¹ Downs, Anthony, An Economic Theory of Democracy (New York: Harper & Row, 1957), chaps. 3 and 14Google Scholar; and Olson, Mancur Jr, The Logic of Collective Action (Cambridge, Mass.: Harvard University Press, 1965).Google Scholar

² Downs and Olson come to the same conclusion about the rationality of political action, though from somewhat different perspectives. Downs focuses on the rationality of abstention based on the very low likelihood that a voter could cast the deciding vote, Olson on the collective nature of the good sought. See Downs, , Economic Theory of Democracy, chap. 14Google Scholar, and Olson, , Logic of Collective Action, pp. 1–16Google Scholar. A lucid comparison of Downs and Olson is Benn, S. I., ‘Rationality and Political Behaviour’, in Benn, S. I. and Mortimore, G. W., Rationality and the Social Sciences: Contributions to the Philosophy and Methodology of the Social Science (London: Routledge and Kegan Paul, 1976), pp. 246–67Google Scholar. As Benn notes, Downs and Olson differ in the logic by which they come to the conclusion that political activity is not rational; however, the substantive outcome is the same. Another particularly clear explication of this perspective and analysis of the issues it raises is contained in Barry, Brian, Sociologists, Economists, and Democracy (Chicago: University of Chicago Press, 1970), chap 2.Google Scholar

³ See for example, Downs, , Economic Theory of Democracy, pp. 271–2.Google Scholar

⁴ On the nature of the interests pursued through collective action in organizations, see Schlozman, Kay Lehman and Tiemey, John T., Organized Interests and American Democracy (New York: Harper & Row, 1986), pp. 23–37, 123–32.Google Scholar

⁵ Aldrich, John H., ‘Rational Choice and Turnout’, American Journal of Political Science, 37 (1993), 246–69, at p. 247.CrossRefGoogle Scholar

⁶ These explanations focus upon, among other things, the size of the group involved, individuals' estimates of the impact of their own participation, and with respect to voting, the closeness of the election. See, for example, Olson, , Logic of Collective ActionGoogle Scholar; Hardin, Russell, Collective Action (Baltimore, Md: Johns Hopkins University Press, 1982), especially chaps 2–3Google Scholar; Fiorina, Morris and Ferejohn, John A., ‘The Paradox of Not Voting: A Decision Theoretic Analysis’, American Political Science Review, 68 (1974), 525–35Google Scholar; and Ferejohn, John A. and Fiorina, Morris, ‘Closeness Counts in Horseshoes and Dancing’, American Political Science Review, 69(1975), 920–5.CrossRefGoogle Scholar

⁷ For numerous examples, see Mansbridge, Jane J., ‘The Rise and Fall of Self-interest in Explanation of Political Life’, in Mansbridge, Jane J., ed., Beyond Self-Interest (Chicago: University of Chicago Press, 1990), pp. 11–13Google Scholar. Not all variants of rational choice theory require this simplifying assumption of selfishness. The most general and abstract version of rational choice focuses on a very limited ‘syntactical’ view of rationality: whether individual choices satisfy certain restrictive axioms that ensure that their choices are transitive. Benn and Mortimore (‘Technical Models of Rational Choice’, in Benn, and Mortimore, , eds, Rationality and the Social Sciences, p. 158Google Scholar) describe this as a ‘technical’ model of rational choice, and they contrast it with the ‘ordinary’ notion of rational action which requires ‘a good and sufficient reason for… choice’. In this technical theory, if the transitivity of preferences assumption is met, preferences can be measured through utility functions. The approach at this level is powerful for its logic, but has little relevance for telling us much about why people act one way or another. This approach ignores altogether the substance of why people do something: their reasons are of no concern to this perspective. People can construct preference orderings which, like grammatically correct nonsense sentences, satisfy the grammar of choice but which lack meaning. This approach has been useful for thinking about the possibilities (and impossibilities) of aggregating preferences. It does not tell us much about actual choices – it just requires that people be consistent in whatever tomfoolery they undertake. Henry E. Brady and Stephen Ansolabehere test this syntactical theory of rational choice using data on candidate preferences. See The Nature of Utility Functions in Mass Publics', American Political Science Review, 83 (1989), 143–63.CrossRefGoogle Scholar

⁸ Moreover, even when they are present, material benefits are more effective for inducing rational individuals to join a collective effort than to work on its behalf once they are members. In addition, tangible benefits are more useful for explaining the persistence of an ongoing organization than for explaining its inception. See, for example, Wilson, James Q., Political Organizations (New York: Basic Books, 1973), chap. 3.Google Scholar

⁹ Riker, William H. and Ordeshook, Peter C. emphasize the satisfaction of performing a citizen duty in their classic attempt to solve the paradox of voting, ‘A Theory of the Calculus of Voting’, American Political Science Review, 62 (1968), 25–42CrossRefGoogle Scholar. Subsequent analyses have added a variety of other psychic gratifications – for example, the desire to support a particular candidate or party, group identification and loyalties, social benefits and the desire to avoid the social costs of not voting. The literature is vast. See, for example, Fiorina, Morris, ‘The Voting Decision: Instrumental and Expressive Aspects’, Journal of Politics, 38 (1976), 390–415CrossRefGoogle Scholar; Uhlaner, Carole J., ‘Rational Turnout: The Neglected Role of Groups’, American Journal of Political Science, 33 (1989), 390–422CrossRefGoogle Scholar; Pomper, Gerald M. and Sernekos, Loretta, ‘The “Bake Sale” Theory of Voting Participation’ (paper presented at the Annual Meeting of the American Political Science Association, 09 1989)Google Scholar; Knack, Stephen, ‘Civic Norms, Social Sanctions, and Voter Turnout’, Rationality and Society, 4 (1992), 133–56CrossRefGoogle Scholar; and Aldrich, , ‘Rational Choice and Turnout’.Google Scholar

¹⁰ There has been extensive empirical work on the benefits provided to members of political organizations. Walker, Jack L. reviews the relevant literature and presents data in Mobilizing Interest Groups in America (Ann Arbor: University of Michigan Press, 1991), chaps 3 and 5CrossRefGoogle Scholar. See also, Wilson, , Political OrganizationsGoogle Scholar; Godwin, Richard K. and Mitchell, Robert Cameron, ‘Rational Models, Collective Goods, and Nonelectoral Political Behavior’, Western Political Quarterly, 35 (1982), 161–81CrossRefGoogle Scholar; Knoke, David and Wright-Isak, Christine, ‘Individual Motives and Organizational Incentive Systems’, Research in the Sociology of Organizations, 1 (1982), 209–54Google Scholar; Hansen, John Mark, ‘The Political Economy of Group Membership’, American Political Science Review, 79 (1985), 79–96CrossRefGoogle Scholar; Knoke, David, ‘Organizational Incentives’, American Sociological Review, 53(1988), 311–29CrossRefGoogle Scholar; and Knoke, David, Organizing for Collective Action: The Political Economics of Organizations (Hawthorne, NY: Aldine-De Gruyter, 1990).Google Scholar

¹¹ The historical emphasis upon the selective, material benefits offered by the patronage-oriented urban party machine has yielded to an understanding of party activism rooted in a more diverse set of rewards. See, for example, Wilson, James Q., The Amateur Democrat (Chicago: University of Chicago Press, 1960)Google Scholar; Conway, M. Margaret and Feigert, Frank B., ‘Motivation, Incentive Systems and the Political Party Organization’, American Political Science Review, 62 (1968), 1159–73CrossRefGoogle Scholar; and Eldersveld, Samuel J., Political Parties in American Society (New York: Basic Books, 1982), chap. 9Google Scholar. Results using survey data about the intention to take part in protests and, with more limited samples, actual experiences with rebellious collective action also demonstrate the importance of intangible rewards in motivating collective action. See Opp, Karl Dieter, ‘Soft Incentives and Collective Action’, British Journal of Political Science, 16 (1986), 87–112CrossRefGoogle Scholar; and Muller, Edward N. and Opp, Karl-Dieter, ‘Rational Choice and Rebellious Collective Action’, American Political Science Review, 80 (1986), 471–87CrossRefGoogle Scholar. Uhlaner, Carole Jean develops a formal model to explain political participation in general in ‘Political Participation, Rational Actors, and Rationality: A New Approach’, Political Psychology, 7 (1986), 551–73CrossRefGoogle Scholar; and ‘“Relational Goods” and Participation: Incorporating Sociability into a Theory of Rational Action’, Public Choice, 62 (1989), 253–85.CrossRefGoogle Scholar

¹² In ‘A Theory of the Calculus Of Voting’, Riker and Ordeshook argue that fulfilling a sense of civic duty provides such a direct benefit. This position is elaborated in Brennan, Geoffrey and Buchanan, James, ‘Voter Choice: Evaluating Political Alternatives’, American Behavioral Scientist, 28 (1984), 185–201CrossRefGoogle Scholar. For a recent elaboration that explains much of voting behaviour in terms of the direct consummatory benefits of political action, see Brennan, Geoffrey and Lomasky, Loren, Democracy and Decision: The Pure Theory of Electoral Preference (Cambridge: Cambridge University Press, 1993).CrossRefGoogle Scholar

¹³ Albert O. Hirschman puts it with characteristic elegance: ‘Once this essential characteristic of participation in collective action for the public good is understood, the severe limitations of the “economic” view about such participation, and about the obstacles to it, come immediately into view. The implication of the confusion between the striving and attaining is that the neat distinction between costs and benefits of action in the public interest vanishes, since striving, which should be entered on the cost side, turns out to be part of the benefit.’ (See Hirschman, Albert O., Shifting Involvements (Princeton, NJ: Princeton University Press, 1982), pp. 85–6Google Scholar.) Brennan and Lomasky (Democracy and Decision, p. 97Google Scholar) point out that the call to sacrifice is common in politics, the reward of political action deriving, at least in part, from participants having borne heavy burdens.

¹⁴ Brennan, and Lomasky, (Democracy and Decision, p. 2)Google Scholar make a similar point. They take a ‘two-hats’ position, that ‘actors have two personae: one for markets and a different one for the ballot box (and analogous collective activities)’.

¹⁵ This approach is taken by Riker, and Ordeshook, in ‘Theory of the Calculus of Voting’Google Scholar, who add the gratification of fulfilling a civic duty to the cost-benefit analysis of voting. This preserves the theory, but makes it so encompassing that it borders on tautology. However, as Barry, Brian (Sociologists, Economists, and Democracy, p. 33)Google Scholar, who first commented on the dilemma, points out, the theory is ‘still a quite potent tautology’. For discussion of the problems with this expansion, see also Benn, , ‘Rationality and Political Behaviour’, pp. 256–60Google Scholar. In Political Participation (Canberra: Australian National University, 1978), p. 77Google Scholar, Benn presents a justification for political action which draws on a ‘notion of non-instrumental rationality’ which includes consideration of a broad range of principles.

¹⁶ Amartya Sen suggests that it is more appropriate to consider this to be behaviour based on normative commitments. See his classic statement in ‘Rational Fools’, reprinted in Mansbridge, , ed., Beyond Self-interest, pp. 25–43Google Scholar. For arguments from many disciplinary perspectives about the significance of altruism and co-operation in human behaviour and extensive bibliographical references, see the essays in that volume.

¹⁷ The Citizen Participation Study is a large-scale, two-stage survey of the voluntary activity of the American public. The first stage consisted of over 15,000 telephone interviews of a random sample of American adults conducted during the last six months of 1989. These 20-minute screener interviews provided a profile of political and non-political activity as well as basic demographic information. In the spring of 1990, we conducted much longer, in-person interviews with a stratified random sample of 2,517 of the original 15,000 respondents chosen so as to produce a disproportionate number of those active in politics (as well as African-Americans and Latinos). The data in this article are from the 2,517 respondents in the follow-up survey. A more detailed description of the sample, the sample weights that allow the sample to be analysed as a random sample and a listing of the relevant measures, can be found in Verba, Sidney, Schlozman, Kay Lehman, Brady, Henry and Nie, Norman, ‘Citizen Activity: Who Participates? What Do They Say?’ American Political Science Review, 87(1993), 303–18, at p. 314.CrossRefGoogle Scholar

¹⁸ In general we asked about activity over the past twelve months or, in the case of campaign work or contributions, the period since the beginning of the last presidential cycle. In order to generate sufficient numbers of those who engage in rare acts, the time frame for attending a protest or serving on a local governing board was the past two years. Respondents who had engaged in an activity more than once over the designated time period were asked, depending upon the act, about either the most recent or the one to which the most time or money was devoted. In the case of ongoing activity – for example, membership on a local board or educational, charitable or social activity in a church – it is impossible to specify a particular event. In these cases, we asked instead what keeps the respondent active.

¹⁹ Our approach to measuring organizational involvement bears some elaboration. In an electronic era, affiliation with an organization may be a matter of writing a cheque in response to a telephone or mass-mail solicitation instead of membership in the traditional sense in which one joins the League of Women Voters or the Rotary. Therefore, we asked about both membership and financial contributions and consider either to be sufficient to qualify as organizational involvement. Respondents were asked about twenty different categories of organizations. For each organizational category for which a respondent indicated involvement, the respondent was asked the number of such organizations and a series of follow-up questions about the organization (or, if more than one in the category, the organization in which the respondent is most involved) and his or her activity in it. Among these follow-up questions was an item about whether the organization ever takes stands on public issues either nationally or locally. We consider affiliation with an organization that takes stands to be a form of political participation. Then we asked respondents a long series of questions about the single organization to which they give the most time or money (or, if different, the organization they consider most important). The organization data on which we report in this article derive from this section of the interview. For an extended discussion of these measures, and the rationale for them, see Schlozman, Kay Lehman, ‘Voluntary Associations in Politics: Who Gets Involved?’ in Crotty, William, Schwartz, Mildred A. and Green, John C., eds, Representing Interests and Interest Group Representation (Lanham, Md: University Press of America, 1994), pp. 67–83.Google Scholar

²⁰ This point is made by, among others, Aldrich, , ‘Rational Choice and Turnout’, p. 261.Google Scholar

²¹ The strongest statement of this point of view is Nisbett, Richard E. and Wilson, Timothy DeCamp, ‘Telling More than We Can Know: Verbal Reports on Mental Processes’, Psychological Review, 84 (1977), 231–59.CrossRefGoogle Scholar

²² Nisbett and Wilson's original claim has been modified somewhat in the substantial literature inspired by their influential article. See, for example, Smith, Eliot R. and Miller, Frederick D., ‘Limits on Perception of Cognitive Processes: A Reply to Nisbett and Wilson’, Psychological Review, 85 (1978), 355–62CrossRefGoogle Scholar; White, Peter, ‘Limitations on Verbal Reports of Internal Events: A Refutation of Nisbett and Wilson and of Bern’, Psychological Review, 87 (1980), 105–12CrossRefGoogle Scholar; Ericsson, K. Anders and Simon, Herbert A., ‘Verbal Reports as Data’, Psychological Review, 87 (1980), 215–51CrossRefGoogle Scholar; Sabini, John and Silver, Maury, ‘Introspection and Causal Accounts’, Journal of Personality and Social Psychology, 40 (1981), 171–9CrossRefGoogle Scholar; Wright, Peter and Rip, Peter D., ‘Retrospective Reports on the Causes of Decisions’, Journal of Personality and Social Psychology, 40 (1981), 601–14CrossRefGoogle Scholar; Kraut, Robert E. and Lewis, Steven H., ‘Person Perception and Self-Awareness: Knowledge of Influences on One's Own Judgments’, Journal of Personality and Social Psychology, 42 (1982), 448–60CrossRefGoogle Scholar; and Gavanski, Igor and Hoffman, Curt, ‘Awareness of Influences on One's Own Judgments: The Roles of Covariation Detection and Attention to the Judgment Process’, Journal of Personality and Social Psychology, 52 (1987), 453–63.CrossRefGoogle Scholar

²³ See Davidson, Donald, Essays on Actions and Events (New York: Oxford University Press, 1980)Google Scholar, especially, ‘Actions, Reasons, and Causes’, and Elster, Jon, Sour Grapes (Cambridge and New York: Cambridge University Press, 1983).CrossRefGoogle Scholar

²⁴ In addition, in order to avoid giving respondents the occasion to invent reasons for relatively inconsequential activity – very small political donations or nominal church memberships or organizational affiliations – we imposed a minimum level of commitment below which the questions about reasons were not asked. Those whose campaign contributions did not total more than $50, whose church activity did not include being a board member or an officer, serving on committees or helping with special projects, or giving at least two hours each week, or whose activity in their most important organization entailed giving neither time nor money were not asked about their reasons for getting involved.

²⁵ It has been suggested to us that the way to ask about gratifications is to use open-ended rather than closed-ended questions. We did not do so for several reasons. Pre-testing indicated that respondents tend to mouth clichés when an open-ended question about rewards of participation is posed. In addition, they are more likely to acknowledge the importance of selfish motives when self-interestedness is legitimated in closed-ended items. Moreover, we followed up the battery of closed-ended items about the relative importance of various gratifications with an open-ended, ‘Anything else?’ Most of the replies to this question simply restated something already covered in the closed-ended questions. As will be seen, we did use open-ended items in asking about the subject matter of activity.

²⁶ Wilson, , Political OrganizationsGoogle Scholar, especially chaps 2–3. Other typologies of the gratifications attendant to organizational support have many elements in common with Wilson's. See for example, Salisbury, Robert ‘An Exchange Theory of Interest Groups’, Midwest Journal of Political Science, 13 (1969), 1–32CrossRefGoogle Scholar; or Knoke, , Organizing for Collective Action.Google Scholar

²⁷ In constructing items about the reasons for activity, we were not always able to frame them in such a way that they would hew to the lines of Wilson's analytic categories. Some of the formulations that recurred as respondents talked about their activity in pre-tests were not readily identifiable in terms of Wilson's typology. For example, across almost all the acts relatively high proportions of activists – ranging from 40 per cent in the case of contributing or contacting to 70 per cent in the case of protesting – reported that ‘the chance to work with (contribute along with, add my vote with, etc.) people who share my ideals’ was very important. In Wilson's terms, it is not clear whether this should be considered a ‘solidary’ gratification deriving from the association with like-minded people or a ‘purposive’ one dependent upon the ideals to which the activist is committed. Because the way that activists actually talk about politics does not always correspond to Wilson's conceptually distinct categories, we posit our categories, not as a theoretical breakthrough, but rather as a version of Wilson's categories altered for utility in a survey. Because our survey items could not always preserve Wilson's analytic distinctions, we do not use his labels – even though we derive our understanding from his categories.

²⁸ These are what Wilson calls ‘purposive’ gratifications, ‘intangible rewards that derive from the sense of satisfaction of having contributed to the attainment of a worthwhile cause’ (Political Organizations, p. 34).

²⁹ David Knoke follows a similar approach when he specifies a separate dimension for ‘lobbying incentives’ in ‘Organizational Incentives’, as does Hansen, who discusses ‘political collective benefits’ in ‘The Political Economy of Group Membership’.

³⁰ The difficulty is using a common set of reasons for all participatory acts grows out of the profound differences among the modes of activity. When pre-tests elicited reactions indicating that respondents considered an item simply too far-fetched (‘I went to the protest for the chance to meet important and influential people? Give me a break.’), we omitted it from the list. In so doing, we avoided alienating respondents but risked missing a possible motive for activity. In addition, certain items are significant for a single context – but relevant only there. For example, a higher proportion, 76 per cent, of the church activists cited the desire to affirm their religious faith as very important than any other item on the list for religious activity. Not to have included this item – one that is, obviously, germane only to one kind of activity – would have been to overlook one of the fundamental wellsprings of religious activity. For a complete listing of the items used for each of the acts and the proportions saying that a particular reason was ‘very important’ for a particular act, see Appendix Table A.

³¹ It is certainly possible to question the judgements made in assigning items to categories. Although we settled on a strategy of including under a particular rubric only items that unambiguously belong there, we experimented with more expansive definitions of several of the categories. The effect of defining the categories more broadly by including items germane to relatively few acts or ambiguously related to a particular theoretical dimension was only to raise the proportion who cited at least one of the components as very important but not to change in any way the pattern of responses reported below. The reader is invited to refer to Appendix Table A for the full array of questions asked.

³² We factor analysed the gratifications from each political act separately. We consistently found three dimensions: civic gratifications, collective policy outcomes, and a dimension that combined selective material and social gratifications. In the absence of clear guidance from the factor analysis, we separated the latter dimension into two – material and social – based on our reading of what the questions were about. For a description of the analysis and its results, contact the authors.

³³ Because activities undertaken alone can be exciting, this is not technically a ‘social’ gratification. However, this is a gratification cited relatively frequently across acts, and it seems to fit in with the other reasons we call ‘social’, all of which involve gratifications derived from an association with others.

³⁴ It might be argued that, since the lists of possible reasons varied in length across the acts, the average number of items deemed ‘very important’ is not the appropriate measure. After all, the respondent was given more possible reasons to consider important for church activity than for voting. This line of reasoning suggests that a more appropriate measure would be the average proportion considering each reason important for the various acts. By this measure, church activity still stands out as the most rewarding. Among political acts, protest scores high with respect to the proportion. of activists rating each of the possible reasons as very important. Before accepting this alternate measure, however, we should note that, as mentioned earlier, the lists varied in length because certain gratifications are simply not plausible for certain acts. Thus, in each case the length of the list says something about the approximate number of potential benefits.

³⁵ Of those whose electoral contributions were directed to an organization rather than a particular candidate, 39 per cent donated to a party organization, 52 per cent to a political action committee (PAC) and the remainder to some other kind of organization. The PAC donations were directed as follows: work-related, 38%; political issue, 40%; general liberal or conservative, 12%; PAC associated with a politician, 7%, other, 4% (total, 101%). The distribution among types of work-related electoral organizations is as follows: union, 32%; company or firm, 27%; trade association, 4%; professional association, 29%; other, 8%; (total, 100%).

³⁶ With respect to the kinds of selective, material benefits in which organizations are said to specialize, 15 per cent of those whose most important organization takes stands in politics and 29 per cent whose most important organization does not take political stands mentioned the recreational opportunities provided by the organization as very important in keeping them active. The analogous figures for the proportions mentioning the direct services provided to members are 36 per cent and 33 per cent.

Considering the narrower group of organizations that are more exclusively political, we see that the proportion mentioning any selective material benefit as a reason for involvement is much lower, only 25 per cent. Similarly, among those whose most important organization fell into the narrower category of explicitly political organizations, only 4 per cent cited recreational opportunities and 15 per cent cited direct services as very important in keeping them involved.

³⁷ The proportion of church activists reporting at least one material benefit as very important is relatively high, 40 per cent. Church activists rarely cited career-oriented motivations, however. Rather, they were much more likely to report that the recreational opportunities, the possibility of getting help on a personal or family problem and, especially, the services – for example, child care or marriage counselling – provided by the church as being very important in keeping them active in their churches. If we were to include the desire for a religious education for their children as a selective, material benefit, then the proportion of church activists mentioning at least one such benefit would soar to 73 per cent.

³⁸ The item asking the kind of electoral organization to which the contribution was made was separated from the battery of items about the reasons for the contribution. Hence, we believe that the difference in the patterns of reported gratifications is not simply a function of question ordering.

³⁹ The profile for the narrow category of explicitly political organizations is, of course, quite different and more similar to that for most political activities.

⁴⁰ The respondent was shown a card containing the following categories: only myself or my family; only other people, but not myself or my family; myself or my family, as well as others like us; all people in the community; all people in the nation or the world. In order to distinguish particularized from policy concerns, respondents who said that the problem affected themselves and others as well were asked whether their activity was aimed at dealing with the problem for themselves and their family only or whether they were trying to deal with it for other people as well. In this way, we were able to delineate more accurately the scope of concern of individuals who mentioned ‘Social Security’ or ‘high property taxes’ – problems that would affect many people in circumstances similar to theirs, but could be either particularistic or policy concerns. If more than one issue or problem was listed in response to the open-ended item, the closed-ended question was asked about the first one mentioned.

⁴¹ The 14 per cent of responses that did not contain recognizable policy issues were distributed as. follows. A small proportion – 4 per cent across all the political activities – could not be coded either because the respondent was confused and inarticulate or because the interviewer was sloppy in recording and editing. Another 10 per cent represented coherent and ‘codeable’ problems or concerns that did not constitute public policy issues. Examples include the following statements describing the issues or problems behind campaign work: ‘To see New York have its first black mayor.’ ‘My husband was running for office. He was the best-qualified candidate.’ ‘We needed another conservative.’

⁴² Let us clarify several aspects of the coding. First, a single political act may have been inspired by more than one issue concern. A person making contact who expressed concern about ‘public housing, teenage pregnancy, and the child care bill’ would have been coded as mentioning three separate issues. In addition, respondents identifying a problem or issue within a particular category might have quite different opinions. For example, the social issues category includes many social conservatives as well as respondents who are pro-choice rather than pro-life or who are concerned about making contraceptives more easily available to teenagers. Finally, these categories are not exhaustive. Issue concerns ranging from gun control to local economic development, for which we have codes, have been omitted from this list.

⁴³ Across all activities the overwhelming share of those who mentioned social issues in connection with their activity were referring to abortion. For example, of those who a got in touch with a public official on an issue of national import, 81 per cent contacted about abortion. Similar – and, in some cases, even higher – proportions obtain for other activities.