Conceptualising and Measuring E - Participation : Paper prepared for presentation at the Internet , Voting , and Democracy Conference ( II ) University of California Irvine , Center for the Study of Democracy

This paper seeks to more clearly define and measure the phenomenon of eparticipation, focusing specifically on the question of whether it is simply an extension of existing forms, differing only in mode, or whether it offers a new and qualitatively different form of political engagement. The analysis is timely since despite over a decades‟ worth of research being conducted into the topic of eparticipation, a clear and commonly accepted definition of the activity itself remains elusive. After developing a typology of e-participation we test it using confirmatory factor analysis. The results indicate that different forms of e-participation can be identified, some of which are conventional and some of which are new. The implications of each for mobilization of citizens are discussed in a concluding section. Introduction This paper seeks to advance the burgeoning literature on online or „eparticipation‟ by developing a more sophisticated conceptual and empirical understanding of this new phenomenon within political science. Specifically we contend that the study of e-participation has been hampered by an unclear understanding within the literature of what the object of study precisely is, and a lack of a strong theoretical foundation with which to define it and its various forms. In the early days of study, there was a tendency to adopt blunt measures of basic internet „use‟ or „access‟. However, as surveys have expanded their range of items measuring internet use, a more multi-faceted picture of online participation has emerged and the detection of mobilising effects increased. In this paper we seek to develop this trend to provide a more diverse and contextualised understanding of e-participation, by categorising the approaches taken to the topic within the existing literature, drawing upon the work on offline participation to develop a more clearly articulated definition of e-participation and its various modes and testing that conceptualisation with original survey data drawn from the UK General election of 2010. The paper is organised into four basic sections. In the first section we outline the current state of research on e-participation by classifying studies into one of the main approaches. Arguing that very few studies have actually sought to define and differentiate e-participation we then introduce the offline participation literature and the central concepts it offers that help to develop our understanding. In a third section of the paper we outline our measurement model of e-participation and test it through simultaneous confirmatory factor analysis. In a final section of the paper we discuss the implications of the different modes of e-participation uncovered in terms of how far they are linked with mobilization of individuals. The study of e-participation The study of online or e-participation is a growing field of enquiry and studies placed under its „banner‟ vary in scope, method and conclusions. Much of the early work was speculative and even visionary in nature with e-participation being understood from a wider holistic „edemocracy‟ perspective, i.e. whether it promotes a communitarian or more direct model of democracy (Rheingold 1993; Negroponte 1996; Dahlberg 2001; Budge 1996). More empirical work began to emerge in the late 1990s, a strand of which focused primarily on the informal and radical nature of e-participation, with attention given to organisational uses of the internet as a means to coordinate protest and collective action. This included case studies of anti-government protests in Mexico among the Zapatistas, anti-war movements in the countries comprising the „coalition of the willing‟. the global anti-capitalist protests in the famous „Battle in Seattle‟ and anti-WTO protests, and collective action (Bennett 2003; Bonchek 1995; Capling and Nossal, 2001; Gillan and Pickerill 2007, Pickerill 2001; Rheingold 2002; Della Porta and Mosca 2005; Earl 2006; Mosca 2010), as well as more theoretical accounts of how the internet reduced problems of collective action Bimber et al. 2005. A second empirical strand of research also emerged in the late 1990s looking more closely at the impact of the internet on individuals‟ political attitudes and behaviour within the representative sphere of politics. The focus was primarily on questions of mobilization and whether the internet increases individuals‟ involvement in politics. Seminal work was done in this area by Bimber (1999; 2001) who compared predictors of e-contact with government versus by phone and mail and concluded „small but subtle‟ effects, in that similar resources mattered for each but that internet contacters were less politically connected and younger. He then followed this up with an examination of the effects of internet access and obtaining online campaign information on voting, donation, and other political acts such as attending a rally, contacting government officials, and working for a candidate. Again he concluded small but significant relationship of online information seeking with donations. Following Bimber‟s analyses there has been an increasing volume of work on the topic of e-participation and particularly the question of whether internet is increasing participation. While the search for effects at the individual level is an important one, the rush to examine it causally we argue has led to an overlooking of more basic theoretical questions about the actual object of study i.e. the „what‟ of eparticipation with a variety of internet use and participation measures employed. This paper constitutes an attempt to address this deficit by profiling the literature on eparticipation to date, identifying the gaps in our current understanding and positing and testing a new multi-dimensional model of e-participation that draws on classic theories of participation. E-participation in the representative sphere Studies placed in the „e-participation‟ field that followed Bimber‟s work can be broadly divided into two approaches to the topic. While both look at the impact of the net on political activities and attitudes, the first set take a more tangential or indirect approach and examine the impact of internet use, defined in general or more specific terms, on offline political behavior, attitudes and levels of civic engagement. A second body of work takes a more direct approach to the topic looking at a range of specifically online political activities, such as e-contact, e-petition. In a nutshell, the studies in the former camp look at participation and the internet whereas the latter are centered on participation on the internet, and define e-participation as an activity in its own right. Some studies cover both areas of enquiry. While the former studies are of interest in terms of building the picture of the internet as a mobilizing force, as sources for better definition and measurement of e-participation their value is quite limited. It is the latter set of studies that we argue are most relevant to the purposes of this paper. Below we briefly profile these literatures, focusing particular attention on the second and its attempts to define and differentiate forms of e-participation. Participation and the Internet: E-participation studies in this category take a broad approach to the topic, relating internet use, understood in basic access or use terms (Shah et al. 2002; Norris 2004; Stoneman 2007; Sylvester and McGlynn 2010) or as more specific types of usage such as information seeking, recreational activities (game playing) or online skills (sending an email attachment, setting up a website) to changes in individuals‟ level of „real world‟ political activity or civic attitudes (Jennings and Zeitner 2003; Kaye and Johnson 2002; Scheufele and Nisbet 2002; Tolbert and McNeal 2003; Tolbert and Mossberger 2003; Johnson and Kaye 2003; Hardy and Scheufele 2005; Moy et al. 2005; Krueger 2002; Best and Krueger 2005; Kenski and Stroud 2006; Xenos and Moy 2007; Mossberger et al. 2007; Quintelier and Vissers 2008). The focus is thus on internet use as a driver to real world participation, and the substantive findings have focused on how far radical or even subtle changes in citizen behavior can be detected in levels of engagement. The „e‟ component is an independent variable sitting on the right side of the equation. In general, the work has pointed to the internet as having a positive if small effect on individuals‟ proclivity to engage in political activity (Boulianne 2009). Participation on the internet Studies coming under this banner are those that conceptualize and operationalize e-participation as a dependent variable and see it as an activity in its own right, to be explained. The central focus has again been on detection of mobilization with e-participation regressed on a range of standard predictors to see how far the usual suspects are supported. The types of activities studied have varied, however, with some studies focusing largely on existing or „converted‟ forms of offline participation such as emailing government or online donation and petition signing (Bimber 1999; Krueger 2002; Anduiza et al. 2010; Saglie and Vabo 2009; Schlozman et al. 2010; Sylvester and McGlynn 2010) and others including activities with no obvious offline counterpart, such as blogging, following a politician on twitter, posting or commenting on online video (Rojas 2010; Leung 2009), while others have covered both types (Schlozman et al. 2010). Finally, there are studies that examine individuals‟ involvement in specific initiatives or experiments in online interaction/discussion consultation (Stanley and Weare 2004) Again while the results have been largely positive in arguing that e-participation is attracting some new faces to the political arena, the lack of a commonly shared or agreed on subject of study means that there is a lack of coherence or continuity in the findings. In this vein, some studies have attempted to draw out or differentiate the „what‟ of e-participation and particularly whether it constitutes a new and different form (independence thesis), or simply is an extension of existing types of participation, differing only in mode (convergence thesis). One of the first attempts to examine this was by Jensen et al. (2007) which used multi-dimensional scaling to test the independence of various online and offline participation items. The findings were seen as supporting the independence rather than convergence. However, there were some substantial differences between the online items measuring community involvement and those measuring offline related more to individuals civic skills. Subsequent work has moved more toward supporting the convergence model. Saglie and Vabo (2009) employed an exploratory factor analysis to interrogate a series of conventional e-participation items, i.e. those with offline equivalents. The results revealed e-participation to be multi-dimensional phenomenon, breaking down into three factors corresponding to contact, information seeking on websites, and use of e-petitions. The authors then went on to use a uni-dimensional scale in the subsequent mobilization analyses. Adopting more advanced techniques, Hirzalla and Van Zoonen (2010) conducted a simultaneous confirmatory factor analysis (SCFA) on 14 items measuring both offline and online participation. Their results show a picture of differentiation but also of convergence between the different types of participation, with 4 separate factors emerging, three of which merge off and online activities – „politics‟ „activism‟ and „sharing‟. For the first two factors it is visiting websites about politics and activist causes respectively that constitute the online participation component in amongst more traditional activities such as contacting a politician or joining a demonstration. The sharing factor is largely composed of online activities – forwarding an email, signing an e-petition and using a discussion forum – with offline discussion adding a fourth dimension. For the authors the findings show that online participation cannot be conceived of as a new and independent mode but is linked and blends with existing forms. As one of the few attempts to test the differentiation and convergence propositions the paper clearly offers a step forward in the literature. However, the theoretical underpinnings to the expectations of differentiation and convergence are limited and the items confirming convergence are, according to much of the class participation literature, pre-participatory in that they focus on information seeking and discussion. We return to this argument below in our discussion participation studies more generally. Other analyses of interest on this topic have taken the SCFA approach one step further and developed and tested full structural equation models to assess both the question of differentiation or dimensionality of e-participation and mobilization. Here the focus has been on positing a pathway effect of some kind, whereby different types of e-participation are linked to other types and ultimately to offline participation. Possibly the most advanced of these is the analysis of Rojas et al (2009) which models a three step causal path from online news consumption to an „expressive‟ form of e-participation that centers on posting comments and opinion to various fora. This more passive form of participation is then seen to prompt more active effort via mobiles and social network sites to mobilize others, which ultimately leads to offline engagement. Similar conclusions are reached by Shah et al. (2005) in a two step SEM in which e-information seeking is confirmed as prompting civic emailing and offline talk which in turn prompts offline participation. Other analyses by Gil de Zuniga et al (2009) or Baumgartner and Morris (2010) have employed hierarchical and two stage least squares regression techniques to model the stages of engagement surrounding e-participation. Both connected online information seeking to more active forms of e-participation, the former study with an additional step of online discussion. In addition, both tested but found no support for the e-activities providing a stimulus to offline participation. The attention given to e-participation, understood as a set of definable activities that take place or are largely reliant upon the internet, is clearly expanding. To date, however, much of the attention has focused on questions of causality, either in terms of direct analysis of who is engaging in e-participation activities, or how various types of online activity link together to produce offline or online participation. In the latter set of studies there has been some attempt to disaggregate the concept of e-participation as a distinct and definable set of activities. This work has pointed to a multi-stage process whereby an initial engagement in more passive and less costly forms of engagement such as reading and then forwarding or commenting on political information, leads onto a more pro-active attempt at mobilization of others or direct involvement in politics. In particular, the identification of a possibly new intervening form of „expressive‟ e-participation development is an important development, in that it appears to challenge existing accounts of participation theory to argue for a new „softer‟ form that acts as an important precursor to more active forms. Within this literature a number of gaps and a significant degree of confusion remains, however. In particular while there is basic consensus that e-participation can be differentiated into different sub-types, the theoretical justification for this differentiation is missing in most accounts or quite limited. Particularly the question of what this new form of expressive or opinion sharing constitutes as a form of participation requires closer analysis. Furthermore, there appears to be disagreement on the extent to which the various forms of e-participation uncovered constitute a fundamentally new participatory repertoire, with little connection to existing forms, or whether they are simply translations of offline activities, differing only in mode. Defining E-participation This paper seeks to address two central research questions to advance the literature on e-participation. First we confront the basic question of differentiation of eparticipation – what sub-forms can be identified and how far do they conform to existing participation theory in terms of their levels of activism and passivity? Secondly, if a distinctive set of internet-based participatory activities can be identified, to what extent do they constitute an extension or equivalent of existing forms of offline participation? Or, are they independent of established modes, requiring a new set of conceptual criteria to be fully understood? Perhaps both convergence and independence are occurring in that the online environment offers replication of existing modes like contact, petition or discussion. But the rise of the social media has opened up new possibilities for activities such as blogging, tweeting and „friending‟ that are wholly new means of political expression and engagement. In order to investigate these questions we embark on a review of the classic studies of participation and the classificatory schemas that have been developed to understanding offline forms. To what extent can the internet-based types of participation so far identified be accommodated within these schemas? Defining and measuring offline political participation On the first question of what constitutes or comes within the rubric of participation, and particularly the degree of activity or passivity associated with it, this appears to be a movable feast with the boundaries changing and expanding over time. Overall the repertoire of activities considered political participation has grown dramatically since the first studies in the field where conducted focusing only on voting and behaviours connected to elections (Lazarsfeld et al. 1948; Berelson et al. 1954; Campbell et al. 1960). Political participation can take a variety of forms and the inclusion of new activities as objects of study during the past decades has been accompanied by a debate in the literature on how to categorize these into relatively homogeneous groups. Verba and Nie (1972) conducted one of the first studies widening the range of activities beyond voting. They showed that individuals tend to specialise in homogeneous sets of activities, labelled as modes of participation. They identified four different modes: turnout, campaign activities, communal activities and parochial participation. What united these activities was they are activities that have “the intent or effect of influencing government action – either directly by affecting the making or implementation of public policy or indirectly by influencing the selection of people who make those policies‟ (Verba et al 1995: 38). The multidimensional nature of participation has become something of a truism since the work of Verba and Nie. Different forms of participation vary so much in the concerns that motivate them, in their duration and intensity, in their target, and in their outcomes, that a complete analysis must go beyond uni-dimensional scales” (Brady, 1999: 741). However the boundaries drawn by Verba and Nie have been questioned and particularly whether actions directed toward government – voting, campaigning, contacting – constitute the only legitimate forms of participation. Only a few years later, several authors incorporated to the repertoire a new set of actions such as attending a demonstration or a protest political meeting, boycotting products, taking part in a strike or in violent activities (Marsh 1977; Barnes, Kaase et al. 1979). Political protest emerged as a new cluster of participatory activities mainly characterized by their unorthodox nature, understood as a lack of adequacy with the normative rules and common practices in a society (Marsh and Kaase 1979: 41). These studies suggested a new theoretical conceptualization which distinguished between conventional and unconventional modes of political participation. In recent years however, the spread of protest actions meant that the label of unconventional participation has become inappropriate. Except for violent actions, protest gradually has been seen to become a relatively common and legitimate form of participation (Parry et al. 1992; Verba et al. 1995; Dalton 2002; Norris 2002; Teorell, Torcal and Montero 2007; Rucht 2007). The rising acceptance of these methods paradoxically undermined a conceptualisation of these activities based on their “unorthodox” nature. In an attempt to bring this typology up to date, the authors of the study on “Citizenship, Involvement and Democracy” (CID) suggested a new conceptualisation of the classical distinction between conventional and unconventional participation (Teorell, Torcal and Montero 2007). Instead of clustering activities according to their orthodox/unorthodox nature, they applied a distinction based on the channel of expression being employed by participants. The activities formerly labelled as conventional which are developed within the framework of representation, like voting or party activities, were labelled as “representational participation”. On the contrary, protest participation and consumerism were defined as belonging to a category called “extra-representational participation”. For the most part the traditional participation literature has tended to treat eparticipation either as falling outside of the boundaries of „true‟ participation, or if it is considered a legitimate form in its own right then this is generally view in terms of „independence‟ rather than convergence. Certainly the work of Teorell et al (2007) forms a good example of the former approach and constitutes the most recent attempt at a post-internet classification of participation. The authors develop a five category typology of participation that encompasses voting, consumer participation, party activity, protest activity and contacting. The approach thus follows the model of Verba et al but includes an extra-representational element in which the latter two elements fit. They then further divide the classification on whether it is exit or voice based, meaning a singular one-off act (voting and boycotting) versus a more directed, specific and ongoing form of input (party, protest, and contact). Despite extending out beyond the Verba et al (1972) scheme it is not clear how e-participation is accomodated. Speculation by other authors within this literature has been somewhat limited but in general authors have either proved sceptical about the extent to which eparticipation constitutes a genuine form. For Verba et al (1995) for instance, a social networking site like Facebook is simply a forum for political talk among friends rather than a place for organized political effort directed toward influencing public officials. The political groups formed are more about affinity and shared interests than concerted political action. „Friending‟ a candidate is not same as working in a campaign they argue. In more recent work, however they do appear more open to the idea of these interactive forms of political engagement serving as catalyst to more concerted political behaviour and are alert to the fact that these forums are changing so rapidly “....that they may well morph into new forms of activity aimed at political influence.‟ (Schlozman et al 2010: 501). If we assume a convergence model then table 1 provides some indication of how the Teorell et al (2007) scheme could be adapted to fit some elements of eparticipation that have been identified within the literature.