec ts of exposure to mediated information T he purpose underlying the first four hypotheses in this study on the other hand, was to provide the first empirical testing of the predictive power of the theory of political information efficacy The results of th e all the hypotheses testing a s well as the answers to all the research questions are summarized in Table 5 1. As shown in the first four rows of this table, the analysis revealed that levels of political information efficacy are not only closely associate d with other cognitions and attitudes such as political knowledge, interest and strength of partisan affiliation but that they are also predictive of the frequency of behaviors in the form of civic participation. T hus, these results enhance the value of this analytic framework b y measuring the predictive power of political information efficacy as an independent variable. To strengthen the predictive power of this analytic framework in terms of patterns of media usage the theory of political information efficacy was placed in the larger uses and gratifications tradition to determine how those who feel most efficacious or confident in their ability to participate in politics use the media. This rationale allows for test ing the theory bility to predict m edia usage to fulfill vote guidance needs. While t he results from hypothesis five show ed that the most efficacious use all media more than those who are less efficacious do, the answers to research question one are mixed. As shown in rows five and six of T able 5 1, the analysis show ed that levels of
82 information efficacy are not predictive of traditional media usage, but that they are a moderately strong predictor of Internet use to satisfy vote guidance needs. T h is finding is not terribly surprising given c college students are digital natives, they generally prefer Internet sources for political information gathering due to their familiarity with, ease of access to, and preference fo r th is platform relative to traditional media channel s. Overall, then, t hese results show that the theory of political information efficacy predicts Internet use moderately well, but that it is not a good predictor of traditional media use to satisfy vote guidance needs. In this study, the need for political information efficacy is conceptualized as precipitating levels of political interest, knowledge, attitudes and behaviors. The correct ordering of these variables, however, is not th e sole or primary objective of this investigatio n In fact, m ost rational behaviorists would probably c ontend that the effects of these variables are both reciprocal and recursive (Finkel, 1985, 1987 ) Thus, while it is eminently valuable to test t he predictive power of relatively new political communication theories the analytic limitations of results based on self reports, correlations, and analyses of variance tests using a median split technique may beg the question of cause and effect. Indeed, po litical scientists have long recognized that levels of political interest, knowledge, partisan affiliation and civic participation co vary (Campbell, Converse, Miller, & Stokes, 1960 ). While none of these results, therefore, is terribly surprising, an intuitive outcome does not lessen the strength of th ese findings i n both explaining and predicting other political cognitions, attitudes, and behaviors. In truth, the biggest revelation emerging from this first set of results is probably that
83 political inf ormation efficacy is a moderately strong predict or of Internet usage, but that it is not a good predictor of traditional media usage to fulfill vote guidance needs. Cognitive Effects of Online Format and Interactivity Levels In hypotheses six through eleven, two online information formats and three level s of interactivity were m anipulated as independent variables with issue and image learn ing serving as dependent variables A summary of these findings is presented in rows seven through twel ve of Table 5 1. Overall, the analysis shows that exposure to news web sites resulted in differentially significant learning about the issues important to both candidates and that exposure to social network sites resulted in differentially significan t learning about image qualities. The se findings link directly back to a u ses and g ratifications view of online political information formats (i.e., Ancu & Cozma 2009 ; Abrahamson 1998 ; Donohew, Finn, & Christ, 1988 ) Although exposure in the experimental setting is compelled rather than a natural choice, the nature of the site should reflect the need it is designed to gratify. Social networking sites, for instance, are used mainly for interpersonal interaction and keeping up with the activit ies network channel that one would be more inclined to watch videos, read comments by other users, and look for personal information; and one might be less inclined to read i social network sites Conversely, news web sites are typically used to satisfy surveillance needs in terms of issue information and vote guidance needs rather than for socialization or entertainment. In this case, it seems natural that one would lapse into an information gathering mode while interacting with news web sites.
84 Alternately, in regard to image learning, the uses and gratifications roles of these sites are reversed. Again, social network sites app ear more personalized and prompt experiences. Alternately, n ews web sites do not necessarily highlight personal o r image information and users of such sites may have their attention m ore focus ed on issue information. Further, it must be noted that controlled, whereas stories on a news web site are not. In fact, the campaigns themselves played a particularly strong role in the fi ndings of this study. While we those exposed to social networking sites, the difference in learning about Rick image characteristics between those exposed to news and those exposed to social network sites was not significant T his finding is directly attributable to the style of campaign run by each candidate. Typical of many insurgent Republican candidates in the 2010 midterm electio ns, was less about describing who he wa s, and more about declaring who he wa s not. Throughout the campaign, Scott at tempted to build support among those who fe lt a growing dissatisfaction with the current national administration, and he positioned himself as an alternative to both career politicians and Obama Democrats (Erickson, 201 0 ) He avoid ed describing himself personally, and instead attacked his opponents and the current system ( Krakauer, 2011 ) In fact, whereas the s Facebook and provides personal background information about the candidate Facebook page does not
85 contain similar information. It seems clear that in terms of image learning the campaigns themselves played a sig nificant role. When measuring the impact of interactivity on issue learning, campaign strategies also influenced the se findings in terms of both main and interaction effects First, the differences among interactivity levels in overall issue learning as we ll as learning about the issues important to Rick Scott w ere not significant Second, the differences among interactivity levels in learning about the issues important to Alex Sink were significant and i n the predicted direction. Thus, those in the user to user interactivity condition reported they learned significantly more about the issue s important to Alex Sink than did those in the user to system interactive condition Third, the interaction between online format and interactivity level was also only si gnificant concerning learning about the issues important to Alex Sink This interaction effect, however did not occur among those exposed to news websites as predicted but among those exposed to social network sites. Thus, the use of interactive features had a significantly greater impact on learning about the issues important to Alex Sink among those exposed to social network sites, but not among those exposed to news web sites. Similarly while interactivity level did not exert any significant main effe cts on candidate image learning, there was an interaction between interactivity level and online format in learning about image qualities among those exposed to social network sites. Once again, the Sink campaign appeared to capitalize on the p erceived purpose of the Internet format as well as the needs it fulfill s by making effective use of interactivity on the candidates social network sites Since the concept of interactivity springs from analyses of interpersonal communication s, a context i n which participants switch back and forth from sender to
86 receiver roles throughout the ir transaction, it seems natural that the greatest effects of interactivity would be found among those exposed to sites designed to satisfy social utility needs. Online Format and Interactivity Effects on Information Efficacy Another noteworthy finding that ran counter to the prediction was that exposure to both social network sites and online news sites result ed in significantly higher levels of political information eff icacy While greater gains in political information efficacy were predicted among those exposed to news web sites, significant and statistically equivalent gains across both online information formats were actually observed A s shown in rows thirteen throu gh fifteen of Table 5 1, none of the hypotheses predicting greater effects on political information efficacy among those exposed to online news sites were supported A lthough not predicted th e s e finding s also provide further support for the weight of the evidence from the data analysis; that interactivity exerts distinct effects on specific online channels. Additional reasons for these null findings may also be due to low levels of information about the candidates in a midterm election. In such a case, exposure to any information, regardless of source, may elevate ability to participate meaningfully in politics most expensive gube rnatorial campaign Further, t he participants were college students who as noted in the results section, correctly answered an average of 5.86 out of 8 political knowledge questions Thus, the sample was much more knowledgeable about politic al institutions, events, and actors than the general public Yet, th ese par ticipants who are relatively sophisticated in terms of both politic al knowledge and usage of digital media reported increases in their levels
87 of political information efficacy that were statistically equivalent across both online information formats This result speaks volumes about the power of social network sites to fulfill vote guidance needs among college students and digital natives Whil e politics, especially outside of presidential elections, may frequently fail to register on the perceptual radar screen of many young citizens, those who interacted with the fi cant increases in th eir efficacy levels that equaled the increases among those exposed to news websites. Although there were no main effects of interactivity on political information efficacy observed in this investigation, there was a significant interaction between interac tivity level and online format A mong those exposed to news web sites the differences in political information efficacy gains by interactive condition were not significant Among those exposed to social network sites, on the other hand, interactivity leve ls did exert a significant influence in the predicted direction. At this point, it would appear that the effects of interactivity may not only be dependent upon campaign strate gies, but also upon the online information format. That is, the nature of social network sites may facilitate interactivity as well as make this online format more amenable to the effects resulting from the use of these features. Again, the finding that interactivity exerts greater influence on social network sites than on news web sites may be explained by the uses and gratifications uniquely fulfilled by the specific online platform. Further, according to the communication mediation model, ego involvement, sense of feeling heard, and social commitment levels may be greater wh en expression occurs over social network sites since such expressions may be more publicly visible on a more platform that is also more important to the participant T his
88 type of expression then, may exert greater influence on cognitions, attitud es, and behaviors. Specifically, past research has shown that citizens who feel their voices are heard, that their interests are at stake, and who express their support of ( or opposition to) a candidate, party, or issue are much more likely to participate in politics than are those who do not feel heard, involved, or committed (Finkel 1985; Gerber & Green, 2000). Candidate Evaluations and the Saliency of th e Election Practically speaking, the results of this study suggest that it is important for political campaigns to make use of both the conventional campaign function in terms of providing information subsidies through traditional channels and an expansion in the use of emergent social network sites As shown in Table 5 1, the answers to research questions two and three suggest that the Scott campaign was unable to improve candidate evaluations in any condition. The Sink campaign, on the other hand, was able to improve her evaluations across online formats, interactivity levels, and participants political party affiliations. Thus, even the argument that the news media was biased against Scott cannot explain why exposure to his campaign ites failed to elevate his evaluations his opponent. Unfortunately for Sink, however, her campaign was unable to increase the perceived saliency of the election. Regardless of how well a campaign cultivates a positive evaluation of a candidate, should those responding favorably fail to sense any urgency to participate, then the outcome is far less than assured. Indeed, by falling just 60,000 votes short of a victory, the low rate of turnout may hav e been critical in the election outcome.
89 Although this investigation importance in the pretest questionnaire, it is clear from the last three rows of Table 5 1 that supporters of Rick Scott felt much greater ur gen c y about participating in the election than did those favoring Alex Sink. Although midterm wins by the party in opposition to a first term president have become a part of a familiar cycle the Republican insurgency wins across the country was significan tly greater in 2010 than any over the past generation Indeed, the outcome of the 2010 elections was described by the Pew Research Center for the People and the Press (2010 b ) of the status quo no consensus about future policies; GOP early poll results revealed that the difference in candidate preferences between likely voters and the general public were significant and stable throughout t he campaign (p. 4) Thus, the inability of the Sink campaign to motivate citizens favorable to her candidacy was fatal. Indeed, the electorate in this contest was described as significantly older, more affluent, and more conservative than both the 2008 vot ing pool and the general public (p. 3). Additionally, the effectiveness of social network sites shown here may be particularly important to political campaigners because of its dual nature both passive and active. While one has to intentionally search for a traditional campaign Website, a advertisement on the site. This type of passive introduction to the campaign reduces the cost of seeking out information, and may be persuasive if recommended by a an action he or she
90 may not have taken without prompting. On the other ha nd, social network sites can also be very active prompting a person to comment and engage in the social nature of the format. Conventional campaigning wisdom holds that this small change from passive to active information processing also makes one more l ikely to become engaged with the campaign by contribut ing or volunteering. Theoretical Implications Not only do the findings about the differential exposure effects of online information formats and interactivity levels have notable practical implications for both professional political and media operatives, but they may also inform analyses of the theoretical parameters of information efficacy. Functionally a set of constructs that are linked together by relational statements that are internall y consistent with each other the most common type of theoretical statement is a prediction that 2 constructs will covary information efficacy, for example, predicts that efficacy levels are posit ively associated with other political cognitions, attitudes, and behaviors. As operationalized in this study, political information efficacy is distinguished from internal political efficacy by one item his or her ability to describe the important issues and candidacies at stake in the upcoming election, rather than perform in public office. Since the utility driving research on this theory is that higher levels of information efficacy are associated with both the number of participatory acts as well as purpose of this investigation was to provide support for this theory. The results showed that efficacy levels are indeed a moderately good predictor of levels of political interest, knowledge, party affiliations, and participation. Future researchers may wish to refine
91 the other three items used to measure levels of political information efficacy so that they are distinct f In addition to testing the predictive power of political information efficacy theory, exposure to different types of formats can, in fact, be an antecedent for enhanced levels of political information efficacy McKinney, Tedesco, 2007, p. 1098). In line with uses and gratifications research, information efficacy theory also fulfills an explanatory function by informing analyse s of the differential effects of exposure to various media content and channels. Previous research has shown that young citizens, more so than older citizens, use issues over personal qualities in their evaluations of political candidates (p. 1101). The re sults of the present inquiry, on the other hand, reveal that increasing the interactivity on sites designed to satisfy social utility needs results in issue learning equivalent to those exposed to online news sites designed to fulfill surveillance and vote guidance needs. Further, individual level gains in political information efficacy resulting from exposure to social network sites are not only equivalent to, but they may also exceed the gains experienced by those exposed to online news sites when compute r mediated communication between users occurs. By expanding the experimental design to measure the effects of interactivity during exposure to different online platforms, this present inquiry has advanced research on political information efficacy by more precisely accounting for the effects of what people actually do with the media. While previous studies measuring the effects of exposure to t elevised political advertisements and debates on efficacy levels have revealed differences between demographic and ideological groups, they have also shown that exposure to Internet
92 sources stimulates information seeking and interpersonal communication more so than does exposure to traditional media channels (Kaid, 2002). When looking beyond studies comparing the expo sure effect of the same content on different media channels, however, t here is a dearth of experimental investigations comparing the differential effects of online information formats and interactivity levels extant in the literature. It seems cl ear that, at least for young voters, traditions of uses and gratifications of political information sources should be revised to inform our under stand ing of the power of social network sites more robustly Young people use the Internet for most of their in formation gathering, and the use of social network sites has become a daily habit in the lives of most young people. In fact, some college students may consult sites such as Facebook for political information more often than news web site s All of th ese factors lend support to the conclusion that social network sites have a profound ability to The rise in the use of social network sites to satisfy needs related to not only soc ial utility, but also surveillance, entertainment, and escape needs illustrates the broader blurring of the distinction between information sources and entertainment witnessed across media channels. For example, the television audience viewing T he D aily S how may be primarily motivated by entertainment needs, but the gratifications obtained from watching the show almost certainly also include surveillance, information, and vote guidance need fulfillment. When looking at computer mediated communications, on the other hand, the uniquely interactive aspects of this media channel have facilitated use of its features to gratify needs to an extent previously achieved only by interpersonal communication. In terms of the uses and gratifications
93 analytical approach, the purpose of media usage and the level of interactivity in the communication interaction have been shown to impact attitudes and learning in predictable ways. T he effects of exposure to political information over I nternet channels in this study are quite compelling. R anked as a major source of mediated po litical information the interactive nature of computer mediated communication as well as t he wide spread use of social network sites have also made this medium a functional alternative to much f ace to face communication the most important and influential source of politica l information T he results of this study also provide a starting point f rom which we may inform analyses of the impact of computer mediated political information on young One of the most important distinctions of this study, however, is the use of both a functional a pproach to interactivity combined with analysis focused on the effects or the gratifications obtained from exposure to experimental stimuli. Ironically, the results in this study disallowing rejection of the null hypotheses may provide some of the most co mpelling implications in terms of delimiting the theory of uses and gratifications when used to analyze interactivity effects on specific Internet platforms or formats. Put simply, the level of interactivity during exposure to online news sites exerted no significant effects on political cognitions, attitudes, or behaviors. This null finding means that none of the differences among the points on the lines representing the news websites condition in Figures 4 1, 4 2, and 4 3 is significant. Although the effe ct size of interactivity among those exposed to social network sites was large enough to produce a significant main effect on issue learning, the relatively
94 stable line representing those exposed to news websites in the Figure 4 1clearly shows that the eff ect in this condition is not significant. The effect size of interactivity on image learning among those exposed to social network sites, however, was not large enough to produce a significant main effect as postulated in hypothesis ten. Graphically presen ted in Figure 4 2, this finding means that the only point on the lines in Figure 4 2 that is significantly different from the others is the one representing those in the user to user condition exposed to social network sites. Similarly, none of the differe nces among the points representing mean gains in political information efficacy on the lines in Figure 4 3 is significantly different, except the one representing those in the user to user interactivity condition exposed to social network sites. The lack of support for a main effect of interactivity on image learning and political information efficacy suggests that the effects of what people do with the media may be limited by the specific channel format with which they are interacting. Although Internet u se requires more audience activity than traditional media usage, these findings suggest that higher levels of interactivity may significantly increase the effects of exposure to social network sites on cognitive, affective, and behavioral variables. Specif ically, while the lack of a significant effect of interactivity among those exposed to news websites may mask its effects among those exposed to social network sites, the interaction between interactivity level and online format exerts a significant effect among those exposed to social network sites in the user to user condition. Thus, the social network sites in the condition requiring the greatest interactivity and those in the same conditio n exposed to news websites was equivalent. Even more significantly, those in the highest
95 interactive condition exposed to social network sites reported greater image learning and gains in political information efficacy than those in any of the other condit ions. These findings, while not consistently in line with the hypotheses, nonetheless remain within the uses and gratifications tradition. As explicated by Katz Blumler & Gurevitch ( 1974 ), uses and gratifications research appropriately investigates the differential effects elicited by exposure to resulting in need gratification and other ( p. 510 ). While the purpose and design of news websites would suggest that those exposed to this online information format would report greater gains in issue learning and levels of political information efficacy, the results of this investigation may social net work sites. Indeed, uses and gratifications research is particularly useful in laying out the framework for analyzing the differences between gratifications sought and gratifications obtained. Much of the criticism leveled at uses and gratifications resea rch focuses on the typologies used to classify motivations for media use. Not only are these typologies imposed across all types of media consumers who may perceive the same media content as fulfilling different needs but many also argue that they limit t he analysis to the results of an artificially imposed schematic. By focusing on the gratifications obtained, however, the results of this investigation support the argument that the process of information gathering among digital natives may involve the use of media channels primarily designed to fulfill social utility needs.
96 Although the uses and gratifications typology allowed for predicting the direction of interactivity effects based on the function of the online format the results suggest that the fun ctional approach to interactivity allows for the use of models for analyzing expression effects Indeed, the findings in this study show that interactivity and expression effects are greatest on social network sites, even when focusing on cognitive variab les such as learning about the issues important to a candidate. Specifically, the interaction between online format and interactivity consistently occurred among those exposed to social network sites in the user to user condition. Not only was this interac tion was observed in regard to issue learning, however, but it was also observed in the analysis of exposure effects on image learning and political information efficacy. These findings suggest that expression is distinctly powerful when it occurs over soc ial network sites. Limitations Despite theoretical and practical implications, this study has some notable limitations. First, participants were encouraged to interact with the web sites to w hich they were exposed. Although they were instructed to explore the site to which they were assigned, they were also able to explore link s from site to a news article or vice versa. In addition by using real candidate s ocial network sites and real news web sites this study was forced to give up some control over content. These differences in content, however, are the cost paid for gaining insight into the effects of computer mediated political information in present tim e. Finally, this study highlights the power of the political information on the Internet, and in particular social network sites. Social network sites appear to act as familiar, reliable, and credible portals to political information for young citizens. No longer is
97 Internet campaigning simply brochure ware; candidates are beginning to make full use appealing to young people. Additionally, unlike the passive role re quired to consume political information through traditional media channels, the Internet requires more active ly involved users. In addition to testing the differential effects of interactivity on these web sites, future research may also explo re how this increased level of activity among Internet users affects source credibility. Clearly, there is also the potential for stronger impacts on knowledge levels, attitudes, and behaviors among those making use of the various platforms availab le on this new communications medium than among those exposed to other channels as well.
98 Table 5 1. Summary of findings Hypotheses/Research Questions Result s H1: PIE predicts political interest Strongly supported H2: PIE predicts poli tical knowledge Strongly supported H3: PIE predicts partisan strength Strongly supported H4: PIE predicts civic participation Strongly supported H5: PIE predicts media use for vote guidance Strongly supported RQ1: Does PIE predict t raditional media and Internet use for vote guidance? PIE does not predict traditional media, but does predict Internet use H6: Online news exposure results in greater issue learning Strongly supported H7: Positive relationsh ip between interactivity and issue learning Moderately strongly supported H8: Interaction occurs in online news and user to user interactivity condition Not supported H9: Social network site exposure results in grea ter candidate image learning Partially supported H10: Positive relationship between interactivity and image learning Not supported H11: Interaction occurs in social network sites and user to user condition Partially supported H12: Exposure to online news results in greater PIE gains Not supported H13: Positive relationship between interactivity and PIE gains Not supported H14: Interaction occurs in online news and user to user condition Not sup ported RQ2: How do online format and interactivity affect candidate evaluations? Uniformly positive for Sink and negative for Scott across conditions RQ3: How does political party affiliation affect candidate evaluations? Across cond itions and affiliations, improved only for Republicans RQ4: How do online format and interactivity affect No effect RQ5: How did party affiliation affect perceptions of Republicans perceived the election to be significantly more salient RQ6: How did candidate preference affect the Significantly greater saliency among those preferring Rick Scott
99 APPENDIX A INTERACTIVE ACTIVITIES PRETEST Instructions When using the Internet to find information about current events and politics, people consult a wide variety of sources and engage in many different types of communications activities. For exam ple, some people may simply read online news stories on websites while others may post their comments about the story on the website, forward the story to others by email, or post information about campaigns and elections on social network sites such as th eir individual wall on Facebook Scholars have classified the range of activities people may engage in when interacting with online websites into three categories ranging from the least to the most interactive: user to system, user to document, and user to user levels of interactivity. User to system interactivity describes the interface between people and the computer itself and this interaction between a single human and a single computer is considered the most basic form of interactivity. These types o f activities involve the information requested from a menu of many options. Most scholars consider this type of activity to require the least amount of effort of the user who inte racts with the computer to access the information and then merely reads or views the documents or videos provided. User to document interactivity describes the interaction between people and the content of a website as it is displayed on the computer scree n. Research on this mid as well as the actual creation of content itself as displayed on the web page. Feedback
100 provided by readers/viewers on webpage content and rel ationships between users and the information they consume are explored at the user to document level of interactivity. User to user interactivity describes the computer mediated communication (CMC) between people over the Internet. Research in the CMC trad ition describes the interaction among humans as the highest level of interactivity and ass umes that the computer and its networks serve primarily as a conduit or channel for communication that may flow back and forth among any number of participants over t ime. Key concepts include feedback between Internet users and relationships between CMC and interpersonal communication are explored at the user to user level of interactivity. Below is a list of communications tasks that people may complete when looking f or political information over the Internet. Please indicate which level of interaction you think each activity should be classified into: 1. User to system interactivity : people use their computer to access and consume information. 2. User to document interactiv ity : people access and consume information and t hen provide feedback on the website about the content of the information. 3. User to user interactivity : people access and consume information and then engage in communication with other people about the info rmation over the Internet. Results The results from 100 undergraduates, who were directed to classify each of the following activities into one of the three levels of interactivity, were used to calculate mean scores for the following activities. Based on the mean scores, activities fell into the three levels as follows (mean scores in parentheses) : User to User Interactivity 1. Forwarding news st ory to another person by email ( 3 .0 )
101 2. Contacting or sending a candidate an email (2.92) 3. Forwarding a video to another person to watch (2.87) 4. Send candidate or campaign a message through a website (2.75) 5. Facebook (2.6) 6. Re Facebook wall post to your Facebook wall (2.49) 7. Post informat ion about a candidate on your Facebook page (2.47) User to Document Interactivity 8 Re Posting a news story on your Facebook page (2.37) 9 Friending a candidate on Facebook (2.35) 10 Commenting on a video on a website (2.16) 1 1 ed to Facebook (2.14) 1 2 Facebook wall post (2.10) 1 3 Com menting on news story on websit e (2.0) 1 4 Signing up to receive text messages from candidates or campaign organizations (1.98) 1 5 Becoming a follower of a candidate, journa l ist, or news service on Twitter (1.92) 1 6 Signing up to receive emails from candid ates or campaign organizations (1.88) 1 7 Bookmarking, tagging, and linking news stories to Delicious or other so cial bookmarking W eb services (1.81) 1 8 Subscribing to YouTube channel (1.75) User to System Interactivity 19 2 0 photo galleries ( 1.22)
102 2 1 R eading about political candidates on Facebook (1.18) 2 2 Watching a video on a website (1.12) 2 3 2 4 Reading about current events on n ews w ebsites (1.0) 2 5 Reading about political candidates on thei r campaign websites (1.0)
103 APPENDIX B PARTICIPANT INSTRUCTIONS Condition One: User to System News Websites Instructions This project requires that you visit four news w ebsites that provide information about the candidates in the Flori weeks on November 2, 2010. You must spend 10 minutes reading the information on EACH website or on websites hyperlinked to it. You will need to note the time you entered the site (from the bottom righ t hand corner of the screen) and note the time you left the w ebsite on the paper provided. DO NOT post any comments on any website or forward any information to anyone. Condition Two: User to Document News Websites Instructions This project requires t hat you visit four news w ebsites that provide information weeks on November 2, 2010. You must spend 10 minutes reading the information on EACH website or on websites hyperlinked to it. You will need to note the time you entered the site (from the bottom right hand corner of the screen) and note the time you left the website on the paper provided. This project requires that you spend the 10 minutes on each website doin g more than just reading information, watching videos, or following any hyperlinks. In addition to consuming the information, you also must interact with the documents. Specifically, you must do a minimum of two of the following activities : 1. Re P ost a news story to your Facebook page 2. Comment on a video Can be positive, negative, or neutral.
104 3. Comment on a news story Can be positive, negative, or neutral. 4. Facebook page. 5. Become a follower of a news feed on Twitter. 6. Bookmark, tag, and/or link news story to Delicious or other social bookmarking W eb services. DO NOT forward any information to anyone. Condition Three: User to User News Websites Instructions T his project requires that you visit four news websites that provide information weeks on November 2, 2010. You must spend 10 minutes reading the information on EACH we bsite or on websites hyperlinked to it. You will need to note the time you entered the site (from the bottom right hand corner of the screen) and note the time you left the website on the paper provided. This project requires that you spend the 10 minutes on each website doing more than just reading information, watching videos, or following any hyperlinks. In addition to consuming the information, you also must interact with others Specifically, you must do a minimum of two of the following activities : 1. Forward News story to another person by email. 2. Forward a video to another person by email. 3. Recommend news story linked to your Facebook page. 4. Contact or send a candidate an email. 5. Post information about a candidate on your Facebook page.
105 Condition Four : User to System Social Network S ites Instructions This project requires that you visit four social networking websites that provide information about the candidates in the Florida governor held in a few weeks on November 2, 2010. You must spend 10 minutes reading the information or watching videos on EACH website or on websites hyperlinked to it. You w ill need to note the time you entered the site (from the bo ttom right hand corner of the screen) and note the time you left the website on the paper provided. Please spend 10 minutes on each website reading information, watching videos, or following any hyperlinks DO NOT post any comments on any website or forw ard any information to anyone. Condition Five: User to Document Social Network S ites Instructions This project requires that you visit four social networking websites that provide This election will be held in a few weeks on November 2, 2010. You must spend 10 minutes reading the information or watching videos on EACH website or on websites hyperlinked to it. You will need to note the time you entered the site (from the bottom right hand corner of the screen) and note the time you left the website on the paper provided. This project requires that you spend the 10 minutes on each website doing more than just reading information, watching videos, or following any hyperlinks. In addit ion to consuming the information, you also must interact with the documents. Specifically, you must do a minimum of two of the following activities: 1. Re post a news story to your Facebook page
106 2. Comment on a video Can be positive, negative, or neutral. 3. Comment on a news story Can be positive, negative, or neutral. 4. Facebook wall post 5. Become a follower of a candidate on Twitter 6. Facebook 7 Bookmark, tag, and/or link news story to Delicious or other social bookmarking web services. DO NOT forward any information to anyone. Condition Six : User to User Social Networking S ites Instructio ns This project requires that you visit four social networking websites that provide held in a few weeks on November 2, 2010. You must spend 10 minutes reading the info rmation on EACH website or on websites hyperlinked to it. You will need to note the time you entered the site (from the bottom right hand corner of the screen) and note the time you left the website on the paper provided. This project requires that you sp end the 10 minutes on each website doing more than just reading information, watching videos, or following any hyperlinks. In addition to consuming the information, you also must interact with other(s). Specifically, you must do a minimum of two of the f ollowing activities : 1. Re post Facebook wall post to your own Facebook wall 2. Forward a video to another person by email. 3. Post information about a candidate on your Facebook page. 4. Re post news story on your Facebook page. 5. Contact or send a candidate an email
107 APPENDIX C STIMULI News Sites 1. The internet service provider Yahoo ISP News Aggregator: Yahoo News Blog Newsroom Story Sep 14, b y Carol Bengle Gilbert. Link to complete story: http://news.yahoo.com/s/ynews /20100914/pl_ynews/ ynews_pl3620 2. Newspaper Website : St. Petersburg Link to full story: http://www.tampabay.com /news /politics/elections/scott sink divided on how to create jobs/1119839 3. T el evision attack: I full story: http://www.nbcmiami.com/news/politics/attack on --candidates swiping at each other -10248 2849.html 4. Online Nonpartisan News Source: Grist story: h ttp://www.grist.org/article/2010 09 09 florida Social Network Sites 1. Alex Sink Facebook page: http://www.facebook.com/#!/alexsink?ref=ts 2 Rick Scott Facebook page: http://www.facebook.com/#!/scottforflorida? v=wall & ref=ts 3. Alex Sink YouTube channel: http://www.youtube.com/user/AlexSinkFlorida 4. Rick Scott YouTube channel: http://www.youtube.com/user/ScottForFlorida
108 LIST OF REFERENCES ABC News/US. (2011). Graduation gaffes lighten the moment. Retrieved May 2 4, 2011, from http://abcnews.go.com/US/graduation gaffes lighten moment/story?id=13667038&page=2 Aberbach, J. D. (1969). A lienation and political behavior. American Political Science Review 63, 86 99. Abrahamson, D. (1998). The visible hand: Money, mark ets, and media evolution. Journalism and Mass Communication Quarterly, 75, 14 18. Abramowitz, A.I., & Saunders, K.L. (1998). Ideological realignment in the U.S. electorate. The Journal of Politics, 60, 634 652. Abramson, P.R., & Aldrich, J. H. (1982). T he decline of electoral participation in America American Political Science Review 76 (3), 502 21. Alexa.com. (2011). Top sites on the web. Retrieved May 11, 2011 from http://www.alex a.com/ Althaus, S.L., & Tewksbury, D. (2000). Patterns of Internet and traditional news media use in a networked community. Political Communication 17, 21 45. American National Ele ction Study. (2009). 2008 2009 Panel study questionnaires. Retrieved Octo ber 23, 2009, from http://www.electionstudies.org/studypages/2008_2009panel/anes2008_2009 pan el_qnaire.pdf Ancu, M., & Cozma, R. (2009). MySpace politics: Uses and gratifications of Friending candidates. Journal of Broadcasting & Electronic Media 53, 567 583. Axelrod, R. (1973). Schema theory: An information processing model of perception and cognition. American Political Science Review, 67, 12 48 1266 Bandura, A. (1977). Self efficacy: toward a unifying theory of behavioral change. Psychological Revi ew 84, 191 215. Bandura, A. (1982). Self efficacy mechanism in human agency. American. Psycho logist 37, 122 147. Bandura, A. (1986). The explanatory and predictive scope of self efficacy theory. Journal of Social Clinical Psycholog y 4, 359 373. Baran, S.J., & Davis, D.K. (2009). Mass communication theory: Foundations, ferment and future Boston: Wadsworth Cengage Learning.
109 Barker, J., & Tucker, R. N. (1990). The i nteractive l earning r evolution: Multimedia in e ducation and t raining London: Kogan Page. Barone, M. (2009). Freewheeling young voters scare both parties. The Examiner Retrieved November 11, 2009 from http://www.washingtonexaminer.com/politics/Freewheeling young voters scare both parties 8495585 69407592.html Bartels, L. M. (19 96). Uninformed votes: Information effects in presidential elections. American Journal of Political Science 40(1), 194 230. Becker, L. B & Dunwoody, S. ( 1982 ). Media use public affairs knowledge and voting in a local election Journalism Quarterly 59 212 218 Becker, L. B. and Schoenbach, K. (1998). Audience responses to media diversification: Coping with plenty In L.B. Becker and K. Schoenbach, (E ds) (1989) When media content diversifies: Anticipating audience behaviors ( pp. 1 27 ) Hillsdale, NJ : E rlbaum. Danah Boyd & Nicole Ellison. Journal of Computer Mediated Communication 13, 516 529. Bennett, S. E. (1997). Why young Americans hate politics, and what we sho uld do about it PS: Political Science and Politics 30(1), 47 53. affairs. PS: Political Science and Politics 31(3), 535 541. Blumler, J. G. (1985). The social character of media gratifications. In K.E. Rosengren, L. A. Wenner, & P. Palmgren (Eds ) Media gratifications research: Current perspectives (pp. 41 59). Beverly Hills, CA: Sage. Blumler, J. G., & McQuail, D. (1968). Television in politics: Its uses and influence Chicago: University of Chicago Press. Boulianne S. ( 2009). Does Internet Use Affect Engagement? A Meta A nalysis of Research. Political Communication 26: 2, 193 211 Boyd, D. M., & Ellison, N. B. (2008). Social network sites: Definition, history, an d scholarship. Journal of Computer Mediated Communication 13 (1), 210 230. Brody, R.A. (1978). The puzzle of political participation in America. In A. King (Ed.) The new American political system (pp. 287 234). Washington, D.C.: American Enterprise Instit ute. Bucy, E.P. (2004). Interactivity in society: Locating an elusive concept. The Information Society 20, 373 383.
110 Bucy, E.P., & Tao, C.C. (2007). The mediated moderation model of interactivity. Media Psychology 9, 647 672. Campbell A. Converse, P.E., Miller, W.E., & Stokes, D.E. ( 1960 ) The American v oter New York: John Wiley & Sons. Cho, W.K.T. (1999). Naturalization, socialization, participation: Immigrants and (non ) voting. The Journal of Politics 61, 140 1155 Cho, J., Shah D.V., McLeod, J.M., McLeod, D.M., S c holl, R.M., & Gottlieb, M.R. (2009). Campaigns, r eflection, and d eliberation: Advancing an O S R O R m odel of c ommunication e ffects Communication Theory 19, 66 88. Clark, H.D., & Acock, A.C. (1989). National elec tions and political attitudes: The case of political efficacy, British Journal of Political Science 19(4), 551 562. Chaffee, S. H., & Berger, C. R. (1987). Levels of analysis: An introduction. In C. R. Berger & S. H. Chaffee (Eds.), Handbook of commun ication science (pp. 143 145). Newbury Park, CA: Sage. Chafkin, M. (2007, June). How to kill a great idea! Inc. Magazine Retrieved August 27, 20 10 from http://www.inc.com/magazine/20070601/features how to kill a great idea.html Chung, D.S., & Yoo, C.Y. (2009). Audience motivations for using interactive features: Distinguishing use of different types of interactivity on an online newspaper. Mass Communication and Society 11, 375 397. Cohen, R. (2003, July 5). Livewire: Web sites try to make inte rnet dating less creepy. Reuters Retrieved October 5, 20 10, from http://asia.reuters.com/ Converse, P.E., & Pierce, R. (1985). Measuring partisanship. Political Methodology 11, 143 166. Culbertson, H. M., Evarts, D., Richard, P. B., Karin, S., & Stem pel, G. H., III. (1994). Media use, attention to media and agenda richness. Newspaper Research Journal 15, 14 29. Delli Carpini, M. X. (2000). Gen.com: Youth, civic engagement, and the new information environment. Political Communication 17 341 3 49. Delli Carpini, M. X., & Keeter, S. (1996). What Americans know about politics and why it matters New Haven, CT: Yale University Press. Delli Carpini, M. X., & Keeter, S. (1993). Measuring political knowledge: Putting first things first. American Journal of Political Science 37, 1179 1206.
111 Deslatte, A. (2010). GOP gubernatorial primary shaping up as expensive and nasty. Orlando Sentinel. Retrieved February 11, 2011, from http://www.orlandosentinel.com/news/local/breakingnews/os scott qualifies for governor 20100617,0,7145725.story D essauer, C. (2004). New, Media, Internet News, and the News Habit. In P. N. Howard & S. Jones (Eds.), Society o nline (pp. 121 136). Thousand Oaks, California: Sage Diddi, A., & LaRose, R. (2006). Getting hooked on n ews: Uses and gratifications and the formation of news habits among college students in an Internet environment. Journal of Broadcasting & Electronic Media 50, 193 210. Donohew, L., Finn, S., & Christ, W. (1988). The nature of news revisited: The role of affec t, schemes, and cognition. In L. Donohew, H. Sypher, & T. Higgin (Eds), Communication, social cognition, and a ff ect (pp. 195 218). Hillsdale, NJ: Lawrence Erlbaum. Donohew, R. L., Nair, M., & Finn, S. (1984). Automaticity, arousal, and informati on exposure. In R. N.Bostrom (Ed.), Communication Yearbook, 8 (pp. 267 284). Beverly Hills, CA: Sage. Downs, A. (1957). An economic theory of democracy New York: Harper and Row. Elliott, P. (1974). Uses and gratifications research: A critique and a so ciological alternative. In J. G. Blumler & E. Katz (Eds.), The uses of mass communications: Current perspectives on gratificationsresearch (pp. 249 268). Beverly Hills, CA: Sage. Erickson, E. (2010). Lessons for c onservatives f rom the R ick Scott c ampaign Retrieved January 18, 2011 from http://www.redstate.com/erick/2010/08/25/lessons for conservatives from the rick scott campaign/ Faber, R. J., Brown, J. D., & McLeod, J. M. (1979). Coming of age in the global village: T elevision and adolescence. In E. Wartella (Ed.) Children communicating: Media and development of thought, speech, understanding (pp.215 249). Beverly Hills, CA: Sage. Finkel, S.E. (1985). Reciprocal effects of par ticipation and political efficacy: A panel analysis American Journal of Political Science 29(4), 891 913. Finkel, S.E. (1987). The effects of participation on political efficacy and political support: Evidence from a West German panel. The Journal o f Politics 49(2), 441 464. Finn, S. (1997). Origins of media exposure: Linking personality traits to TV, radio, print, and film use. Communication Research 24; 507 529.
112 Florida Department of State Division of Elections. (2010). August 24, 2010 Primary E lection. Republican Primary. Retrieved February 11, 2011, from http://enight.dos.state.fl.us/Index.asp?ElectionDate=8/24/2010&DATAMODE= Fredin. E.S., & David, P. (1998). Browsing and the hypermedia interaction cycle: a model of self efficacy and goal dynamics. Journalism and Mass Communication Quarterly 75, 35 54. Gans, C. (2004 ). Low voter turnout and the decline of American civic participation. In M. S. McKinney. L. L. Kaid, D. G. Bystrom, & D. B. Carlin (Eds.), Communicating politics: Engaging the public in democratic life (pp. 80 85). New York: Peter Lang. Gerber, A.S. &Gr een, D.P.(2000). The effects of canvassing, telephone calls, and direct mail on voter turnout: A f ield e xperiment. American Political Science Review 653 663. Gilbert, C.B. (2010, September 14). 2010 Florida governor's race: Alex Sink vs. Rick Scott is mo st expensive in Florida history Yahoo News Retrieved February 7, 2011, from http://news.yahoo.com/s/ynews/20100914/pl_ynews/ynews_pl3620 Graber, D. (1984). Processing the news: How people tame the information tide New York: Longman. Haeckel, S.H. (1 998). About the nature and future of interactive marketing. Journal of Interactive Marketing 12 (1), 63 71. Hanssen, L., Jankowski, N. W., & Etienne, R. ( 1996 ). Interactivity from the perspective of communication studies I n N. Jankowski & L. Hanssen (Eds .), The contours of multimedia: Recent te chnological, theoretical and empirical developments (pp. 61 73). Luton, UK: University of Luton Press. Henke, L. L. (1985). Perceptions and use of news media by college students. Journal of Broadcasting & Electroni c Media, 29, 431 436. Hindman, M.S. (2009). The myth of digital democracy Princeton, NJ: Princeton University Press. Holbert, R. L. (2005). Debate viewing as mediator and partisan reinforcement in the r elationship between news use and vote choice. Jou rnal of Communication 55, 85 102. Huckfeldt, R., Sprague, J., & Levine, J. (2000). The dynamic of collective deliberation in t he 1996 election: Campaign effects on accessibility, certainty, and accuracy. American Political Science Review 94 641 651 Hyland, J.L. (1995). Democratic t heory: The philosophical foundations New York: Manchester University Press.
113 Institute for Southern Studies (2009). Election 2009: How voter turnout decided the Virginia governor's race. Retrieved November 11, 2009, from http://www.southernstudies.org/200 9/11/how voter turnout decided the virginia governors race.html Iyengar, S., & Jackman, S. (2003). Technology and politics: Incentives for youth participation Paper presented at the International Conference on Civic Education Research, New Orleans, LA. Jackson, R.A. (1993). Voter mobilization in the 1986 midterm election. The Journal of Politics 55, 1081 1099. Jenkins, H. (2006). Convergence culture New York: New York University Press. Jensen, J.F. (1998). Interactivity: Tracing a new concept in media and communication studies. Nordicom Review 19 (1), 185 204. Johnstone, J. (1974). Social interaction and mass media use among adolescents: A case study. In Blumler, J., and Katz, E. (eds.), The uses of mass communications: Current perspectives on gratifi cations research Sage, Beverly Hills, CA. Just, M. R., Crigler, A. N., Alger, D. E., Cook, T. E., Kern, M., & West, D. M. (1996). Crosstalk: Citizens, candidates, and the media in a presidential campaign Chicago: University of Chicago Press. Kaid, L L. (2002). Political advertising and information seeking: Comparing the exposure via traditional and Internet media channels. Journal of Advertising 31, 27 35. Kaid, L.L. (2004). Handbook of political communication research Mahwah, NJ: Lawrence Erlba um Associates. Kaid, L.L., McKinney, M.S., & Tedesco, J.C. (2007). Introduction: Political information efficacy and young voters. American Behavioral Scientist 50, 1093 1111. Karp, J.A., & Banducci, S.A. (2008). Political efficacy and participation in twenty s even democracies: How electoral systems shape political behavior British Journal of Political Science 38, 311 334. Katz, E. (1959). Mass communication research and the study of popular culture: An editorial note on a possible future for this journal. St udies in Public Communication, 3, 1 6. Katz, E., Blumler, J., & Gurevitch, M. (1974). Uses and gratifications research. Public Opinion Quarterly 37, 509 523. Katz, E., & Lazarsfeld, P. (1955). Personal influence: The part played by people in the flow of mass communications Glencoe, Ill: Free Press.
114 Kayany, J.M., Wotring, C. E., & Forrest, E. J. (1996). Relational control and interactive media choice in technology mediated communication situations. Human Communication Research 22 (3), 399 421. Kaye, B. K., & Johnson, T. J. (2002). Online and in the know: Uses and gratifications of the Web for political information Journal of Broadcasting & Electronic Media 46 (1), 54 71. Kaye, B. K., & Johnson, T. J. (2003). From here to obscurity? Media substitution t heory and traditional media in an online world. Journal of the American Society for Information Science and Technology 54, 260 273. Kaye, B. K., & Medoff, N. J. (2001). The World Wide Web: A mass communication perspective Mountain View, CA : Mayfield Kenski, K., & Stroud, N.J. (2006). Connections between Internet u se and p olitical e fficacy, k nowledge, and p articipation Journal of Broadcasting & Electronic Medi a 50(2), 173 192. Kernel, S. (1977). Presidential popularity and negative voting: An alte rnative explanation of the midterm congressional decline of the president's party. The American Political Science Review, 71, 44 66. Kim, J.Y. (2006). The impact of Internet use patterns on political engagement: A focus on online deliberation and virtual social capital Information Polity 11, 35 49. Kiousis, S. (2002). Interactivity: A concept explication. New Media & Society 4, 355 383. Kirby, E.H., & Kawashima Ginsburg, K. (2009, August 17). The youth vote in 2008. The Center for Information and Research on Civic Learning and Engagement. Retrieved October 23, 2009, from http://www.civicyouth.org/?page_id=241 Klapper, J. T. (1963). Mass communication research: An old road resurveyed. Public Opinion Quarterl y 27 515 527. Krakauer, S. (201 0 ). GOP Florida g overnor c n ew c ampaign a m osque Retrieved January 18, 2011 from http://w ww.mediaite.com/online/gop florida governor candidates new campaign ad obamas mosque/ Kuklinski, J.S, & Quirk, P.J. (2000). Reconsidering the rational public: Cognitions, heuristics, and mass opinion. In A. Lupia and M.D. McCubbins (Eds.) Elements of reas on: Cognition, choice, and the bounds of rationality ( pp. 153 182) New York: Cambridge University Press. Lasswell, H. (19 48 ). The structure and function of communications in society. In L. Bryson (Ed). The comnmunication of ideas (pp. 32 51). New York : H arper.
115 Lazarsfeld, P. F., Berelson, B.R., & Gaudet H. ( 1948 ) The p eople's c hoice New York: Columbia University Press. Lazarsfeld, P. F.,&Stanton, F. (1942). Radio research, 1942 1943. NewYork: Duell, Sloan & Pearce. Lometti, G. E., Reeves, B., & Bybee, C R. (1977). Investigating the assumptions of uses and gratificationsresearch. Communication Research, 4, 321 328. Levy, M. (1977). Experiencing television news. Journal of Communication 27, 112 117. Margolis, M., & Resnick, D. (2000). Politics as usu Thousand Oaks, CA: Sage Publications. Markus, H., & Zajonc, R. B. (1985). The cognitive perspective in social psychology. In G. Lindzey & E. Aronson (Eds.), The handbook of social psychology (3rd ed., pp. 137 229). Ne w York: Random House. McLeod, J. M., & Becker, L. B. (1981). The uses and gratifications approach. In D. D. Nimmo & K. R. Sanders (Eds.), Handbook of political communication Beverly Hills, CA: Sage. McQuail, D. (2005). Mass communication theory (5th ed .). London: Sage. McQuail, D., J. G. Blumler, and J. R. Brown. ( 1972 ) The television audience: a revised perspective. In D. McQuail (Ed.) Sociology of mass communications Harmondsworth, Penguin McMillan, S. J. (2002). Exploring models of interactivity from multiple research traditions: Users, documents, systems. In L. Lievrouw & S. Livingstone (Eds.), The handbook of new media: Social shaping and consequences of ICTs (pp. 163 182). London: Sage. M iller, A.H., & Miller, W.E. (1975) Issues, Candidates and Partisan Divisions in the 1972 American Presidential Election British Journal of Political Science 5, 393 434. Mills, R. (2010). Florida governor election: AP calls race for Rick Scott, Bill McCollum not conceding. Naplesnews.com. Retrieved February 11, 2011, from http://www.naplesnews.com/news/2010/aug/24/florida governor election results Scott McCollum/ Mills, R., & Hale, L.W. (2010). Scott vs. Sink: Down to the wire, candidat es trade barbs, tout credentials. Naplesnews.com. Retrieved February 7, 2011, from http://www.naplesnews.com/news/2010/oct/30/alex sink rick scott florida governor down to wire/
116 Morrell, M.E. (2003). Survey and experimental e vidence for a r eliable and v ali d m easure of i nternal p olitical e fficacy The Public Opinion Quarterl y 6 7(4), 589 602 National Association of Secretaries of State. (1999). New millennium project part 1: American youth attitudes on politics, citizenship, government, and voting Lexington KY: Author. Newhagen, J. E. (1998, July). Hitting the Agenda Reset Button for the Internet: The problem of Matching Research with Development Paper presented at the International Communication Association Conference, Jerusalem, Israel. Newhagen, J. E., & Rafaeli, S. (1996). Why communications scholars should study the Internet: A dialogue. Journal of C omputer M ediated C ommunication 1(4), 4 13. Niemi, R. G., Craig, S. C., & Mattei, F. (1991). Measuring internal political efficacy in the 1988 National Election Study. American Political Science Review 85 1407 1413. Niemi, R.G., & Weissberg, H.F. (2001). Controversies in voting behavior Washington, D.C.: Congressional Quarterly, Inc. Norris, P (2004). Electoral Engineering: Voting Rules and Political Behavior New York: Cambridge University Press, 2004. news media. Journalism Quarterly 50, 543 548. Pan, Z., Shen, L., Paek, H., & Sun, Y. (2 006). Mobilizing political talk in a presidential campaign: An examination of campaign effects in a deliberative framework. Communication Research 33 315 345. Page, B. I. (1996). Who deliberates? Mass media in modern democracy Chicago: University of Chicago Press. Patterson, T.E. (1994). Out of orde r New York: Random House. Patterson, T.E. (2002). The vanishing voter : Public involvement in an age of uncertainty New York: Random House Pavlik, J. (1996). New media technology: Cultural and commercial perspectives Boston: Allyn & Bacon. Perse, E.M., & Courtright, J.A. (1993). Normative images of communication media: Mass and interpersonal channels in the new media environment. Human Communication Research 19, 485 503.
117 Perse, E. M.,& Dunn, D. G. (1998). The utility of home computers and media use: Implications of multimedia and connectivity. Journal of Broadcasting & Electronic Media 42 435 456. Pew Internet & American Life Project (2008 a ). The I nternet and the 2008 election. Retrieved October 13, 2010, from http://www.pewinternet.org/pdfs/PIP_2008_election.pdf. Pew Research Center. (2010 a February). Millenials: Portrait of g eneration n ext Retrieved November 11, 2010, from http://pewsocialtrends.org/files/2010/10/millennials confident connected open to change.pdf Pew Research Center ( 2010b, November). A clear rejection of the status quo, no consensus about future policies: GO P wins big despite Retrieved January 9, 2011, from http://pewresearch.org/pubs/1789/2010 midterm elections exit poll analysis Pew Research Center for the People and the Press (2007). Public knowledge of current affairs little changed by news and information revolutions: What Americans know: 1989 2007. Retrieved October 23, 20 10 from http://people press.org/report/319/public knowledge of current affairs little changed by news and information revolutions Pew Research Center for the People and the Press (2008 b ). Internet overtakes newspapers as news outlet Retrieved October 23, 2009, from http://pewresearch.org/pubs/1066/internet overtakes newspapers as news outlet Project for Excellence in Journalism. (1998). Changing definitions of news. Retrieved October 11, 2010 from http:/ /www.journalism.org/resources/research/reports/definitions/differences .asp n ewsroom i nvestment 2006 a nnual r eport Convergence Retrieved October 11, 2010, from http://www.jour nalism.org/node/1486 Pingree, R. J. (2007). How messages affect their senders: A more general model of message effects and implications for deliberation. Communication Theory, 17, 439 461. Plutzer, E. (2002). Becoming a habitual voter: Inertia, resources, and growth in young adulthood. American Political Science Review 96, 41 56.
118 Poll ock III, P.H. (1983). The participatory c onsequences of i nternal and external political e fficacy: A r esearc h n ote T he Western Political Quarterly 36 ( 3 ), 400 409 Putnam, R.D. (2000). Bowling alone: The collapse and revival of American community New Yo rk: Simon & Schuster. Raacke, J & Bonds gratifications theory to exploring friend CyberPsychology & Behavior 11, 169 174. Rafaeli, S. (1988). Interactivity: From new media to communication. In R.P. Hawkins, J.M. Wiemann, & S. Pingree (Eds.), Advancing communication science: Merging mass and interpersonal process (pp. 110 134). Newbury Park, CA: Sage. Rafaeli, S., & Ariel, Y. (2007). Assessing interactivity in computer me diated research. In A. Joinson, K. McKenna, T. Postmes, & U.D. Reips (Eds.), The Oxford handbook of Internet psychology (pp. 71 88). New York: Oxford University Press. Real Clear Politics. (2010). Florida Governor Scott vs. Sink. Retrieved February 11, 2011, from http://www.realclearpolitics.com/epolls/2010/governor/fl/florida_governor_scott_vs _sink_vs_chiles 1607.html Renshon, S.A. (1974). Psychological needs and political behavior: A theory of personality and political efficacy New York: The Fr ee Press. Rheingold, H (1993). The virtual community: Homesteading on the electronic frontier Howard Rheingold. Published: New York, NY: Harper Perennial. Riker, W., & Ordeshook, P. (1968). A theory of the calculus of voting. American Political Scienc e Review 62, 25 42. Rosenstone SJ, Kinder DP, Miller WE (1997) American national election study Inter university Consortium for Political and Social Research Ann Arbor, MI Rubin, A. M. (1981). An examination of television viewing motivations. Communica tion Research 8, 144 165. Rubin, A.M. (1983). Television uses and gratifications: The interactions of viewing patterns and motivations. Journal of Broadcasting & Electronic Media 27, 37 51. Rubin, A. M. (1984). Ritualized and instrumental television vi ewing. Journal of Communication 34, 67 77. Rubin, A. M. (1985). Media gratifications through the life cycle. In K. E. Rosengren, L. A. Wenner, & P. Palmgreen (Eds.), Media gratifications research: Current perspectives (pp.195 208). Beverly Hills, CA: Sage
119 Rubin, A M. (1993). Audience activity and media use. Communication Monographs 60 98 103. Rubin A.M. and Rubin, R. ( 1982 ) Older persons' TV viewing patterns and motivations. Communication Research, 9 287 313. Ruggiero, T. E. (2000). Uses and gratifications in the 21 st century. Mass Com munication & Society 3, 3 37. Sachs, R.A. (2010). Retrieved December 3, 2010, from http://www.ronsachs.com/news2010_11_02.shtml Schneider, S.M., & Foot, K.A. (2006). Web campaigning by U.S. presidential primary candidates in 2000 and 2004. In A.P Williams & J.C. Tedesco (Eds.), The Internet election: Perspectives on the Web in campaign 2004 Lanham, MD: Rowman & Littlefield. Selnow, G.W. (1998). Electronic whistle stops: The impact of the Internet on American politics Westport, CT: Praeger. Severin, W.J., & Tankard, J.S. (2001 ). Communication theories: Origins, meth ods, and uses in the mass media New York: Addison Wesley Longman, Inc. Shah, D. V., Cho, J., Eveland, W. P. Jr., & Kwak, N. (2005). Information and expression i n a digital age: Modeling Internet effects on civic participation. Communication Research 32 531 565 Shani, D. (2010). On the origins of political interest (Doctoral dissertation). Princeton University, Princeton, NJ. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental s tudies: New procedures and recommendatio ns. Psychological Methods 7 422 445 American Life Project. Retrieved October 3, 2009, from http:/ /www.pewinternet.org/~/media//Files/Reports/2009/TheInternetsRole in Camp aign_2008.pdf Stacy, M., & Sedensky, M. (2010). Money, message, mad electorate make Scott Fla. Gov. The Washington Post. Retrieved February 11, 2011, from http://www.washingtonpost.com/wp dyn/content/article/2010/11/03/AR2010110300781.html Strmb ck, J., & Esser, F. (2009): Shaping politics: Mediatizat ion and media interventionism. In Lundby, Knut (ed.) Mediatization. Concept, Changes, Consequences (pp. 205 223). New York: Peter Lang.
120 Stromer Galley, J. (2000). On line interaction and why cand idates avoid it. Journal of Communication 50(4), 111 132. Stromer Galley, J. (2004). Interactivity as product and interactivity as proce ss. The Information Society 20, 391 394. Stromer Galley, J., & Baker, A.B. (2006 ). Joy and sorrow of interactivity on the campaign trail: Blogs in the primary campaign of Howard Dean. In A.P Williams & J. Tedesco (Eds.) The Internet election: Perspectiv es on the web in campaign 2004 (pp.111 132). Lanham, MD: Row m an & Littlefield Publishers, Inc. Sundar, S. S., Kalyanaraman, S., & Brown, J. (2003). Explicating Website interactivity: Impression formation effects in political campaign sites. Communication Research 30 (1), 30 59. Sunstein, C. R. (2001). Republic.com Princeton, NJ: Princeton University Press. Swanson, D.L. (1977). The uses and misuses of uses and gratifications. Human Communication Research 3, 214 221. Swanson, D.L. (1979). Political comm unication research and the uses and gratifications model: A critique. Communication Research 6, 37 53. Sweetser, K.D. & Kaid, L.L. (2008) Stealth s oapboxes: Political i nformation e fficacy, c ynicism and u ses of c elebrity Weblogs among r eaders New Media & Society 10(1) 67 91 Sweetser, K. D., & Weaver Lariscy, R. (2007). Candidates make good friends: An analysis of Paper presented at the National C ommunication Association Conference Chicago, IL. Szuprowicz, B. O. (1995). M ultimedia networking New York: McGraw Hill, Inc. information efficacy. American Behavioral Scientist 50, 1183 1 194. Teixeira, Ruy A. 1992. The Disappearing America n Voter. Washington, D.C.: The Brookings Institution. Trammell, K. D., Williams, A. P., Postelnicu, M., & Landreville, K. (2005). Evolution of o nline campaigning: Increasing interactivity in candidate Web sites through hyperlinks and blogs. Mass Com munication & Society 9 (1), 21 44. University of Florida Healthy Gators Coalition. (2010). Healthy Gators Student Survey Report 2010. Retrieved January 9, 2011, from http://healthygators.ufsa.ufl.edu/wp content/uploads/2011/03/HG Student Survey 2010 Revis ed.pdf
121 U.S. Bureau of the Census (2009, September 1). Voting and Registration. Retrieved Oct ober 23, 2009 from http://www.census.gov/population/www/socdemo/voting.html Verba S & Nie N 1972. Participation in America New York: Harper & Row. Vincent, R.C., Basil, M.D. ( 1997 ) College students news gratifications, media use and current event knowledge. Journal of Broadcasting and Electronic Media 41, 380 392. Wattenberg, Martin P. 1996. The Decline of American Political Parties, 1952 1994 Cambridge, MA: Harvard University Press. Williams, F., Rice, R.E., & Rogers E.M. (1988). Research Methods and the New Media New York: Free Press. Wolfinger, R. E., & Ros enstone, S.J. (1980). Who v otes? New Haven, CT: Yale University Press. Wolfinger, R.E., Rosenstone, S.J., & McIntosh, R.A. (1981). Presidential and congressional voters compared. American Politics Research 9, 245 256. Wright, C. (1 960 ) Functional analysis and mass communication. Public Opinion Quarterly 24. Wu, G (1999) Perceived i nteractivity and a ttitude t o ward w ebsite In M. Roberts (Ed.), Proceedings of the American Academy of Advertising (pp. 254 262). Gainesville FL : Univer sity of Fl orida Press Zaichowsky, J.L. (1985). Measuring the involvement construct. Journal of Consumer Research 12, 341 352
122 BIOGRAPHICAL SKETCH O rigina ting from the heart of the Blue Ridge Mountains in southwestern Virginia, David Lynn Painter calls Florida degrees in Communication Arts from the University of West Florida. U pon completion of his doctoral degree from the College of Journalism and Communication at the University of Florida, he became a Course Di rector in the New Media Journalism Master of Arts program at Full Sail University.