Personal mobilisation, civic norms and political participation

Geoforum 45 (2013) 178–189

Contents lists available at SciVerse ScienceDirect

Geoforum journal homepage: www.elsevier.com/locate/geoforum

Personal mobilisation, civic norms and political participation Charles Pattie a,⇑, Ron Johnston b a b

Department of Geography, University of Sheffield, Sheffield S10 2TN, United Kingdom School of Geographical Sciences, University of Bristol, Bristol BS8 1SS, United Kingdom

a r t i c l e

i n f o

Article history: Received 28 June 2011 Received in revised form 5 November 2012 Available online 25 December 2012 Keywords: Electoral geography Political participation Contextual effects

a b s t r a c t Previous analyses of the role of context in political participation have tended to focus on voting, and on electoral context (in particular, the closeness of the local campaign and the marginality of the electoral district). However, neither this form of political engagement nor these measures of context capture the range of possible influences. This paper therefore analyses the role of civic norms and personal mobilisation on participation in a range of different forms of political activity. In general, individuals are responsive both to actual mobilisation and to their perceptions of how others in their acquaintance circles and neighbourhoods are likely to act. The more the local environment encourages participation (whether in the form of more frequent invitations to get involved or of stronger perceived norms for participation), the more likely individuals are to get involved. However, contextual influences on participation are stronger for the less politically motivated than for the more motivated. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction Active public political participation is a cornerstone of democracy. High rates of involvement, whether voting in an election, taking part in community politics, or working for a campaign or organisation, are taken as signs of a healthy polity. Low participation rates, meanwhile, are often presented as signs of democratic malaise (although the causes may be disputed: Stoker, 2006; Hay, 2007). As a result, falling rates of electoral participation in many western polities since the 1950s have been the focus of anxious debate (Norris, 2002, 2011; Franklin, 2004). Understanding the factors which explain participation is therefore an important enterprise. Over the years, considerable effort has been devoted to understanding citizen engagement in politics. We know a substantial amount about who votes, who volunteers, who contacts the authorities, who protests, and the conditions under which they do so. Much of the relevant literature focuses on one of two scales (or occasionally both): the individual or the wider society. Most studies of political behaviour focus almost entirely at the former scale, however, studying the decision-maker in terms of her/his characteristics only with little or no relevance to the varying spatial contexts – household, neighbourhood, workplace, local organisations, etc. – within which the decision on whether to act, and how, is made. Also part of the picture, ⇑ Corresponding author. E-mail address: c.pattie@sheffield.ac.uk (C. Pattie). 0016-7185/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.geoforum.2012.11.002

however, are influences which might operate at an intermediate level between the individual and the wider society as a whole, in particular the potential impact of those in individuals’ social networks and local communities. Citizens’ decisions on whether or not to participate may depend not only on their own resources and predilections, but also on their assessments of what others in their communities are likely to do, or (to the extent that internalised social norms operate) what they think others in the community will see as normal and acceptable. Where participation rates in general have declined, however, this contextual effect, if present, may contribute to a negative feedback mechanism: if citizens perceive their peers to be unlikely to participate, this may further discourage them from doing so. In this paper, therefore, we investigate the relative importance of local mobilisation and civic norms as contextual influences on participation. We use the five main theories of citizen engagement, discussed below, as the foundations for political decision-making, but then – using survey data suitable for the task – add in a number of variables that, if our arguments regarding contextual effects are valid (and there is considerable evidence for one form of behaviour – voting – to suggest that they do: Johnston and Pattie, 2006) should extend our appreciation of the influences on who does what, where.

2. Understanding civic norms and mobilisation The decision to participate in politics is likely to be affected not only by one’s own personal circumstances and outlooks, but also

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

by what other citizens do, and (not quite the same thing) what one thinks others are likely to do. One theoretical manifestation of this is the well-known paradox of voting (Downs, 1957). At its simplest, this suggests that a rational individual should participate if the personal benefits from doing so outweigh the costs. But voting is a communal act, and depends on the mutual participation of many others. In most circumstances, so many individuals vote that the participation of any one individual is very unlikely to be crucial to the outcome of the contest. Our rational voter might be well advised, therefore, to discount the benefits likely to accrue from participation by the probability that their own participation will determine the outcome. As the latter is generally minuscule, the incentives for participation for this hypothetical rational voter should always be far too small to justify taking part: a rational voter who assumes many others will vote should, on the basis of that knowledge, abstain. At the extreme (hence the paradox), if all voters are rational in this way, none should vote, defeating the object of the exercise. In practice, of course, this hypothetical situation rarely if ever occurs in election voting (though it may be more realistic for other, more demanding forms of political participation). That said, there certainly is evidence that individuals’ participation decisions are in part affected by (perceptions of) how others will behave. A near-analogy to the paradox of voting is the tendency for constituency electoral turnout to correlate negatively with the seat’s marginality: the closer the competition in a constituency, the higher the turnout there (Denver and Hands, 1985; Denver, 1995; Pattie and Johnston, 2005; Johnston and Pattie, 2006; Johnston et al., 2011). At least some individuals use their knowledge of the state of local party competition to decide whether it is worth their taking part in the election. This and similar effects are liable to rest on a range of mechanisms. Direct mobilisation is clearly important. There is substantial evidence that individuals who are specifically asked to participate in a range of political actions are much more likely to do so, ceteris paribus, than are individuals who are not personally invited (Brady et al., 1995; Verba et al., 1995; Pattie et al., 2004). Some invitations to get involved come from friends, family, acquaintances and so on. Many others come from organised political groups. For instance, party canvassing activity affects electoral participation. The more actively political parties campaign in particular electoral districts, and the more voters they contact there, the greater the political rewards for them and the higher the local turnout (Jacobson, 1978, 1990; Huckfeldt and Sprague, 1995; Pattie et al., 1995; Denver and Hands, 1997; Pattie and Johnston, 2009, 2010; Johnston et al., 2012). As parties tend to focus their local campaign and canvass efforts in the most marginal constituencies, this helps create the correlation noted above between the closeness of the contest and the turnout. Similarly, experimental studies show that ‘getout-the-vote’ interventions are effective: those contacted are more likely to take part than those who are not (e.g. Green and Gerber, 2004; Gerber et al., 2008; Nickerson, 2008; John and Brannan, 2008). Furthermore, the effect of direct mobilisation by parties and political groups might be cumulative: individuals who are contacted by party campaigns are more likely than those who are not contacted to go on and try and persuade their friends and acquaintances to vote as well (McClurg, 2004). We therefore expect that direct mobilisation in the form of specific invitations to participate from organised political groups and also from friends, families and relatively casual contacts will tend to encourage a range of forms of political engagement. The influence of others on individuals’ propensity to participate is not restricted to direct mobilisation, however. Most people want to fit in with their peer groups and communities, and

179

will behave accordingly. Peer pressure is an important factor in the political socialisation of young adults (Langton, 1967; Langton and Karns, 1969; Tedin, 1980; Boehnke et al., 1998; Pancer et al., 2007; McDevitt and Kiousis, 2007). Perceptions of civic norms regarding participation also matter. The more individuals feel that voting is a civic duty, widely valued by other citizens, the more likely they are themselves to participate (Clarke et al., 2004, 2009). To some extent, these perceptions of civic norms reflect widespread social beliefs and ideologies. But they also rest on more immediate assessments of what individuals’ peer groups might think. This latter can be affected by conversations and contacts between friends and family (on the political impact of contextual effects in political participation generally, see e.g. Cox, 1969; Agnew, 1987, 1996; Books and Prysby, 1991; Huckfeldt and Sprague, 1995; Zuckerman, 2005; Zuckerman et al., 2007; Mutz, 2006; Johnston and Pattie, 2006; Pattie and Johnston, 1999, 2000, 2001, 2008). Thanks to population mobility and the relative ease of long-distance communications, many of these discussions take place over geographically highly dispersed networks. But even in modern societies, a very substantial part of most people’s social networks remains quite local (Johnston and Pattie, 2011). In a study of conversation partnerships in a US city, Baybeck and Huckfeldt (2002, 265) found that 33% of discussion partners live less than a kilometre apart, and three-quarters live within 10 km of each other. Other studies report similar levels of proximity between discussion partners: for instance, Eagles et al. (2004) found that discussants in their study lived on average 2.8 miles apart. Many conversations between citizens therefore reinforce local as well as society-wide norms and expectations. Furthermore, most individuals are surprisingly knowledgeable about their local political environments, and relevant information on them is garnered not only from conversations with other local residents but also from more diffuse observations (general knowledge, the local media, local party campaign posters and leaflets, local election results, casual encounters with relative strangers in the community, and so on: Burbank, 1995; Baybeck and McClurg, 2005; Mutz, 1998; Cho and Rudolph, 2008). This matters as many political activities, from voting in a constituency to joining a local pressure group or campaign, depend for their effectiveness on specifically local participation. If one is thinking about whether to join a campaign opposing the closure of a local school, for instance, one might reasonably want to have some idea of how many others locally would be likely to join too before one made the final decision to participate, and this is likely to be more important in that particular instance than a sense of whether people in the country as a whole valued political participation. The literature on get-out-the-vote experiments sheds some interesting light here too, suggesting that individuals are responsive to whether others in their local communities are likely to take part in politics, and to what they think others locally might think of them if they decide not to get involved. Experimental results suggest that get-out-the-vote messages which also contain information on local participation rates are more effective than those which simply stress the importance of participation: individuals are more likely to take note if they are told that others locally are likely to participate too. Even more effective were messages which pointed out that individual voting was a matter of public record, hence allowing others in the locality to check on an individual’s compliance. But most effective of all were messages which encouraged voting but also threatened to name in the local press those individuals who did not vote: individuals exposed to such material were the most likely to participate, and the effect was relatively long-lasting (Gerber et al., 2008; Davenport et al., 2010).

180

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

The potential shame of being exposed locally as flouting the civic norm of voting proved a powerful spur to participation (Panagopoulos, 2010; Gerber et al., 2010). However, some individuals are liable to be more influenced by direct mobilisation and by civic norms than others. For instance, where people find themselves in a minority locally, their likelihood of voting will depend on their prior partisanship (McClurg, 2006). Strong partisan supporters of a political party will ignore their minority status and will tend to vote. Weak partisans and the non-partisan, however, may react to being in the minority by withdrawing from political engagement entirely.1 Furthermore, most studies of local context’s impact on political participation have focussed on voting in elections. But there is no necessary reason why it should not also have the same bearing on other forms of political action; indeed, it may be more efficacious in encouraging some forms of participation than others. There is still much to find out. In this paper, therefore, we move the debate on by looking at the role of civic norms and local mobilisation on individuals’ chances of participating in a range of different forms of political action other than voting, taking into account their personal characteristics that existing theories suggest provide important influences on citizen engagement decision-making. 3. Delimiting political participation and local influences To analyse the impact of local community influences on political participation, we employ the 2010 British Election Study (BES) face-to-face post-election cross-section survey. The BES asked respondents about their actual or potential participation in a variety of political activities. Most obviously, it asked whether they had voted in the 2010 General Election: 77% reported doing so. Clearly, this is higher than the actual election turnout of 65%. Some of the discrepancy can be explained by some respondents misreporting their behaviour by giving the interviewers what they felt was the socially desirable answer, but the most important cause is the difficulty surveys face in getting non-voters to participate (Swaddle and Heath, 1989). In addition, respondents were also asked to assess how frequently they voted in General Elections: 48% claimed to vote in every election, 24% said they did so in most, 11% said they voted in ‘some’, 8% claimed to vote in ‘not very many’, and 9% said they never voted in General Elections.2 We can therefore obtain some insights not only into who voted in 2010, but also into who had the voting habit more generally. Voting is only one aspect of political participation, however, and may be influenced by rather different factors to other forms of activity. Helpfully, the BES survey also asked respondents whether or not they had volunteered at some point over the previous few years to take part in either community affairs or local politics: only 13% reported doing so.3 People are more likely to participate in activities – such as voting – which are relatively easy than in those – like voluntary activity – which require greater investment of personal effort (Pattie et al., 2004). BES respondents were also asked to consider how likely, on an 11-point scale running from ‘no chance’

1 Strength of partisanship is measured here using responses to a set of questions asking if individuals generally thought of themselves as identifying with a particular, named party and, if they did, how strongly they identified with it. Here, we concentrate on the strength of partisan identification. Individuals were placed in one of four groups: those who identified very strongly with a political party (coded 3); those whose identified with it fairly strongly (coded 2); those whose identification was not very strong (coded 1); and those who did not identify with any party (coded 0). 2 Respondents were asked: ‘‘Since you have been eligible to vote in general elections, have you voted in: all general elections; most of them; some of them; not very many of them; or, none at all?’’ 3 The question was: ‘‘Over the past few years, have you volunteered to get involved in politics or community affairs?’’

(scored at 0) to ‘definite’ (scored 10), they thought they would participate in future in a variety of other activities ranging from voting in EU and local elections to engaging in acts of ethical consumption.4 Their responses give insights into their participation potential (Barnes and Kaase, 1979). Respondents were most likely to think they would vote in future elections or discuss politics with friends and families (Fig. 1). On the average, they thought it very unlikely indeed that they might consider helping a political party – either financially or as a volunteer. Other possible forms of participation, such as ethical consumption, protest activity or collective action, lay between these extremes. The participation potential measures fall into three distinct groups.5 One group takes in the likelihoods of engaging in acts of ethical consumption, taking part in a demonstration, and of working collectively with a like-minded group. The second group contains the two potential voting measures. And the final group encompasses the chances of working for or donating money to a political party. Political participation is not all of a piece, therefore. Just because individuals are predisposed to some forms of engagement does not necessarily mean they will also be involved in others. Voters are not necessarily the same people as ethical consumers or party activists. Each form of participation needs to be considered in its own right. Rather than examining all nine participation potential measures individually, later analyses therefore look in more detail at one participation potential measure from each group: the potential for boycotting products, the likelihood of voting in an EU election and the likelihood of donating money to a political party.6 Contextual influences, meanwhile, can take a variety of forms. Here, we consider two different forms: the civic norms regarding participation – perceived or actual – associated with friends, family and the local community; and more formal attempts to mobilise individuals, either from political parties or by invitations to get involved in community affairs. Individual peer pressure is measured using three variables. Two involve respondents’ perceptions of the norms within their local communities. They were asked whether they agreed that their family and friends thought voting a waste of time and that most people in their community voted.7 Responses to both questions reveal that most individuals think they are surrounded by others who vote: 71% disagree with the first proposition, and the same proportion agree with the second. But substantial minorities think their local environments are relatively devoid of voters: 18% say their friends and families see voting as a waste of time, and 11% think they live in an area where most are not voters. In addition to these general contextual perceived norms, individuals can be mobilised more directly by those they know via (for instance) conversation networks. To capture the extent to which individuals are directly encouraged to participate by those

4 In this case, respondents were asked: ‘‘Let’s think about the next few years. Using a scale from 0 to 10, where 0 means very unlikely and 10 means very likely, how likely is it that you will. . .’’ take part in a named political activity. The activities were: voting in the next EU Parliament election; voting in the next local government election; working actively with a group of people to address a public issue or solve a problem; participating in a protest to show concern about a public issue; donating money to a political party; working for a party or candidate in an election campaign; joining a consumer boycott of a product or store; deliberately buying a product for political, ethical or environmental reasons; or discussing politics with family or friends. Each respondent was asked separately about the likelihood of their taking part in each activity. 5 This is confirmed by the results of a principal components analysis (not reported here: details available from the lead author), which produces three components, together accounting for around two-thirds of the variation in the original variables. 6 These are the three participation potential measures which load most strongly on each of the three components identified by the PCA referred to in Footnote 1. 7 The questions were phrased: ‘‘Please tell me how far you agree or disagree with each of the following statements: Most of my family and friends think that voting is a waste of time; Most people around here usually vote in general elections’’. Responses were coded on a 5-point scale, from 1 (strongly agree) to 5 (strongly disagree).

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

181

Fig. 1. Likelihood of taking part in different political activities. (Source: 2010 BES face-to-face post-election cross-section.)

they talk to, we use a question asking respondents if people in different groups (friends, family members, co-workers or others) had tried to convince them how to vote in 2010.8 The great majority – 84% – said no-one had done so: 14% said they had received encouragement to vote from someone in at least one of the groups, and the remainder said they had been on the receiving end of the efforts of people from more than one group. More formal attempts to mobilise people come from the political parties via their election campaign activities. Over half (54%) of the BES respondents said they had been contacted in some way by at least one political party during the 2010 election campaign, suggesting this is a potentially important avenue for mobilisation. Much of that contact was via election literature (almost all of those who said they had been contacted by a party during the 2010 election campaign reported seeing leaflets), a considerable quantity of which probably went straight from the letterbox to the bin without being read in any real depth.9 That said, some respondents (21%) also reported more personalised contact, through canvassing, phone calls or other media. Finally, and outside elections, individuals can be invited – whether by friends and contacts or by other individuals – to take part in some community or local political activity: 18% of the BES respondents said they had received at least one such invitation in the previous few years.

4. Controlling for other influences on participation To understand the relative impacts of civic norms and direct mobilisation on political participation, however, we need to take into account other influences on participation. The analyses which follow therefore include a variety of control variables, reflecting five of the main theories of political engagement (for a summary, see Pattie et al., 2004, chapter 5). The first, the socio-economic status account, emphasises the importance of resources: the basic argument is that the resource-rich are more likely to participate, and the resource-poor less likely to do so (Parry et al., 1992). Resources take a variety of forms. They can be economic: wealth – 8 The wording was: ‘‘Did anyone, for example, a friend, a member of your family, someone at work, or some other person try to convince you which party to vote for in the recent general election?’’ 9 Official figures from the UK Electoral Commission show that nearly 75% of all local campaign spending during the 2010 ‘long campaign’ (from 1 January to 12 April 2010) went on advertising – mostly on posters – and leaflets, as did just under 85% of the equivalent spending in the short campaign (from 12 April to 6 May 2010: Johnston et al., 2011).

roughly indexed here by social class – can buy access, confidence and influence.10 They can be intellectual: the ability to weigh up, marshal and utilise information can facilitate participation, whereas the lack of such skills can be a hindrance and can undermine individuals’ confidence in their abilities to cope. Here, we use respondents’ highest formal educational qualifications as an indicator. Four levels of educational attainment are assessed: degree-level qualifications; post-school qualifications below degree level; secondary school level qualifications; and no formal qualifications. In addition, time is a resource for participation. One aspect of this is related to age: older people are generally more likely to participate than the young (though this trend reverses for the very old and frail). Retired people, for instance, can have more time available for participation than their younger counterparts. Our second model of participation stresses relative deprivation as a spur to participation (Runciman, 1966; Gurr, 1970). The claim is that where people think they are receiving less than they deserve, they will take action to secure what they see as their just deserts. The greater individuals’ sense of relative deprivation, therefore, the more likely they should be to engage in politics. Here, perceptions of relative deprivation are measured using individuals’ level of agreement with two propositions: that government ‘generally treats people like me fairly’; and that ‘there is often a big gap between what people like me expect out of life and what we actually get’. Responses to both are coded on a fivepoint Likert scale, from strongly agree (coded 1) to strongly disagree (coded 5). Third, we control for rational choice motivations (Downs, 1957; Riker and Ordeshook, 1968; Dunleavy, 1991). At its simplest, rational choice reduces the decision to participate to a calculation of relative benefits and costs. In these analyses, two forms of benefits are taken into account: benefits to the individual (measured by how much they agree with the proposition that politics is a good way of getting benefits for themselves and their families); and collective benefits (this time, using the proposition that politics is a good means by which to obtain benefits for ‘groups that

10 Class is defined here using a sixfold schema based on the work of John Goldthorpe and widely used in studies of British society (Goldthorpe, 1987), which reflects gradations of status and autonomy in the workplace. The classes are: the salariat (the middle class professions); routine non-manual workers (RNM: routine clerical and office jobs); the petty bourgeoisie (the self-employed); supervisors (manual workers with some degree of control over other workers – foremen and supervisors, for instance); manual workers (the manual working class); and ‘other’ occupations which do not fit into the schema.

182

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

people care about’, such as pensioners). Perceptions of the costs of participation, meanwhile, are indexed by responses to the proposition that ‘it takes too much time and effort to be active in politics and public affairs’. Responses to all three questions are measured on the same five-point Likert scale as before. Advocates of social capital theories argue that participation in shared community activities (many of which will be social rather than political) generates a virtuous circle of growing trust in other members of society and hence of further engagement (Putnam, 1993, 2000). Two variables control for social capital here. The first records how trustworthy respondents thought ‘most people’ were: answers were on an 11-point scale, from 0 (‘‘most people can’t be trusted’’) to 10 (‘‘most people can be trusted’’). The second variable measures how active respondents were ‘over the past few years’ in voluntary organisations such as community associations, charities or sports clubs (hence covering both more political and more social sorts of organisations): four possible levels of self-assessed activity were possible, ranging from ‘very active’ (coded 1) to ‘not at all active or not involved’ (coded 4). The final set of control measures takes into account the civic voluntarism model (Brady et al., 1995; Verba et al., 1995). Resources play an important part in this model (and are already controlled for). But in addition, it emphasises the role of other factors. Individuals’ sense of personal efficacy matters. (It is measured here by respondents’ assessments of how much influence they believe they have on politics, on an 11-point scale where high scores indicate a great deal of influence.) The more influence they think they have, the more likely they are to participate. Similarly, civic voluntarism points to individuals’ sense of engagement with politics. We assess it using two measures: respondents’ self-reported interest in politics in general (answers range from ‘a great deal’ of interest, coded 1, to ‘none at all’, coded 5); and the strength of their partisan identification (coded from 0 for those who say they do not identify with any party to 3 for those who say they identify ‘very strongly’ with a party: for the purposes of this paper, the partisan direction of their attachment – which party they identify with – is unimportant). Mobilisation – whether or not individuals are actively approached to take part in political activities – is also an important element of the civic voluntarism model but as this is the particular focus of the paper, it is treated separately from the rest of the model. The core of our analyses is shown in six separate models, one for each of the forms of actual and potential participation identified above. Table 1 reports two logistic regression models, one each for actual turnout at the 2010 General Election and for volunteering to take part in community activities. OLS regression is used in Table 2 to model self-reported frequency of turnout in General Elections. And Table 3 reports three further OLS models for our three participation potential measures: boycotting a product; voting in a future EU election; and donating money to a political party. By and large, the independent variables perform as expected, though they do indicate that different forms of political engagement are driven by different sorts of factors. The patterns are of sufficient interest to warrant separate discussion before moving onto the central results on civic norms and mobilisation. The civic voluntarism model has the most consistent effects across all six of the forms of participation considered here. Other things being equal, the more individuals felt they could influence politics and the more interested in it they were, the more likely they were to vote or to volunteer (Table 1), the less often they abstained (Table 2), and the more likely they were to consider boycotting a product for a political end, voting in a future EU election, or giving money to a political party (Table 3). Understandably, however, strength of partisanship was associated with those activities related to elections and party politics, but not with forms of engagement outside of the repertoire of electoral competition.

Strong partisans were more likely to vote, less likely to abstain, and more likely to consider voting in an EU election than were the non-partisan but they were no more likely to volunteer for community work or to consider engaging in ethical consumption. The other theories were less consistent in their influence. Once other factors were controlled for, the relative deprivation, rational choice and social capital models all failed to account for actual voting in 2010, for instance. And where they did have effects, the nature of the impact varied. For instance, relative deprivation explanations also did nothing to account for the likelihood of donating to a party. But they were related to other forms of participation. Other things being equal, the greater an individual’s sense of relative deprivation, the more likely they were to report abstaining frequently, and the less likely they were to consider voting in future EU elections. The less well-served people felt by government and society, therefore, the less likely it was that they would take part in conventional politics. However, a sense of relative deprivation did encourage participation in non-electoral forms of action, such as community volunteering or ethical consumption. The implication is that at least some of those who thought society treated them unfairly eschewed state-focussed activities like voting, taking matters into their own hands instead. If one cannot trust the state to play fair, one may feel better served by doing things for oneself. Rational choice accounts, meanwhile, fared relatively poorly. In most models, only one (or in the case of actual voting, none) of the coefficients proved significant and more often than not it was the collective benefits measure which came through. The less effective people thought politics was as a means of obtaining benefits for socially vulnerable groups, the less likely they were to volunteer in their communities, or to consider engaging in a consumer boycott, donating to a political party, or voting in a future EU election.11 Perceptions of personal benefits, meanwhile, proved significant in only one model: the frequency of abstention. The less effective people thought politics was as a means of obtaining personal benefits, the more likely they were to abstain frequently. Contrary to rational choice expectations, the perceived cost of participation (at least as indexed by anxieties over the time participation might take) was insignificant in four out of the six models. The two exceptions are in the participation potential models for voting an EU election and for donating to a political party. In both cases, the greater the sense that participation would be costly, the less likely participation became. Social capital accounts were similarly patchy in their capacity to account for participation. Trust in others was associated only with frequency of abstention (greater trust was associated with less frequent abstention – i.e. more frequent voting) and with the prospects for engaging in ethical consumption (higher levels of trust were related to a greater chance of engaging in a consumer boycott – presumably in part because greater trust in others fostered a sense that others would also participate, enhancing a boycott’s potential impact). Respondents’ levels of activity in their local communities, meanwhile, were also related to taking part in a boycott (the lower the level of reported community involvement, the lower the probability of engaging in a boycott), and also to volunteering for local politics (the less active were less likely to volunteer than the more active). But social capital was not a good discriminator of those who did vote in 2010, or of those who might

11 Elections to the European Parliament are classic second-order contests: the stakes are relatively low compared to elections to Westminster, turnout is invariably small, and the elections tend to produce ‘surprise’ results as those few who do vote use them as an opportunity to vent their feelings regarding the incumbent national government (Marsh, 1998). They therefore present a stiffer test for public preparedness to vote than do Westminster elections as most voters see them as unimportant and irrelevant.

183

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

consider voting in an EU election or giving money to a political party. At first glance, the socio-economic and demographic factors associated with the resources model seem equally sporadic in their influences. Age, for instance, is significant (and correctly signed) in only half of the models, while education plays an independent part in only two (consumer boycotting and voting in an EU election are both encouraged by education). But on closer inspection, much of their effects on participation are mediated through their impact on other factors taken into account via some of the other models included here. For instance, interest in politics and a sense of political efficacy are both related to age, class and education: other things being equal, older, more middle class and more formally qualified individuals feel more personally powerful and have a greater interest in politics than younger, more working class and less formally educated individuals. In analyses (not reported here) which include only gender, age, class and education, all apart from gender prove consistently significant: the older, more middle class, or better-educated an individual is, the more likely it is that he or she will engage in each of the six political activities examined in this paper. Gender, meanwhile, proves insignificant in five out of

the six models. The exception is that for working on behalf of a party: women are less likely to report considering doing this than are men. 5. Modelling the impact of civic norms and mobilisation These results confirm that, while there are some general patterns underlying many forms of political participation, there are also discontinuities. A sense of relative deprivation is more likely to encourage activities which involve taking the initiative oneself than those which involve choosing others to do things on one’s behalf, for instance. And rational choice motivations are more consequential for forms of participation which involve a substantial investment of personal effort than for those which are easier to carry out – though even here there are important nuances, as revealed by the opposing effects of personal and collective benefits on the likelihood of donating money to a political party. We might therefore anticipate that the effects of perceived civic norms and direct mobilisation on political participation will also vary depending on the form of political activity being considered. And so it proves. Both measures of contextual civic norms prove

Table 1 Accounting for General Election turnout and community volunteering, 2010 II: logistic regressions (standard errors in brackets). Source: 2010 BES face-to-face cross-section. Turnout, 2010 election

Female Age

0.19 (0.13) 0.03 (0.00)**

0.08 (0.16) 0.01 (0.01)

Education (comparison = no qualifications) School qualifications Post-school qualifications Degree-level qualifications

0.18 (0.19) 0.15 (0.19) 0.29 (0.22)

0.28 (0.29) 0.09 (0.29) 0.19 (0.31)

Class (comparison = salariat) RNM Petite bourgeoisie Supervisors Manual Other

0.35 0.11 1.38 0.66 0.51

0.44 1.27 0.80 0.11 0.19

Relative deprivation Government treats people fairly (high = strongly disagree) Big gap between expectations and rewards (high = strongly disagree)

0.09 (0.07) 0.00 (0.06)

0.20 (0.09)* 0.12 (0.08)

Rational choice Politics a good way to get family benefits (high = strongly disagree) Politics takes too much time and effort (high = strongly disagree) Politics a good way to get group benefits (high = strongly disagree)

0.03 (0.07) 0.03 (0.07) 0.04 (0.08)

0.07 (0.08) 0.01 (0.08) 0.20 (0.10)*

Social capital Can most people be trusted (high = can be trusted) How active in local groups? (high = not at all active)

0.06 (0.03) 0.05 (0.06)

0.08 (0.04) 0.40 (0.07)**

Civic voluntarism Influence on politics (high = a great deal of influence) General interest in politics (high = none at all) Strength of party identification (high = very strong)

0.14 (0.03)** 0.67 (0.07)** 0.16 (0.06)**

0.12 (0.03)** 0.19 (0.09)* 0.16 (0.09)

Civic norms Friends and family think voting waste of time (high = strongly disagree) Most people round here vote (high = strongly disagree)

0.30 (0.06)** 0.27 (0.08)**

0.13 (0.09) 0.17 (0.10)

Mobilisation Someone tried to convince respondent how to vote Respondent invited to take part in community politics Respondent canvassed by any party during election Constant Model improvement Significance Nagelkerke R2 % Correctly classified N * **

Volunteering for local politics/ community affairs

Significant at p = 0.05. Significant at p = 0.01.

0.45 0.00 0.58 0.66 556.85 0.00 0.33 83.1 2273

(0.19) (0.33) (0.35)** (0.19)** (0.29)

(0.15)** (0.19) (0.12)** (0.89)

0.04 2.57 0.12 3.97 668.04 0.00 0.45 89.2 2274

(0.23) (0.47)** (0.65) (0.23) (0.37)

(0.13) (0.16)** (0.16) (0.91)**

184

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

significant in the voting-related equations. In line with the neighbourhood effect literature, there is evidence that individuals are influenced by the activities of those around them, bringing their own behaviour into line with what they perceive as the local norm. The more people disagree with the proposition that their friends and families think voting is a waste of time and the more they think that most people in their area vote, the more likely they themselves are to have voted in the 2010 UK General Election (Table 1) and to consider voting in a future EU election (Table 3), and the less likely are they to say they abstain often rather than vote in most elections (Table 2). Experiencing directly applied attempts at mobilisation by friends and acquaintances, in the form of whether or not respondents had encountered someone who tried to persuade them how to vote in 2010, was linked to only one of the voting-related forms of participation, however. Those who reported some of their acquaintances had tried to persuade them how to vote in 2010 were more likely to have voted in that election than were those who reported no such attempt at personal persuasion (Table 1). But whether or not respondents had experienced personal attempts at persuasion in 2010 made no real difference to their expressed intentions of voting in a future European contest or to the self-reported frequency with which they abstained (Tables 3 and 2 respectively). That is hardly surprising. Why, for instance,

should a discussion with an acquaintance concerning the most recent UK General Election have any bearing on one’s chances of voting in a future Europe-wide ballot (a vote, moreover, which in 2010 was still 4 years in the future)? And why should pressure to vote from acquaintances in 2010 make any difference to how often one had or had not voted in past elections. Where the perceived influence of other citizens was generalised in the form of civic norms (most people normally vote; my friends and family thinking voting is a generally good idea), it had effects in all three voting models. But peer pressure focussed purely on the 2010 campaign only had a bearing on the likelihood of voting at that contest. There are general and long-standing forms of (perceived) influence from other citizens and there are other, more immediate and time-specific forms. As here, the former are liable to have generalised effects whereas the latter are likely to have more limited and focussed impacts. That said, being on the receiving end of an attempt by an acquaintance to persuade one how to vote in 2010 was also positively associated with the potential for taking part in ethical consumption activities, in the form of engaging in a consumer boycott (Table 3). This may link to more formalised forms of mobilisation, which are another ‘success story’ among the possible influences on participation. In every equation, at least one mobilisation measure proves significant and correctly signed. And in all

Table 2 Accounting for frequency of abstention in General Elections: OLS regressions (standard errors in brackets). Source: 2010 BES face-to-face cross-section. Frequency of abstaining in General Elections (high = never vote) Female Age

0.02 (0.05) 0.02 (0.00)**

Education (comparison = no qualifications) School qualifications Post-school qualifications Degree-level qualifications

0.13 (0.07) 0.07 (0.07) 0.02 (0.08)

Class (comparison = salariat) RNM Petite bourgeoisie Supervisors Manual Other

0.16 0.07 0.33 0.09 0.28

Relative deprivation Government treats people fairly (high = strongly disagree) Big gap between expectations and rewards (high = strongly disagree)

0.02 (0.03) 0.07 (0.02)**

Rational choice Politics a good way to get family benefits (high = strongly disagree) Politics takes too much time and effort (high = strongly disagree) Politics a good way to get group benefits (high = strongly disagree)

0.05 (0.03)* 0.00 (0.02) 0.04 (0.03)

Social capital Can most people be trusted (high = can be trusted) How active in local groups? (high = not at all active)

0.04 (0.01)** 0.01 (0.02)

Civic voluntarism Influence on politics (high = a great deal of influence) General interest in politics (high = none at all) Strength of party identification (high = very strong)

0.06 (0.01)** 0.25 (0.03)** 0.11 (0.02)**

Civic norms Friends and family think voting waste of time (high = strongly disagree) Most people round here vote (high = strongly disagree)

0.18 (0.02)** 0.15 (0.03)**

Mobilisation Someone tried to convince respondent how to vote Respondent invited to take part in community politics Respondent canvassed by any party during election Constant R2 N * **

Significant at p = 0.05. Significant at p = 0.01.

0.08 0.05 0.19 3.11 0.31 2310

(0.07)* (0.11) (0.15)* (0.07) (0.11)*

(0.05) (0.06) (0.05)** (0.27)**

185

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

three of the participation potential models reported in Table 3, both measures of formal mobilisation are significant. Where people are approached and asked to take part in political activities, they are much more likely to do so (or, in the case of participation potential, to consider doing so in the future) than are those who are not approached. Being invited matters: where general exhortations may be insufficient, targeted, personal requests for help seem to be harder for individuals to ignore. The implications for policy agendas such as British Prime Minister David Cameron’s Big Society initiative are clear. Attempts to encourage greater public involvement in community life are liable to depend heavily for their success not only on resources but also on their capacity to employ direct and personal invitations for individuals to get involved. General exhortations alone are much less likely to be effective (Pattie and Johnston, 2011). Comparing the models for actual participation reported in Tables 1 and 2 with Table 3’s participation potential measures reveals some intriguing differences, however. Actual participation seems to be affected by specific types of invitation. Reporting being canvassed by a party is associated with a heightened chance of turning out in 2010 and a lowered chance of being a frequent abstainer. But it is not related to individuals’ chances of undertaking voluntary or community work. Conversely, those who report being specifically asked to get involved in community politics are much more likely than those who are not asked to then actually do so (indeed, this is just about the most important single factor in Table 1’s

volunteering model). But being invited to take part in community politics has no independent effect on respondents’ chances of actually voting. Contrast this with the participation potential models in Table 3 where, as noted above, both forms of formal invitation – being invited to take part in community politics and being canvassed by a party at the 2010 election – are positively associated with the perceived likelihood of taking part in all three forms of political activity. This implies that while actual participation is encouraged by specific and targeted mobilisation, the abstract idea of participating in any way in future can be strengthened by many forms of contact. There would appear to be a sort of halo effect of all forms of mobilisation on participation potential in general.

6. Differential susceptibility to civic norms and mobilisation Civic norms and direct mobilisation do not affect all individuals in the same way, however. The theoretical roots of this idea can be traced all the way back to early accounts of voting behaviour. In a pioneer study of voting in one community during the 1940 US Presidential election, Lazarsfeld et al. (1948: see also Bereleson et al., 1954) showed that personal contacts between voters were influential in individuals’ voting decisions: people listened and reacted to the views of others. However, they also showed that relevant information flows between individuals were asymmetric.

Table 3 Accounting for different types of potential participation, 2010 II: OLS regressions (standard errors in brackets). Source: 2010 BES face-to-face cross-section. Boycott product

**

Donating to political party

Female Age

0.13 (0.14) 0.01 (0.00)

Education (comparison = no qualifications) School qualifications Post-school qualifications Degree-level qualifications

0.36 (0.21) 0.66 (0.21) 0.57 (0.24)

Class (comparison = salariat) RNM Petite bourgeoisie Supervisors Manual Other

0.19 0.15 0.70 0.50 0.34

Relative deprivation Government treats people fairly (high = strongly disagree) Big gap between expectations and rewards (high = strongly disagree)

0.35 (0.08)** 0.08 (0.07)

0.16 (0.08)* 0.16 (0.07) *

0.02 (0.05) 0.01 (0.05)

Rational choice Politics a good way to get family benefits (high = strongly disagree) Politics takes too much time and effort (high = strongly disagree) Politics a good way to get group benefits (high = strongly disagree)

0.11 (0.07) 0.07 (0.07) 0.21 (0.08)*

0.07 (0.07) 0.14 (0.07)* 0.21 (0.09)*

0.09 (0.05) 0.13 (0.05)** 0.12 (0.06)*

Social capital Can most people be trusted (high = can be trusted) How active in local groups? (high = not at all active)

0.18 (0.04) 0.13 (0.06)

0.06 (0.04) 0.06 (0.06)

0.03 (0.02) 0.00 (0.04)

Civic voluntarism Influence on politics (high = a great deal of influence) General interest in politics (high = none at all) Strength of party identification (high = very strong)

0.17 (0.03)** 0.30 (0.08)** 0.02 (0.07)

0.16 (0.03)** 0.93 (0.08)** 0.18 (0.07)**

0.25 (0.02)** 0.12 (0.05)* 0.11 (0.04)*

Civic norms Friends and family think voting waste of time (high = strongly disagree) Most people round here vote (high = strongly disagree)

0.13 (0.07) 0.10 (0.09)

0.37 (0.07)** 0.19 (0.09) *

0.01 (0.05) 0.01 (0.06)

Mobilisation Someone tried to convince respondent how to vote Respondent invited to take part in community politics Respondent canvassed by any party during election Constant R2 N *

Vote in EU election

Significant at p = 0.05. Significant at p = 0.01.

0.49 0.79 0.36 1.13 0.14 2288

0.07 (0.14) 0.00 (0.00)

** *

(0.20) (0.31) (0.45) (0.20)* (0.33)

(0.14) (0.18) (0.14) (0.81)

** *

** ** **

0.39 (0.21) 0.50 (0.21) 0.32 (0.24) 0.16 0.53 0.68 0.53 0.35

0.12 0.59 0.38 6.60 0.24 2283

0.12 (0.09) 0.01 (0.00)*

*

(0.20) (0.32) (0.45) (0.20)** (0.33)

(0.14) (0.19) (0.14) (0.82)

** ** **

0.12 (0.14) 0.00 (0.14) 0.07 (0.16) 0.05 0.31 0.09 0.12 0.19

0.16 0.46 0.23 0.24 0.12 2315

(0.13) (0.21) (0.30) (0.13) (0.22)

(0.09) (0.12) (0.09) (0.53)

** *

0.15 0.32 0.23 0.24 0.01 1.35 1.46 0.01 0.07 0.17 0.26 0.09 0.07 0.04 0.09 0.20 0.39 (0.35)* (0.33) (0.24)* (0.29) (0.46) (0.55) (0.48) (0.07) * (0.08) (0.15) (0.19) (0.13) (0.06) (0.08) (0.13) (0.18) (0.14) 0.75 0.63 0.63 0.27 0.40 0.38 0.06 0.16 0.05 0.09 0.23 0.18 0.01 0.14 0.15 0.11 0.05 (0.34) (0.33) (0.24) (0.29) (0.45)** (0.54)* (0.47) (0.07) (0.09) (0.14) (0.18) (0.13) (0.06) (0.08) (0.13) (0.18)* (0.13) 0.64 0.44 0.36 0.08 1.31 1.32 0.79 0.08 0.15 0.21 0.18 0.03 0.11 0.04 0.18 0.39 0.13 (0.12) (0.12)** (0.08) (0.10) (0.15) (0.18) (0.16) * (0.02) (0.03) (0.05) (0.06) (0.05)** (0.02)* (0.03) (0.04) (0.06) (0.05) 0.18 0.36 0.07 0.03 0.11 0.33 0.38 0.01 0.00 0.05 0.10 0.17 0.04 0.04 0.05 0.09 0.04 (0.54) (0.49)* (0.29) (0.30) (0.41) (0.52)** (0.52) (0.10) (0.11) (0.15) (0.17)* (0.18)* (0.09) (0.10) (0.14) (0.18) (0.18) 0.65 1.01 0.08 0.08 0.39 1.43 0.72 0.09 0.04 0.03 0.38 0.41 0.11 0.08 0.17 0.17 0.07 (0.28) (0.32) * (0.24) (0.29) (0.63) (0.60) (0.49) * (0.06) (0.07) (0.15) (0.18)* (0.12)* (0.06) (0.08) (0.16) (0.18) (0.13)

**

*

Significant at p = 0.05. Significant at p = 0.01.

0.37 0.81 0.39 0.46 0.79 0.75 1.18 0.09 0.07 0.21 0.45 0.26 0.02 0.03 0.03 0.04 0.03 Partisanship Interest in politics Friends think voting waste of time Most people vote Acquaintance tried to persuade respondent Invited to volunteer Canvassed Partisanship  friends think voting waste of time Partisanship  most people vote Partisanship  acquaintance tried to persuade respondent Partisanship  invited to volunteer Partisanship  canvassed Interest in politics  friends think voting waste of time Interest in politics  most people vote Interest in politics  acquaintance tried to persuade respondent Interest in politics  invited to volunteer Interest in politics  canvassed

Vote in EU election Boycott product OLS regression

Frequency of abstaining in General Elections (high = never vote) Turnout 2010 election

Volunteering for local politics/community affairs Logistic regression

Table 4 Interaction effects in the operation of civic norms and voter mobilisation: interactions with partisanship (standard errors in brackets).

(0.22) (0.21) (0.16) (0.19) (0.30) (0.36)** (0.31)** (0.04) (0.05) (0.09) (0.12)* (0.09) (0.04) (0.05) (0.08) (0.12) (0.09)**

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

Donating to political party

186

Some individuals, usually strongly partisan, were, in effect, local ‘opinion leaders’, who acted for their peers as ‘interepreters’ of political stories in the mass media and whose views other less partisan and politically engaged individuals, found influential. This ‘two-step’ theory of political communication implies that some citizens (the less partisan) are open to influence by others (the more partisan ‘opinion leaders’), and hence that the influence of local contextual factors are liable to be different for different groups of voters. We therefore expect that perceptions of civic norms and exposure to direct mobilisation campaigns should have a greater influence on those who are less politically engaged than on those who are more so. The latter group will generate their own reasons to participate, almost irrespective of what is happening around them. The former, meanwhile, may need the extra stimulus provided by a sense of the social norms in their communities or by a direct invitation to get involved in order for them to overcome their relative reluctance to participate. Recent studies confirm that those who are less politically-motivated are more influenced by perceived norms and by direct mobilisation than those who are more overtly political in their outlook (McClurg, 2004, 2006). And experimental research has shown that ‘get-out-the-vote’ campaigns (especially when they emphasise that most people in the community do vote) have a larger positive impact in encouraging electoral participation among those who generally voted infrequently than among those who generally vote often (Gerber and Rogers, 2009). We can illustrate this by adding interactions between, on the one hand, the civic norms and mobilisation measures examined above and, on the other, strength of partisanship and general interest in politics to the regression models reported in Tables 1–3 (Table 4: only the relevant coefficients for variables involved in the interactions are reported). Most of the interactions proved insignificant. However, a consistent overall story emerges from most of those which were significant. Of the ten identified in Table 4, eight involved the effects of direct attempts to mobilise individuals, either via an invitation to volunteer or being canvassed by a party, and seven were related to partisanship (only three were interactions with interest in politics). As expected, party canvassing, has a greater impact on less partisan individuals than on the more partisan: the stronger individuals’ sense of partisanship, the weaker the mobilising effect of canvassing on turnout, volunteering and frequency of abstention; canvassing by the parties is more likely to influence and mobilise the less-engaged because strong partisans have for the most part already decided to vote. To get an idea of the relative size of these effects, the results reported in Table 4 can be used to predict how particular types of individuals should score on each of the forms of political action considered here. To focus on the interactions between partisanship and interest in politics on the one hand and civic norms and mobilisation on the other, we hold constant all other variables in the model except for those involved in the interaction terms. To do so, we consider a group of hypothetical citizens who are all 40 year old men with no formal qualifications but who are in middle class, salariat, occupations, whose scores on most of the relative deprivation, social capital and rational choice controls place them at the middle of the relevant scales, and who say no-one tried to persuade them how to vote in 2010. Among those who shared these characteristics, were in the middle of the ‘political interest’ scale, and who also reported that they did not support any political party, the probability of voting for those who were not canvassed by any party was just 0.56. But if they were canvassed, their probability of voting rose substantially to 0.79. The equivalent ‘canvassing bonus’ for those who were very strong partisans, while also positive, was much more muted: among this group, the probability of voting in 2010 increases from 0.76 if they were not canvassed to 0.81 if they

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

were. Graphs showing predictions from all the significant interactions are displayed in Figs. 2 and 3 (in these estimations, where we consider interactions with partisanship, we assume the individual has a middling interest in politics; where we consider interactions involving interest in politics, we assume the individual is non-partisan). In most cases, the same basic story is borne out: the less interested individuals are in politics, the more they need the extra stimulus provided by a personal invitation, a party canvass, or a sense that their peers think participation is worthwhile to encourage them to take part (Fig. 2). Not all of the interaction terms work in quite this way, however. Indeed, in some cases, being invited to take part increases the chances of participation by the more engaged to a greater extent than it does for the less engaged (Fig. 3). Being invited to take part in community politics is more likely to encourage the strongly partisan to vote in 2010 and to volunteer for local politics than for the less partisan. Similarly, the more interested individuals were in politics, the more being invited to take part in a community initiative incentivised joining in with a consumer boycott. (The least

187

interested in politics, meanwhile, became less likely to consider joining a boycott when they were invited to take part in community politics. And being exposed to a party canvass made the most interested in politics more likely to consider donating to a party, but it made the least interested less likely to do so (perhaps by irritating them?) What unites most of these contrary effects is that they tend to involve potentially high-cost forms of political engagement, either by getting involved in community politics – a higher-cost activity for the individual, not least in terms of time involved, than being canvassed to vote – or giving financial resources to political parties (in the donation model). While the less partisan and politically uninterested may react positively to attempts to mobilise them to take part in relatively low-cost voting, it is the more partisan and political who react most positively to mobilisation aimed at more personally costly forms of action. And it seems that the impact of being personally invited to take part in community affairs or being canvassed by a party on one’s chances of making a donation to a political party are higher for those who are generally interested in politics than for those who are not.

Fig. 2. Significant interactions: civic norms and mobilisation encourage participation by less engaged.

188

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

Fig. 3. Significant interactions: civic norms and mobilisation discourage participation by less engaged.

7. Conclusions Perceived civic norms and mobilisation attempts are important spurs to actual and potential political engagement. Other things being equal, individuals who are personally invited to take part in some activity are more likely to do so than those who are not – and most of those invitations, whether formal or informal, are delivered locally, either by family members and friends in the decision-makers’ social networks or by organisations that operate within local contexts seeking support for their activities. Furthermore, the stronger their sense that others in their families and communities both take part in politics and value doing so, the greater their own propensity to join in. In part, this may reflect a desire to conform. Few like to be (seen to be) left out. But it also may reflect rational calculations. Only the most committed are likely to participate willingly if they think they are likely to be in a small and marginalised minority, for instance. While there is a case to be made for ‘demonstration effects’ (for instance, by making sure that all viewpoints are not only represented but are visibly seen to be so, in the hope of building further support in the future: Dunleavy, 1991, 79ff.), this places substantial demands on individuals, with little prospect of reward. Knowing there are others who value taking action too helps share the load. The effect is not indiscriminate, however. Mobilisation aimed at getting out the vote is likely to help achieve that end, but its impact on participation in other forms of political activity may be muted. Not surprisingly, people differentiate between invitations to specific action and more general and diffuse exhortations. While the former mobilise for the stated purpose, the latter are more easily discounted. In general, perceptions of local civic norms and direct mobilisation are particularly likely to encourage those generally less engaged by politics. Strong partisans and those interested in politics are already very likely to participate and do not require the extra stimulus to do so. Many of them would take part whether or not they were asked and irrespective of whether they think their

fellows are liable to join them. But for the less engaged, the sense that others with whom they are likely to identify value participation and have asked them to take part does have a tonic effect. People are more likely to participate in civic activities – the British Prime Minister’s so-called ‘Big Society’ – when invited to personally rather than respond to more general calls for action and involvement. Of particular note, we argue, is the finding that canvassing is particularly effective at mobilising the less politically engaged. In an era of low turnout elections, and with growing anxiety about the legitimacy of election results in consequence, this is particularly important. There is, it seems, no substitute for party organisation on the ground: being invited to participate in an election is, it seems, often the crucial final step towards actually doing so – but that pre-supposes there are people available to make the invitations. This presents the major parties with a challenge, however. In common with many western democracies, party membership is in serious long-term decline in the UK, and it is at the grassroots of the constituency party that this is most apparent. In many parts of the country, local parties are perilously small. No matter how dedicated the members, they are thinly stretched during election campaigns and are unlikely to be able to conduct a full canvass even of their potential supporters in an average-sized constituency. To some extent, voluntary work by sympathisers who are not party members can help plug the gap – but only to some extent. The results reported here imply something of a chicken and egg problem for those concerned about low levels of political participation, therefore. To encourage the less engaged to take part, it helps to have local party activists (whether or not they are party members) and local communities with shared civic norms encouraging participation. But where are they to come from if people will not get involved? Civic norms and local mobilisation clearly encourage participation. But they are themselves vulnerable and can be undermined. Do they provide us with a ‘virtuous circle’ towards greater engagement, or, if they collapse too, might we face a vicious spiral to ever-lower levels of participation? Either way, the

C. Pattie, R. Johnston / Geoforum 45 (2013) 178–189

local context within which people decide whether or not to act politically is liable to remain important. References Agnew, J., 1987. Place and Politics: The Geographical Mediation of State and Society. Allen and Unwin, London. Agnew, J., 1996. Mapping politics: how context counts in electoral geography. Political Geography 15, 129–146. Barnes, S., Kaase, M., 1979. Political Action. Sage, London. Baybeck, B., Huckfeldt, R., 2002. Spatially dispersed ties among interdependent citizens: connecting individuals and aggregates. Political Analysis 10, 261–275. Baybeck, B., McClurg, S.D., 2005. What do they know and how do they know it? An examination of citizen awareness of context. American Politics Research 33, 492–520. Bereleson, B.R., Lazarsfeld, P.F., Jennings, M.K., 1954. Voting: A Study of Opinion Formation in a Presidential Election. University of Chicago Press, Chicago. Boehnke, K., Hagan, J., Merkens, H., 1998. Right-wing extremism among German adolescents: risk factors and protective factors. Applied Psychology: An International Review 47, 109–126. Books, C.L., Prysby, J.W., 1991. Political Behavior and the Local Context. Praeger, New York. Brady, H.E., Verba, S., Schlozman, K.L., 1995. Beyond SES: a resource model of political participation. American Political Science Review 89, 271–294. Burbank, M.J., 1995. The psychological basis of contextual effects. Political Geography 14, 621–635. Cho, W.K.T., Rudolph, T.J., 2008. Emanating political participation: untangling the spatial structure behind participation. British Journal of Political Science 38, 273–289. Clarke, H., Sanders, D., Stewart, M., Whiteley, P., 2004. Political Choice in Britain. Oxford University Press, Oxford. Clarke, H., Sanders, D., Stewart, M., Whiteley, P., 2009. Performance Politics and the British Voter. Cambridge University Press, Cambridge. Cox, K., 1969. The voting decision in a spatial context. In: Board, C., Chorley, R., Haggett, P., Stoddart, D.T. (Eds.), Progress in Geography, vol. 1. Edward Arnold, London. Davenport, T.C., Gerber, A.S., Green, D.P., Larimer, C.W., Mann, C.B., Panagopoulos, C., 2010. The enduring effects of social pressure: tracking campaign experiments over a series of elections. Political Behavior 32, 423–430. Denver, D., 1995. Non-voting in Britain. In: Font, J., Virós, R. (Eds.), Electoral Abstention in Europe. Institut de Ciènces Politiques i Sociale, Barcelona. Denver, D., Hands, G., 1985. Marginality and turnout in general elections in the 1970s. British Journal of Political Science 15, 381–388. Denver, D., Hands, G., 1997. Modern Constituency Electioneering: Local Campaigning in the 1992 General Election. Frank Cass, London. Downs, A., 1957. An Economic Theory of Democracy. HarperCollins, New York. Dunleavy, P., 1991. Democracy, Bureaucracy and Public Choice: Economic Explanations in Political Science. Harvester Wheatsheaf, New York. Eagles, M., Bélanger, P., Calkins, H.W., 2004. The spatial structure of urban political discussion networks. In: Goodchild, M.F., Janelle, D.G. (Eds.), Spatially Integrated Social Science. Oxford University Press, Oxford. Franklin, M., 2004. Voter Turnout and the Dynamics of Electoral Competition in Established Democracies Since 1945. Cambridge University Press, Cambridge. Gerber, A.S., Rogers, T., 2009. Descriptive social norms and motivation to vote: everybody’s voting and so should you. The Journal of Politics 71, 178–191. Gerber, A.S., Green, D.P., Larimer, C.W., 2008. Social pressure and voter turnout: evidence from a large-scale field experiment. American Political Science Review 102, 33–48. Gerber, A.S., Green, D.P., Larimer, C.W., 2010. An experiment testing the relative effectiveness of encouraging voter participation by inducing feelings of pride or shame. Political Behavior 32, 409–422. Goldthorpe, J.H., 1987. Social Mobility and Class Structure in Modern Britain, second ed. Oxford University Press, Oxford. Green, D.P., Gerber, A.S., 2004. Get Out the Vote: How to Increase Voter Turnout. Brookings Institution Press, Washington, DC. Gurr, T.D., 1970. Why Men Rebel. Princeton University Press, Princeton, NJ. Hay, C., 2007. Why We Hate Politics. Polity Press, Cambridge. Huckfeldt, R., Sprague, J., 1995. Citizens, Politics and Social Communication: Information and Influence in an Election Campaign. Cambridge University Press, Cambridge. Jacobson, G.C., 1978. The effect of campaign spending in Congressional elections. American Political Science Review 72, 469–491. Jacobson, G.C., 1990. The effects of campaign spending in House elections: new evidence for old arguments. American Journal of Political Science 34, 334–362. John, P., Brannan, T., 2008. How different are telephoning and canvassing? Results from a ‘get out the vote’ field experiment in the British 2005 General Election. British Journal of Political Science 38, 565–574. Johnston, R.J., Pattie, C.J., 2006. Putting Voters in Their Place: Geography and Elections in Great Britain. Oxford University Press, Oxford. Johnston, R.J., Pattie, C.J., 2011. Social networks, geography and neighbourhood effects. In: Scott, J., Carrington, P. (Eds.), The Sage Handbook of Social Network Analysis. Sage Publications, London, pp. 301–311. Johnston, R.J., Pattie, C.J., Cutts, D., Fieldhouse, E., Fisher, J., 2011. Local campaign spending at the 2010 General Election and its impact: exploring what wider regulation has revealed. The Political Quarterly 82, 169–192.

189

Johnston, R.J., Cutts, D.J., Pattie, C.J., Fisher, J., 2012. We’ve got them on the list: contacting, canvassing and voting in a British general election. Electoral Studies 31. Langton, K.P., 1967. Peer group and school and the political socialisation process. American Political Science Review 61, 751–758. Langton, K.P., Karns, D.A., 1969. The relative influence of the family, peer group and school in the development of political efficacy. Western Political Quarterly 22, 813–826. Lazarsfeld, P., Berelson, B., Gaudet, H., 1948. The People’s Choice: How the Voter Makes Up His Mind in a Presidential Campaign. Columbia University Press, New York. Marsh, M., 1998. Testing the second-order election model after four European elections. British Journal of Political Science 28, 591–607. McClurg, S.D., 2004. Indirect mobilization: the social consequences of party contacts in an election campaign. American Politics Research 32, 406–443. McClurg, S.D., 2006. Political disagreement in context: the conditional effect of neighbourhood context, disagreement and political talk on electoral participation. Political Behavior 28, 349–366. McDevitt, M., Kiousis, S., 2007. The red and blue of adolescence: origins of the compliant voter and the defiant activist. American Behavioral Scientist 50, 1214–1230. Mutz, D.C., 1998. Impersonal Influence: How Perceptions of Mass Collectives Affect Political Attitudes. Cambridge University Press, Cambridge. Mutz, D.C., 2006. Hearing the Other Side: Deliberative Versus Participatory Democracy. Cambridge University Press, Cambridge. Nickerson, D.W., 2008. Is voting contagious? Evidence from two field experiments. American Political Science Review 102, 49–57. Norris, P., 2002. Democratic Phoenix: Reinventing Political Activism. Cambridge University Press, Cambridge. Norris, P., 2011. Democratic Deficit: Critical Citizens Revisited. Cambridge University Press, Cambridge. Panagopoulos, C., 2010. Affect, social pressure and prosocial motivation: field experimental evidence of the mobilizing effects of pride, shame and publicizing voting behaviour. Political Behavior 32, 369–386. Pancer, S.M., Pratt, M., Hunsberger, B., Alisat, A., 2007. Community and political involvement in adolescence: what distinguishes the activists from the uninvolved? Journal of Community Psychology 35, 741–759. Parry, G., Moyser, G., Day, N., 1992. Political Participation and Democracy in Britain. Cambridge University Press, Cambridge. Pattie, C.J., Johnston, R.J., 1999. Context, conversation and conviction: social networks and voting at the 1992 British General Election. Political Studies 47, 877–889. Pattie, C.J., Johnston, R.J., 2000. ‘People who talk together vote together’: an exploration of the neighborhood effect in Great Britain. Annals of the Association of American Geographers 90, 41–66. Pattie, C.J., Johnston, R.J., 2001. Talk as a political context: conversation and electoral change in British elections, 1992–1997. Electoral Studies 20, 17–40. Pattie, C.J., Johnston, R.J., 2005. Electoral participation and political context: the turnout/marginality paradox at the 2001 British General Election. Environment and Planning A 37, 1191–1206. Pattie, C.J., Johnston, R.J., 2008. It’s good to talk: talk, disagreement and tolerance. British Journal of Political Science 38, 677–698. Pattie, C.J., Johnston, R.J., 2009. Still talking, but is anyone listening? The changing face of constituency campaigning in Britain, 1997–2005. Party Politics 15, 411– 434. Pattie, C.J., Johnston, R.J., 2010. Constituency campaigning and local contests at the 2010 UK General Election. British Politics 5, 481–505. Pattie, C.J., Johnston, R.J., 2011. How big is the Big Society? Parliamentary Affairs 64, 403–424. Pattie, C.J., Johnston, R.J., Fieldhouse, E.A., 1995. Winning the local vote: the effectiveness of constituency campaign spending in Great Britain. American Political Science Review 89, 969–983. Pattie, C.J., Seyd, P., Whiteley, P., 2004. Citizenship in Britain: Values, Participation and Democracy. Cambridge University Press, Cambridge. Putnam, R., 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton University Press, Princeton. Putnam, R., 2000. Bowling Alone: The Collapse and Revival of American Community. Simon and Schuster, New York. Riker, W.H., Ordeshook, P.C., 1968. An Introduction to Positive Political Theory. Prentice-Hall, Englewood Cliffs, NJ. Runciman, W.G., 1966. Relative Deprivation and Social Justice. Routledge & Kegan Paul, London. Stoker, G., 2006. Why Politics Matters: Making Democracy Work. Palgrave MacMillan, Basingstoke. Swaddle, K., Heath, A., 1989. Official and reported turnout in the British General Election of 1987. British Journal of Political Science 19, 537–551. Tedin, K.L., 1980. Assessing peer and parent influence on adolescent political attitudes. American Journal of Political Science 24, 136–154. Verba, S., Schlozman, K.L., Brady, H.E., 1995. Voice and Equality: Civic Voluntarism in American Politics. Harvard University Press, Cambridge, Mass.. Zuckerman, A.S., 2005. Returning to the social logic of politics. In: Zuckerman, A.S. (Ed.), The Social Logic of Politics: Personal Networks as Contexts for Political Behavior. Temple University Press, Philadelphia. Zuckerman, A.S., Dasovic´, J., Fitzgerald, J., 2007. Partisan Families: The Social Logic of Bounded Partisanship in Germany and Britain. Cambridge University Press, Cambridge.