Beyond the Self: Social Identity, Altruism, and Political Participation JameS H. FOWler University of California, San Diego Cindy D. Kam University of California, Davis Scholars have recently extended the traditional calculus of participation model by adding a term for benefits to others. We advance this work by distinguishing theoretically a concern for others in general (altruism) from a concern for others in certain groups (social identification). We posit that both concerns generate increased benefits from participation. To test these theories, we use allocations in dictator games towards an unidentified anonymous recipient and two recipients identified only as a registered Democrat or a registered Republican. These allocations permit a distinction between altruism and social identification. The results show that both altruism and social identification significantly increase political participation. The results also demonstrate the usefulness of incorporating benefits that stem from sources beyond material self-interest into rational choice models of participation. "Avarice, or the desire for gain, is a universal passion which operates at all times, in all places, and upon all persons" (Hume, Of the Rise and Progress of the Arts and Sciences, [1742] 1991, 113) Alarge body of work, ancient and modern, posits that self-interest is the primal force for political attitudes and behaviors. Although this parsimonious assumption explains many observed political phenomena, it fails to illuminate some of the most important features of political life (Citrin and Green 1990; Mansbridge 1990; Sears and Funk 1991). For example, rational choice scholars have typically approached the problem of political participation by using models based on pure self-interest (Aldrich 1993; Downs [1957] 1985; Feddersen and Pesendorfer 1996; Ledyard 1982; Palfrey and Rosenthal 1985). These models encounter a well-known difficulty: although an individual may derive personal benefits from a certain political outcome, the probability that a single act of participation will significantly affect the outcome is very small in large populations. This gives individuals an incentive to avoid the costs of participation and free ride on the efforts of others, producing the well-known paradox of participation. If self-interest does not motivate political participation, then what does? One possibility is that indi- The Journal of Politics, Vol. 69, No. 3, August 2007, pp. 813-827 © 2007 Southern Political Science Association viduals consider benefits to others, beyond the self, when deciding whether or not to participate. Even Downs, so often portrayed as the archetypal champion of self-interest as a motivating factor of political choice, states that a concern for the welfare of others might influence political attitudes and behaviors: "In reality, men are not always selfish, even in politics. They frequently do what appears to be individually irrational because they believe it is socially rational— i.e., it benefits others even though it harms them socially" ([1957] 1985, 27). Extending the foundation for political choice beyond the self is not an easy task. It forces analysts to confront an important question: When individuals decide whether and how to act, to whom do they refer? On whose benefit will they act? In this article, we distinguish theoretically two different kinds of other-regarding considerations that influence political participation. Some people are motivated by social identification, which creates a desire to improve the welfare of certain groups in society, possibly at the expense of other groups. These individuals will likely participate when they believe that their actions will give them an opportunity to help their preferred group(s). Other people are motivated by altruism, a willingness to pay a personal cost to provide benefits to others in general, regardless of the identity of the beneficiaries. These individuals will ISSN 0022-3816 813 814 likely participate when they believe that their actions will give them an opportunity to make everyone better off. Our core expectation is that altruists and social identifiers will participate more than egoists—that is, individuals who are primarily self-interested. We test the social identifier and altruism theories of participation using a unique experimental design. Subjects are asked a number of standard questions regarding their socioeconomic status, political attitudes, and participation behavior. They then play three "dictator" games (Forsythe et al. 1994), in which they divide a set of lottery tickets between themselves and an anonymous individual. The recipient is completely anonymous in all three games. However, in two of the games, subjects are informed that the recipient is a registered Democrat or a registered Republican. We use these dictator games to uncover the degree to which each subject is generally concerned about the well-being of others, as evidenced by allocations to the unidentified anonymous recipient, and the degree to which each subject socially identifies with the political parties, as shown by allocations to the Democrat and Republican. These experiments yield several novel results for behavior in the dictator game and its relationship to political participation. First, we show that behaviors in these dictator games reveal a key characteristic of social identification: a preference for the ingroup versus the outgroup. Democrats and Republicans both give more to the recipient from their own party than the opposing party; independents give more to the anonymous recipient than the partisan recipients, while partisans do just the opposite. Second, behaviors in these dictator games reveal that strength of social identity magnifies preferences for the ingroup. Subjects who identify themselves as strong Democrats and strong Republicans tend to give much less to the recipient from the opposing party than partisans identifying with weaker affiliations. Third, we uncover a bias against Republicans. The Republican recipient tends to receive less than the Democrat or the unidentified anonymous recipient, even when the donor is a Republican. Finally, both altruism and social identity increase political participation. People who share with an anonymous individual in the dictator game participate in politics more than those who do not share. People who vary the amount they give depending on the partisan affiliation of the recipient also participate more than those who give (or withhold) the same amount to (from) everyone. These results suggest that other-regarding behavior plays an important role in the decision to participate. Our work has broad implications for existing scholarship in several fields. Since it is the first exami- JAMES H. FOWLER AND CINDY D. KAM nation of the impact of partisanship on dictator game allocations, this work should be of interest to behavioral and experimental economists. It should also be of interest to psychologists and sociologists, since our uniquely designed dictator game provides a novel means of tapping social identity. Most existing work on social identification does not force individuals to sacrifice their own material well-being in order to affirm support for their ingroups, but in the dictator game, social identifiers must deliberately deprive themselves of personal rewards so that they can affirm the position of someone else in their group. In our design, affirming social identification has a cost. We demonstrate that subjects are in fact willing to bear this cost, and we demonstrate the political consequences of this behavior. Finally, our work should be of interest to political scientists, since we not only introduce an innovation in the measurement of dispositions towards groups and others in general, but we also identify the political implications of these dispositions by using them to predict political participation. Our work therefore allows us to address the literature on rational choice by demonstrating that the core motivational elements of rational choice theory need not rest entirely or solely on self-interest, that other-regarding behavior can and should be taken into account, and that rationality in no obvious or necessary way requires material self-interest to be privileged as the primary motivator in models of political behavior. Social Identity, Altruism, and Participation Traditional rational models of participation based on self-interest posit that individuals receive a benefit B from some political activity if their preferred outcome occurs. However, the participatory acts that yield this outcome are individually costly (e.g., Aldrich 1993; Downs [1957] 1985; Feddersen and Pesendorfer 1996; Ledyard 1982; Palfrey and Rosenthal 1985). The sticking point for these models is that a single act of participation usually has only a very small probability P of affecting some political outcome. For example, if the participatory act is voting, then the outcome can only be changed when there is an exact tie, or when the vote can create a tie. If the participatory act is a contribution of money or time to a candidate or political organization, it may be just one of thousands or even millions of other contributions. Thus, the expected benefit of participation PB is typically less than the cost C, even when populations are not too large and even when the cost of participation is very low. BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION 815 Riker and Ordeshook's (1968) D term seems to offer one solution to the paradox of voting. The D term suggests that individuals who participate in politics derive a benefit associated with the act of voting, resulting from satisfying a sense of citizen obligation, affirming their allegiance to the political system or reinforcing their own sense of efficacy. This benefit associated with completing the act of voting is orthogonal to the benefits derived from the policy outcome of the political action. Thus, political participation is an expressive act in which the desired policy outcomes are essentially irrelevant in the participation calculus, given how small P and B are.1 The D term provides one answer to the paradox of voting, but it is not the only answer. We argue that citizens can consider political action to be instrumental not only for themselves but for others as well. Empirical research suggests this to be the case: activists frequently participate in politics in order to enact changes in public policy—that is, they act for instrumental reasons—no matter how "irrational" this motivation seems (Schlozman, Verba, and Brady 1995). Further, activists frequently note that the political stakes of participation affect individuals beyond themselves and their families (Schlozman, Verba, and Brady 1995). That is, they act instrumentally, not just for their own benefit, but for the benefit of others. As such, the policy outcomes of political actions should affect individual decision making. We explicitly address the possibility that an individual may care about the impact of policies as these policies apply beyond the self. We do so by incorporating social identity and altruism into the calculus of participation. Note that the benefits associated with altruism and social identification are distinct from those captured by the D term. The D term can be conceived of as system affirmation or fulfillment of a moral obligation to participate. Moreover, the D term is independent of political outcomes—people with a strong sense of social obligation will participate even if they think the 'Riker and Ordeshook's (1968) approach is decision-theoretic and based on the assumption that the D term is exogenous. However, two recent attempts to endogenize the D term in a game theoretic model show that "ethical" preferences can help to explain turnout even when voters are well informed and fully strategic (Coate and Conlin 2004; Feddersen and Sandroni 2006a, 2006b). These models suggest that voters act as social planners by trying to maximize social welfare, and they gain utility from "doing their part." However, both of these models assume that voters prefer the lowest turnout possible, and neither of these models considers the possibility that voters might care about the distributive implications of political outcomes that provide benefits to some groups at a cost to others. act of participating will have no influence on benefits derived from policy outcomes. In contrast, we argue that altruism and social identity will encourage political action in order to benefit others, generally or specifically; altruism and social identity affect B. According to social identity theory, individuals yearn to acquire and maintain a positive self-identity (Tajfel 1981). This sense of self is derived in large part from formal membership with or psychological attachment to social groupings. In contrast with a theory based purely on self-interest, social identity theory suggests that individuals gain utility from affiliating with social groups, from bestowing benefits upon the ingroup, and from withholding benefits from the outgroup. Social identity theory resonates with Converse's (1964) observation that the fundamental way in which many citizens understand politics is through groups. Social identity theory implies that individuals will make political choices by using specific groups rather than the self as a reference point. Social identity predicts policy preferences (Campbell etal. [1960] 1980; Kinder and Winter 2001; Price 1989), and under some conditions, social identity spurs collective action (for a review, see Huddy 2003). So far, however, the literature has not linked social identity with the policy-oriented benefits of participation in an attempt to address the paradox of participation.2 We argue that social identifiers may be spurred into political action when they believe that political outcomes will positively affect members of their group. When individuals perceive political outcomes as distributive—as opportunities to transfer resources from outgroups to their ingroup—social identifiers should be more likely to participate than individuals who are self-interested. Moreover, as people identify more strongly with their ingroup or more strongly against some outgroup, they should experience greater benefits from distributive politics and thus be more likely to participate. While social identity theory suggests that individuals partition the world into ingroups and outgroups, in a wide range of contexts, human beings 2One exception worth noting is Uhlaner's (1989) treatment of group members, group leaders, and candidates in her formal model of turnout. She argues that group leaders can manipulate the costs and benefits of voting, e.g., through ostracism of abstainers or social invitations directed at compliers. This approach differs from ours because group leaders manipulate benefits obtained from the act of voting—where the act of voting is still expressive and not instrumental, and the instrumental functions of voting remain untouched by the actions of group leaders. 8i6 JAMES H. FOWLER AND CINDY D. KAM have been observed to be motivated by the welfare of others in general (Fehr and Fischbacher 2003; Monroe 1996; Piliavin and Charng 1990). They engage in acts of altruism, or "behavior intended to benefit another, even when this risks possible sacrifice to the welfare of the actor" (Monroe 1996, 6). In contrast with social identifiers, altruists do not typically target individuals from certain groups for benefits. Monroe (1996) explains that individuals who are willing to engage in uncommon acts of altruism express a sense of univer-salism in viewing the human condition. Instead of viewing an individual (and the self, in particular) as tied to specific social groupings, altruists "share a view of the world in which all people are one" (198). Thus, while social identifiers are more likely to help members of their ingroup, altruists are unlikely to discriminate in whom they help.3 Scholars have recently incorporated altruism into the traditional calculus of participation model by assuming that each citizen also cares about the benefits that others secure from the preferred outcome (Edlin, Gelman, and Kaplan 2007; Fowler 2006; Jankowski 2002, 2007). Although a single participatory act may have little effect on a political outcome, the number of people who benefit may be quite large. Thus, those who exhibit a sufficient degree of concern for the welfare of others will be willing to engage in costly political participation. Moreover, as people become more concerned for the welfare of others, they should experience greater benefits when political outcomes portend improvements for the welfare of others generally. Thus, altruists will be more likely to participate than individuals who are self-interested.4 Political outcomes might be construed by individuals as improving the general welfare and/or as favoring particular social and political groups. Consequently, the decision to participate in politics may be motivated by both a desire to make things better for everyone (altruism) and a desire specifically to acquire as many benefits as possible for the ingroup (social identification). Thus the benefit from participation may be derived by some combination of self-interest, altruism, and social identity. 3Note that we do not address the evolutionary or social origins of altruism and other forms of nonself-interested behavior as has been done elsewhere (e.g., Samuelson 1993). Our primary interest here is in how variation in altruism and social identification predicts political participation. 4For a formalized sketch of how altruism and social identity might be inserted into the classic paradox of voting model, see the online appendix at http://journalofpolitics.org/articles.html. Finding Altruists and Social Identifiers among Dictators Our study contributes to existing empirical work by adopting an innovative measure of altruism and social identity. Previous attempts to examine the relationship between other-regarding behavior and participation have relied on questions in the National Election Study (NES) pilots. Knack (1992) creates an index of "social altruism" from questions about charity, volunteer work, and community involvement on the 1991 NES Pilot Study and finds a positive relationship between the index and voter turnout. However, the questions used in the index are very close to those used by scholars who argue that organizational involvement (not altruism) enhances political participation (Verba, Schlozman, and Brady 1995). Jankowski (2007) finds a relationship between voter turnout and "humanitarian" norms (i.e., agreement with the statement that "One of the problems of today's society is that people are often not kind enough to others"). These questions certainly reflect expectations about the altruism of others, but it is not clear how they relate to the respondent's own willingness to bear costs to provide benefits to others. Typical measures of social identification rely upon self-reports (Kinder and Winter 2001) or are based on group membership (Price 1989). The above studies rely on respondents' expressed preferences for helping others generally or for identifying with a group. In neither case do respondents actually experience a cost in order to give a benefit to someone else. In contrast, preferences for helping others are revealed in what experimental economists call the "dictator game" (Forsythe et al. 1994). In this game, the experimenter gives player 1 a certain amount of money and then asks the subject to divide that money between herself and player 2.5 If player 1 is motivated only by her own economic gain, she should keep all the money for herself and allocate nothing to player 2. However, this is not what players normally do. In a survey of dictator game results, Camerer (2003) shows that the mean allocation to player 2 ranges from 10% to 52%. Anonymity conditions tend to decrease the mean allocation, but even in the most anonymous treatments (Hoffman et al. 1994) about 40% of the allocations still exceed 0. 5Unlike the ultimatum game (cf. Hibbing and Alford 2004), the dictator game does not give player 2 an opportunity to accept or reject the offer. In the dictator game, player 2 simply pockets the money that player 1 allocates to her and the game is over. BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION 817 Interpretations of Excess Giving in the Dictator Game Excess giving in dictator games is a replicable empirical regularity. Scholars offer several explanations for this excess. We adopt the most prominent explanation: altruism—that individuals engage in "other-regarding" behavior. The altruism explanation suggests that dictators give to others because they want to improve the well-being of other individuals, even when doing so impinges on their own material interests. In his thorough overview of dictator games, Camerer notes that "there is some pure altruism" that explains excess giving (2003,56). For example, in their study of altruism and dictator games, Eckel and Grossman (1996) manipulate the target of the giving; they find that subjects are much more likely to give when the target is the Red Cross. Eckel and Grossman conclude that "altruism is a motivating factor in human behavior in general and in dictator games in particular" (1996, 182). A companion explanation for excess giving is a taste for fairness; this fairness hypothesis is often discussed interchangeably with altruism, but they are distinguishable from each other. The altruism explanation hinges upon the idea that individuals care about others' welfare. The fairness explanation in its simplest formulation is standards-oriented: an individual cares that the division of goods satisfies some standard of equity (typically, in the standard dictator game, one-half). Further elaborations of the fairness explanation tilt the balance even more towards the self: Fehr and Schmidt (1999) offer an extension of this line of reasoning in suggesting that individuals care about not just equity in outcomes across individuals but also about the absolute difference between an individual's allocations vis-a-vis that of other individuals. They specify an asymmetric utility function, where individuals receive the most utility when payoffs are equal, slight (and increasing) disutility from being advantaged when compared to others ("guilt," per Camerer 2003, 102), and sharper (and increasingly sharper) disutility from being disadvantaged compared to others ("envy," per Camerer 2003, 102). Yet, even after incorporating a sense of fairness, guilt, and envy into account, Fehr and Schmidt still note that "Altruism is consistent with voluntary giving in dictator and other public good games" (1999, 854).6,7 6Using a similar approach, the Bolton and Ockenfels (2000) Equity, Reciprocity, and Competition (ERC) Model specifies that individuals receive utility from their own level of material standing In an ingenious design that compares the altruism and fairness interpretations, Andreoni and Miller (2002) examine choices in a series of dictator games with different payoffs. In some treatments, player 2 is given $.20 or $.30 for every $.10 player 1 allocates. In other treatments, player 1 must allocate $.20 or $.30 for every $.10 player 2 receives. By varying the payoffs, Andreoni and Miller are able to distinguish between individuals who give in order to equalize payoffs (whom they call "Rawlsians") and those who give in order to maximize total payoffs to both players (whom they call "utilitarians"). The results show that about two-thirds of those who incorporate the recipient's utility in their decision can be described as "utilitarians." Thus, while a concern for fairness undoubtedly plays an important role, altruism, or consideration of others' welfare, appears to be the dominant motivation behind giving in the dictator game. Another explanation for excess giving rests on the notion of reciprocity. Hoffman et al. (1994) and Hoffman, McCabe, and Smith (1996a) argue that excess giving occurs in order to satisfy norms of reciprocity. Dictators give to others because future rewards are contingent upon the individual's "social reputation as a cooperative other-regarding person" (Smith 2000,84). Dictators thus give more than would be expected because they are concerned, in the short run, that appearing "greedy" will decrease the likelihood that they would be invited back for more experiments, or they are concerned in the long run of other negative consequences for themselves. To dispute this reciprocity argument, Johannesson and Persson (2000) manipulate the target recipient in a dictator game, specifying that the recipient is one of the other subjects recruited for the study or a randomly selected and from their relative standing compared with others. The core argument, thus, is that individuals do not really care about making others better off; instead, they take their own standing and the relative standing of others into account. However, the ERC still includes the notion of a "social reference point," a standard against which decisions are measured. In dictator games, this social reference point is an equal division of the payouts. The notion of equity, or fairness, has a pivotal place in this formulation. A subtle distinction between the Fehr and Schmidt (1999) and Bolton and Ockenfels (2000) models is that in the former, individuals care about the absolute level of difference between themselves and others, whereas in the latter, individuals care about their relative shares in the allocations rather than absolute differences in these shares (Camerer 2003, 104). 7Fehr and Schmidt (1999) note that it is harder to account for behaviors in other games using altruism. This is not a central concern of ours in this paper, as we do not believe that other games provide as appropriate a means of tapping altruism. 8i8 JAMES H. FOWLER AND CINDY D. KAM individual from the general population. They argue that "If donations in dictator games are motivated solely by reciprocity, donations should therefore drop to zero with this experimental treatment" (2000,138). Johannesson and Persson are unable to reject the null hypothesis of no difference between the two groups, which suggests that excess giving in the dictator game cannot be ascribed to reciprocity on its own. A final explanation for excess giving is that subjects do not understand the game and are just making random allocations. Andreoni and Miller (2002) address this concern by examining within-subject patterns of choices in their series of dictator games with different payoffs. They find that 98% of the subjects make choices that are consistent with the general axiom of revealed preferences across eight treatments, suggesting that most of them understand the game and are not choosing randomly. These results from the literature on giving in the dictator game suggest that while there are several factors that might explain giving, dictator game allocations may be a good proxy for an individual's concern for the well-being of others. The well-being of others is probably more important to a person who chooses to allocate 20% than one who allocates 0%. In fact, the utility function used in Andreoni and Miller (2002) to explain behavior in the dictator game yields a monotonie relationship between the equilibrium allocation in the dictator game and the weight a player places on the other player's utility. In other words, the more a player cares about the well-being of others, the more she will allocate to the other player in the dictator game. Tapping Altruism and Social Identity in a Unique Dictator Game Behavior in dictator games can reveal other-regardingness at a general level (that is, altruism). We can also use dictator games to investigate whether or not individuals exhibit politically relevant group-based preferences. Past experiments have varied the characteristics of the anonymous recipient with some interesting results. People are more willing to give to charities than an anonymous individual (Eckel and Grossman 1996), to women (Saad and Gill 2001), and to people who have been introduced to them (Bohnet and Frey 1999). In their experiment, Bohnet and Frey manipulate the amount of information provided about the target to the dictator. They find that more information "transforms anonymous, faceless entities into visible, specified human beings, i.e., identifiable victims" (1999, 339). They argue that this pattern of increased giving suggests that, "the more we know, the more we care" (citation from Schelling 1968). Camerer notes that the "identification effect is target specific and is not the result of general sympathy toward others" (2003, 76). Our design enables us to capture this distinction between concern for others, generally, and a concern for specific groups. We are interested in whether or not people give more to members of one political group than another or whether they give the same amount to an anonymous individual versus individuals affiliated with groups. By varying information about the political group to which the target recipient belongs, we can uncover the extent to which social identity might drive allocation decisions. In contrast to Bohnet and Frey (1999), we compare giving in the anonymous game with giving to anonymous individuals affiliated with political parties. As a result, we find that there is an important qualification to the observation that decreasing social distance increases giving: it is not just the more you know, but both the more you know plus how you feel about the target. Dictators can, as Camerer notes, show "empathy or contempt" (2003, 76). We select individuals from partisan groups as target recipients because political parties are among the most relevant groupings in political life. As Campbell et al. argue, "the strength and direction of party identification are facts of central importance in accounting for attitude and behavior" ([I960] 1980, 121). Identification with parties is typically measured with a 7-point Likert scale of subjective identification, although it has also been measured with self-reports on closeness to parties and implicit associations (Huddy 2003). These measures of social identification allow individuals to claim allegiance or closeness to groups, but they do not require individuals to sacrifice anything personally in making such a claim. Using dictator game allocations as a measure of social identification is a methodological innovation. The dictator game enables us to measure an individual's willingness not only to claim allegiance to a party but also to affirm that allegiance by withholding material benefits from the selfin order to transfer benefits to a different individual who happens to be a member of the ingroup. Further, the nature of political competition makes it more socially acceptable for individuals to confer benefits to in-partisans and deny benefits to out-partisans (as opposed, to, say, racial groupings which might invoke social desirability concerns). This social acceptability thus improves our ability to distinguish between self-interested, social-identity-based, and altruistic behavior. BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION 819 In December 2004, about 350 subjects were recruited from undergraduate political science and sociology courses at a large Western public university to participate in a computer-based survey.8 Subjects were offered credit towards their course grade to participate in the study; 306 (about 85%) of them elected to do so. Each individual answered a number of standard questions regarding their socioeconomic status, political attitudes, and participation behavior and then played three dictator games.9 In one game, subjects are told, "You know nothing about this anonymous individual." In the other two games subjects are told, "The only thing you know about this individual is that he or she is a registered Republican [Democrat]." The order of these treatments is randomized, and a variable indicating the order is included in the analysis below. Allocations to the anonymous recipient reveal the degree to which each subject cares about the well-being of others generally, while allocations to the Democrat and Republican reveal the extent to which subjects are motivated by social identity. In a typical dictator game, subjects are given a small amount of money ($5 to $10) and they then give back the portion of the money they choose to allocate to the other player. This procedure can be very costly for larger samples, so we employ a different technique. Subjects are given 10 lottery tickets that each have an equal chance of winning a prize of $100.10 They are then given two identical opaque envelopes. They are asked to place the tickets they wish to keep for themselves in one envelope and the tickets they wish to share with an anonymous individual in the other envelope. They seal both envelopes, place the envelope designated for the anonymous individual in a locked 'Subjects range in age from 18 to 43 years; the average age is 21. The sample consists of 56% women and 43% minority; it is quite similar to the undergraduate body from which it is drawn (the undergraduate body is 56% female and 51% minority). The median family income is about $80,000 a year. The average subject leans left and Democratic—the modal response to the liberal-conservative 7-point scale is a "2," or "liberal" (30% of the sample), and 57% of subjects identify as Democratic. 'For a summary and exact question wording, see the online appendix. 10One important difference between our method and the typical dictator game is the stake size. Note that the expected value of the prize is only $100/N = $.33. Though economists sometimes criticize low-stakes experiments like this one, a survey of the experimental economics literature by Camerer and Hogarth (1999) shows that stake size has only a small effect on average behavior and the biggest effect of stakes on behavior is changing from zero to positive stakes. Furthermore, Forsythe et al. (1994) and Carpenter, Verhoogen, and Burks (2005) show specifically for the dictator game that changing from low stakes to high stakes has no effect on mean allocations. mailbox under their computer, and then keep the other envelope for themselves. They then type on the computer the number of tickets they kept for themselves.11 Computers and the locked mailboxes are separated by partitions to protect the anonymity of choices each subject makes. After the study a ticket number for each of the three dictator game prizes was drawn and announced by email to participants.12 All three prizes were claimed. Partisanship and Dictator Game Allocations We begin with mean allocations for each of the three kinds of recipients. In general, results from the dictator game in this experiment appear to be similar to those of other researchers. Forsythe et al. (1994) specifically compare "with pay" dictator games in which subjects are given $5 or $10 to divide and "without pay" dictator games in which subjects are asked to make hypothetical choices. They find that more people keep everything for themselves in the "with pay" treatment (30.4% vs. 13.0%) and the mean allocation is lower (22.6% vs. 38.7%). By comparison, subjects in this experiment were even more likely (38.0%) to keep everything for themselves than those in both treatments. However, the mean allocation (29.9%) falls between the two treatments. This suggests that the lottery mechanism used in our design is replicating at least some of the incentives from dictator games that use cash stakes.13 UA chi-square test of the distribution of computer responses and the distribution of tickets that were physically placed in the mailboxes suggests that these two distributions are not statistically different. 12In many dictator games the recipients are also subjects. This was not true in our experiment—recipients are drawn randomly from the U.S. population (this is also the case in Johannesson and Persson 2000, who send the allocations to a randomly drawn individual in the Swedish population). Increasing the social distance between the dictator and the recipient should minimize the potential effect of reciprocity and thus make altruism a more compelling explanation for excess giving (Johannesson and Persson 2000). We did not hand all the envelopes with donated tickets to randomly chosen individuals. Instead, we waited to see if a donor claimed the prize for a given dictator game. If they did not, then we used random digit dialing to locate an individual and request their name and address (and partisanship for the Republican and Democrat treatments) and mail them the prize. 13Another way to compare the results of this experiment to the existing literature is by examining the relationship between dictator game allocations and demographic variables. Camerer (2003) notes that most demographic factors have little effect on dictator game allocations, but there are two notable exceptions. Carpenter, 820 JAMES H. FOWLER AND CINDY D. KAM Table 1 Partisanship and Allocations in the Dictator Game Donor Anonymous Recipient Democrat Recipient Republican Recipient N All 29.9% 30.1 27.3 306 Democrat 29.6 31.5 26.0 173 Republican 29.2 29.6 32.7 78 Difference .4 1.9 -6.7 p-value .41 .23 .02 Note: p-values reflect probability that true relationship is opposite to the sign of the difference (Wilcoxon signed rank test). Before moving to tests of the relationship between social identity, altruism, and participation, we discuss how social identity manifests itself in the dictator game results. No previous studies have examined the effect of partisanship on dictator game allocations, so we probe this relationship in some detail. Our experimental design provides us with a unique opportunity to test several hypotheses about social identification behavior as it applies to partisanship: Preference for the ingroup over the outgroup. Partisan identifiers will be more generous when asked to allocate rewards between themselves and a member of their own party compared with a member of the opposition party. Partisan identifiers will also be more generous to an ingroup member compared with someone not in the ingroup (the anonymous individual). This implies that Democratic identifiers will give more to a Democratic target than a Republican target, and Democratic identifiers will give more to a Democratic target than an anonymous individual. Likewise, Republican identifiers will give more to a Republican target than a Democratic target, and Republican identifiers will give more to a Republican target than an anonymous individual. Independents will give less to a Democratic target and a Republican target than to the anonymous individual, since partisan targets are more obviously an "outgroup" to independents than an anonymous individual would be. Strength of social identity. The stronger the partisan attachment, the more the ingroup should be rewarded and the more the outgroup should be deprived. As such, we would expect strong Democrats to give more to a Democratic target than weak Democrats would and strong Republicans to give more to a Verhoogen, and Burks (2005) find that subjects with higher family incomes tend to give less, while Eckel and Grossman (1998) find that women tend to give more. Our results replicate both findings. Consistent with results from other dictator games, subjects from families with low incomes (below the median) give 6.4% more than subjects from families with high incomes to the anonymous recipient. Further, in this experiment, women give away 6.1% more tickets than men. Republican target than weak Republicans would. Conversely, strong Democrats will likely withhold more from a Republican target than weak Democrats will; strong Republicans will withhold more from a Democratic target than weak Republicans will. Bias against Republicans. Experimental work suggests that individuals may discriminate against members of different groups when they are choosing whether or not to bear a personal cost to help them. Additionally, considerations of deservingness enter into dictators' decisions. For example, Eckel and Grossman (1996) note that altruism increases when the recipient appears to be more "deserving" or in need of resources (for more on deservingness, see Hoffman, McCabe, and Smith 1996b; Burrows and Loomes 1994). The Republican Party has typically been associated with business interests and the wealthy, whereas the Democratic Party has typically been associated with the working class and the less advantaged (Bastedo and Lodge 1980; Campbell et al. [1960] 1980; Miller, Wlezien, and Hildreth 1991). These associations imply that, on average, individuals might be less generous towards a Republican target compared with a Democratic target. Table 1 shows mean allocations in the dictator game by partisanship of the donor and recipient. Notice first that the Republican recipient receives 2.8% less on average than the Democrat from all donors. However, this difference may be due to the larger number of Democrats in the sample. When we take into account the partisanship of the donor, mean allocations tend to diverge along party lines. Subjects who identify themselves as Democrats and Republicans both give about the same amount to the anonymous recipient, but they tend to give more to the recipient from their own party, suggesting ingroup favoritism occurs. Notice that the Republican recipient inspires the largest divergence, receiving 6.7% more from Republican donors than Democratic donors, or about a fifth of the mean allocation. Table 2 indicates that both direction and strength of partisanship are significantly related to dictator BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION 821 Table 2 Strength of Partisanship and Allocations in the Dictator Game Partisan Recipients Donor Anonymous Recipient In-Party Out-Party N Strong Partisan 24.4 31.7 23.3 127 Weak Partisan 34.6 32.0 31.0 124 Independents 32.0 26.7 55 Table 3 Within-Subject Difference in Giving to Anonymous and Partisan Recipients Amount Given to Amount Given to Amount Given to Republican Minus Amount Republican Minus Amount Democrat Minus Amount Given to Democrat Given to Anonymous Given to Anonymous Donor Mean p-value Mean p-value Mean p-value All -2.8% .01 -2.2 .09 .3 .28 Democrat -5.5 .00 -3.6 .03 1.9 .19 Republican 2.8 .16 4.1 .10 1.2 .33 Independent -2.1 .32 -6.2 .19 -6.2 .05 Note: p-values reflect probability that true relationship is opposite to the sign of the difference (Wilcoxon ranked sign test). game allocations. Strong partisans give most to in-party targets, and they give significantly less to the anonymous recipient (Wilcoxon signed rank test, p = .007) and to the out-party target (p = .001). Weak partisans show about the same degree of favoritism towards the in-party target, but they show much less hostility towards the out-party target, compared with strong partisans. Weak partisans also show much more generosity towards the anonymous recipient compared with strong partisans. This evidence suggests that strength of partisanship does not necessarily affect generosity towards the ingroup, but it does affect punishment of the outgroup. The stronger the partisanship, the greater the propensity to withhold benefits from those not explicitly affiliated with the ingroup. We also see that independents make distinctions as well—but differently from partisans. Independents are much more inclined to be generous towards an anonymous individual compared with a partisan identifier. This is consistent with our expectation that independents see themselves as a group separate from the major political parties and thus are less generous towards these explicit outgroups compared with the anonymous individual. Recall that a single subject participates in three dictator games, so our design enables us to determine how the partisanship of the target recipient affects the within-subject tendency to give more to some political groups and less to others. About 61.7% of the subjects gave exactly the same amount to the registered Democrat and registered Republican that they gave to the anonymous recipient. The remaining 38.3% of the subjects discriminated across targets, changing their allocation in at least one of the games based solely on information about the partisanship of the recipient. Table 3 shows each of the three possible combinations of within-subject differences in the amount given to the anonymous and partisan recipients and how this breaks down by partisanship of the donor. First, note that the average subject gave somewhat less to the Republican than the Democrat or anonymous donor, yielding additional evidence for an anti-Republican bias in giving. The partisan identity of the donor also seems to have an effect on allocations. Democrats give significantly less to the Republican than to the Democrat or anonymous recipient. Republicans give more to the Republican than to the Democrat or anonymous recipient, but the significance of the difference is weak. Once again, the raw data appears to suggest an in-party bias, with the strongest difference in behavior exhibited by Democrats towards Republicans. Finally, people who did not identify themselves as either a Democrat or Republican tend to give less to both the Republican and Democratic recipient. In fact, the mean difference for 822 JAMES H. FOWLER AND CINDY D. KAM Table 4 Determinants of Within-Subject Differences in Giving in Dictator Games with Anonymous and Partisan Recipients Amount Given to Amount Given to Amount Given to Republican Minus Amount Republican Minus Amount Democrat Minus Amount Given to Democrat Given to Anonymous Given to Anonymous Coef. S.E. P Coef. S.E. P Coef. S.E. P Donor Characteristics: Partisan .11 (.05) .01 .12 (.05) .01 .01 (.04) .42 Identification Partisan Strength -.03 (.06) .32 .14 (.06) .02 .16 (.06) .00 High Income .01 (.03) .41 .04 (.04) .11 .04 (.03) .13 Female -.05 (.03) .08 -.07 (.03) .03 -.02 (.03) .24 White .01 (.04) .37 -.02 (.04) .28 -.03 (.03) .15 Order Variables: Republican First -.04 (.04) .15 .00 (.04) .46 .04 (.04) .10 Democrat First -.04 (.04) .15 -.04 (.04) .14 .00 (.04) .47 Constant .00 (.06) .48 -.12 (.06) .03 -.12 (.06) .01 Log scale variable -1.28 (.04) .00 -1.24 (.04) .00 -1.38 (.04) .00 Deviance / Null 43 / 57 108 / 123 23 / 35 Deviance Note: N= 300. Interval regression, where dependent variable is within-subject difference in allocation in the dictator game. All independent variables are dichotomous except the partisan variables which are scaled from 0 to 1. Order variables indicate which dictator game subject played first. both is exactly the same at 6.2%, or about one-fifth of the mean allocation. Although these differences are only weakly significant, they lend qualified support to the strength of social identification observed above. Partisans tend to receive less from nonpartisans and vice versa. To further assess the effect of partisanship on differences in dictator game allocations, Table 4 presents results from three sets of multiple regressions that also control for demographic factors.14 The first set of results analyzes partisan discrimination: the extent to which subjects make distinctions between the Republican and Democratic targets. We see that partisan identification has a positive and significant effect on the difference in the amount given to the Republican versus the Democrat, providing additional evidence for ingroup preference and outgroup hostility. The direction of partisan identification also has a positive and significant effect on the difference in the amount allocated to the Republican versus the 14In Tables 4 and 5 we use interval regression because the dependent variable is truncated at its minimum and maximum value. This estimation method is common in the literature on dictator games (e.g., see Carpenter, Verhoogen, and Burks 2005) and is conducted using maximum likelihood. We scale all variables from 0 to 1 for ease of comparison across coefficients, and we report the residual deviance of the model and compare it to the null deviance of a model with a constant. anonymous recipient. Given that there is no such effect for the difference in giving between the Democrat and the anonymous recipient, these two findings suggest that subjects give less to Republicans than other kinds of recipients and an anti-Republican bias exists. The regressions reveal partisan strength bias: strong partisans give 14% and 16% more than independents do to the Republican and Democratic targets, respectively. These results suggest that strong partisans see themselves as part of two ingroups, rewarding members of their own party at the expense of the opposing party, and rewarding members of any party over those who do not affiliate with a party. Altruism, Social Identity, and Political Participation Our main expectation is that those who are motivated by altruism and by social identity will participate in politics more than those who are motivated by material self-interest. To test this expectation, we create a 7-point scale of participatory acts, including voting, contributing to a candidate, joining a political organi- BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION Table 5 Altruism, Social Identification, and Political Participation 823 Dependent Variable: Political Activity Index Simple Model Model with Controls Coef. S.E. p Coef. S.E. p Other-Regarding Variables Altruism .74 (.39) .03 .66 (.29) .01 Social Identifier .57 (.24) .01 .40 (.18) .01 Political Variables Partisan -.46 (.26) .04 Identification Partisan Strength 1.05 (.34) .00 Political Interest 2.66 (.40) .00 Political Information .18 (.34) .30 External Efficacy .11 (.50) .41 Civic Duty .34 (.25) .08 Socioeconomic Status High Income -.22 (.20) .13 Female -.29 (.18) .06 White .26 (.19) .08 Citizen .85 (.45) .03 Skills Give Presentation .05 (-22) .41 Write Letter .89 (.19) .00 Make Decisions .40 (.21) .03 Chair Meeting .08 (.23) .36 Constant 1.86 (.18) .00 -2.25 (.61) .00 Log Scale Variable .72 (.05) .00 .36 (.05) .00 Deviance / Null Dev. 1216 / 1226 1081 / 1226 Note: N = 300. Interval regression, where dependent variable is the sum of political activities in which an individual participates (an integer from 0 to 7). All independent variables are dichotomous or scaled from 0 to 1. zation, donating to a political organization, attending a local board meeting, volunteering for a local board, and protesting.15 Participation is a function of the benefits that individuals receive, and benefits can be decomposed 15About 73.4% said they voted in the 2004 general election (compared to official turnout of 81.4% for the city in which the study took place) while 20.6% said they had given money to a candidate. About 36.0% claimed to belong to a political organization but only 24.9% had given money to one. Two questions about local politics show that 19.3% regularly attend board meetings while 25.2% had volunteered at least once to serve in some capacity for a board. Finally, 42.2% said they had participated in at least one political protest. A participation index was created using an equally weighted sum of responses to each of these seven questions. The average subject participated in 2.41 of these activities, with 12.5% of them never participating in any activity and 2.3% participating in all of them. The correlation between the participation index and the first component of a principal components analysis of these seven activities is .981 (±.004, 95% confidence), suggesting that the index captures the main dimension that these activities share in common. into three categories: benefits to the self, benefits to society generally, and benefits to a preferred group. First, the benefits from participation should increase as altruism increases, and second, the benefits from participation should increase as social identification increases. We operationalize the altruism incentive by using the proportion of tickets allocated towards the anonymous target in the dictator game, since this best captures the extent to which individuals are willing to give to others in general.16 For the social identity incentive, we use a dummy variable that is 0 if an individual gives the same amount in all three dictator ^Alternative specifications such as averaging the allocations for all three dictator games or using a dummy variable for individuals who gave more than the median amount yielded substantively identical results. 824 JAMES H. FOWLER AND CINDY D. KAM games, and 1 otherwise, indicating the individual discriminates in giving based on partisanship of the target.17 The raw data provide initial support for the notion that both altruism and social identity drive political participation. Those who gave more than the median allocation (30% of their tickets) to the anonymous recipient participated in 2.64 of 7 activities compared to 2.24 activities for those who gave 30% or less. Social identifiers (those who gave different amounts depending on the partisanship of the recipient) participated in 2.74 activities compared to 2.21 for those who gave the same amount to all three recipients. One-sided f-tests suggest that both of these differences are significant (p = .03, p — .01, respectively). We begin by estimating a simple model in which the participation index is regressed on the altruism and social identifier variables. This simple model appears in the first column of results in Table 5. Notice that even when we consider both altruism and social identity together in a single model, they continue to be positively and significantly related to participation. Table 5 also shows that when multiple covariates widely thought to affect participation are included in the model, altruism and social identity continue to have a strong and significant effect on participation.18 To make these results more concrete, subjects who give everything to the anonymous recipient in the dictator game participate in .66 more activities than subjects who keep everything for themselves. In other words, altruists appear to be more likely to participate in politics than egoists. Moreover, subjects who change the amount they give based on the partisanship of the recipient also participate in .40 more activities than those who give the same amount to each recipient. Thus, social identifiers participate in politics more than individuals who weigh benefits to all groups equally. Since variables in the model are dichotomous 17We also included various measures to capture the strength of social identification by incorporating the difference of all three allocations into a single variable, such as their variance or standard deviation. These alternative measures yielded substantively identical results, but we use the dummy variable approach here for transparency. 18We do not attempt to provide a comprehensive explanation of political participation. We aim to add the concepts of altruism and social identity to existing explanations of participation. Some of the factors that influence participation also influence allocations in the dictator game. In our regression model, we include a series of controls to rule out confounding factors that would bias our estimates of the effects of altruism and social identity (see Verba, Schlozman, and Brady 1995 for a comprehensive treatment of the control variables). Coding and question wording can be found in the online appendix. or scaled to range from 0 (sample minimum) to 1 (sample maximum), we can roughly compare effect sizes between independent variables by looking directly at the coefficients. Notice that the altruism and social identifier variables have a stronger effect than many other variables thought to be important. Only partisan strength, political interest, citizenship, and letter-writing skills are stronger predictors. Thus, these findings suggest that self-interest is not the only consideration that drives political participation. Rather, regard for others, generally, and regard for specific others, affiliated with groups, both predict participation. Conclusion While there can be no doubt that much of human behavior is motivated by self-interest, the results in this article suggest that other-regarding behavior may also contribute to political participation. Altruists who want to help others regardless of their group affiliation may have a larger incentive to participate than those who are merely self-interested. However, this will only be true when political outcomes are perceived as generating benefits for everyone—if political outcomes are perceived as being distributive, altruists gain nothing from shifting resources from one group to another. In contrast, social identifiers gain the most from participation when politics is distributive, since this gives them an opportunity to help acquire benefits for their ingroup, and better so if this occurs at the expense of outgroups. Since political outcomes are frequently viewed as improving the general welfare as well as posing more generous gains to some groups over others, both altruists and social identifiers ought to participate more frequently than egoists, who are purely self-interested. Our results show that social identity has an important effect on allocation decisions. Subjects exhibit a preference for the ingroup over the outgroup. Democrats and Republicans both give more to the recipient from their own party than the opposing party, and independents give more to the anonymous recipient than the partisan recipients, while partisans do just the opposite. These preferences are magnified by the strength of social identity. Subjects who identify themselves as strong Democrats and strong Republicans tend to give much less to the recipient from the opposing party than other partisans. We then use the dictator game allocations to test the altruism and social identity theories of participation and find that the evidence supports both theories. BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION 825 People who share with an anonymous individual in the dictator game participate more in politics than those who do not share. People who vary the amount they give depending on the partisan affiliation of the recipient also participate more in politics than those who give the same amount to everyone. Participation in political life is driven by considerations beyond the self. We use dictator games in a laboratory setting to measure self-oriented, social-identity-oriented, and altruistic dispositions. The primary explanation for giving in dictator games focuses on other-regardingness, and this is the interpretation that we take. However, we note that other interpretations exist: subjects may give in order to fulfill an external standard of fairness, or because they feel it is their "duty" to make donations—that is, they donate in order to comport with some external standard for appropriate behavior. Or, they might give to fulfill norms of reciprocity, or they might give randomly. We think the existing evidence rallies primarily around other-regardingness as an explanation, and hence we interpret our results as such. Our study, like most studies utilizing experimental economics, examines dictator behaviors among college students. A standard criticism of studies utilizing convenience samples is that these samples are atypical of the general population, and any results are thus limited in their applicability to the general public. We note, however, that one must establish that the student sample is atypical from the general population in ways that are relevant to the study in question (Sears 1986). Our core contribution is in identifying an innovative way to measure altruism and social identity and showing the empirical relationship between these measures and political activity. How would our results translate to the general public? Perhaps college students who are in a repeated-interactions environment would display, on average, higher levels of excess giving (to an anonymous individual and to a partisan) than members of the general population. It follows that perhaps the level of altruism and social identity may be higher in our convenience sample compared with a representative sample. However, we have no reason to expect that the relationship between altruistic and identity-based giving would differentially predict political participation. Hence, although we acknowledge that our empirical example may hold limited generalizability, we do not dismiss the possibility that these results could be replicated in the general population. (And, in fact, we are in the process of replicating these results in a general population study). Altruism and social identity are likely to have broader applications beyond political participation, and our innovative measures might serve other researchers' purposes in this regard. At a very general level, altruism and social identity might have implications for individuals' understandings of politics and subsequent beliefs about political processes. Social identifiers may see politics as a competition among groups for governmental outputs, and thus they would favor political processes that would allow groups opportunities to press for their own cases. Altruists may see politics as a forum for the production of policies to improve the public good, and thus they might favor political processes that foster wide participation and dialogue. More narrowly, altruism and social identity could have implications for policy preferences. Altruists may oppose policies that are targeted at specific groups and instead favor policies that are more generally applied, much as humanitarians might (Feldman and Steenbergen 2001). Social identifiers are likely to support policies that disproportionately help their own group, to oppose policies that help other groups, and perhaps to provide the most support for policies that increase the standing of their own group at the expense of other groups. Finally, the altruism and social identifier theories of participation have important implications for rational choice. The rationality assumption means only that people have preferences that are complete and transitive. Notice that the words "self-interest" appear nowhere in this definition (Jackman 1993). While it is true that most rational models are based on material self-interest, a concern for others need not be excluded from these models. Social identity theory suggests people gain utility by helping their ingroup, often at the expense of an outgroup. Theories of altruism suggest that people gain utility by providing benefits to others, even when it is personally costly. Rational calculations need not be limited to narrow definitions of material self-interest, especially since such models have failed to explain observable behavior. The evidence clearly suggests that individuals look beyond the self when deciding whether or not to participate in politics. Acknowledgments We would like to thank Arthur Schram, Stephen Coate, Michael Conlin, Ben Highton, Macartan Humphreys, Robert Jackman, Mark Lubell, and Phil Paulino for helpful comments. A prior version of this 826 JAMES H. FOWLER AND CINDY D. KAM paper was prepared for delivery at the 2005 annual meeting of the Midwest Political Science Association. Manuscript submitted 12 January 2006 Manuscript accepted for publication 16 October 2006 References Aldrich, John H. 1993. "Rational Choice and Turnout." American Journal of Political Science 37 (1): 246-78. Andreoni, James, and John Miller. 2002. "Giving According to Garp: An Experimental Test of the Consistency of Preferences for Altruism." Econometrica 70 (2): 737-53. Bastedo, Ralph W., and Milton Lodge. 1980. "The Meaning of Party Labels." Political Behavior 2 (3): 287-308. Bohnet, Iris, and Bruno S. Frey. 1999. "Social Distance and Other-Regarding Behavior in Dictator Games: Comment." American Economic Review 89 (1): 335-39. Bolton, Gary E., and Axel Ockenfels. 2000. "ERC: A Theory of Equity, Reciprocity, and Competition." American Economic Review 90 (1): 166-93. Burrows, Paul, and Graham Loomes. 1994. "The Impact of Fairness on Bargaining Behavior." In Experimental Economics, ed. John D. Jey. Heidelberg: Physica-Verlag, pp. 21-41. Camerer, Colin F. 2003. Behavioral Game Theory: Experiments in Strategic Interaction. Princeton, NJ: Princeton University Press. Camerer, Colin F., and Robin M. Hogarth. 1999. "The Effects of Financial Incentives in Experiments: A Review and Capital-Labor-Production Framework." Journal of Risk and Uncertainty 19 (1-3): 7-42. Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E. Stokes. [1960] 1980. The American Voter: Unabridged Edition. Chicago: University of Chicago Press. Carpenter, Jeffrey, Eric Verhoogen, and Stephen Burks. 2005. "The Effects of Stakes in Distribution Experiments." Economics Letters 86 (3): 393-98. Citrin, Jack, and Donald Philip Green. 1990. "The Self-interest Motive in American Public Opinion." Research in Micropolitics 3: 1-28. Coate, Stephen, and Michael Conlin. 2004. "A Group Rule-Utilitarian Approach to Voter Turnout: Theory and Evidence." American Economic Review 94 (5): 1476-504. Converse, Philip E. 1964. "The Nature of Belief Systems in Mass Publics." In Ideology and Discontent, ed. David E. Apter. New York: Free Press, pp. 206-61. Downs, Anthony. [1957] 1985. An Economic Theory of Democracy. New York: Harper. Eckel, Catherine C, and Philip J. Grossman. 1996. "Altruism in Anonymous Dictator Games." Games and Economic Behavior 16 (2): 181-91. Eckel, Catherine C, and Philip J. Grossman. 1998. "Are Women Less Selfish than Men? Evidence from Dictator Experiments." Economic Journal 108 (448): 726-35. Edlin, Aaron, Andrew Gelman, and Noah Kaplan. 2007. "Voting as a Rational Choice: Why and How People Vote to Improve the Weil-Being of Others." Rationality and Society. Forthcoming. Feddersen, Timothy J., and Wolfgang Pesendorfer. 1996. "The Swing Voter's Curse." American Economic Review 86 (3): 404-24. Feddersen, Timothy J., and Alvaro Sandroni. 2006a. "Ethical Voters and Costly Information Acquisition." Quarterly Journal of Political Science 1 (3): 287-311. Feddersen, Timothy J., and Alvaro Sandroni. 2006b. "A Theory of Participation in Elections." American Economic Review 96 (4): 1271-82. Fehr, Ernst, and Urs Fischbacher. 2003. "The Nature of Human Altruism." Nature 425 (6960): 785-91. Fehr, Ernst, and Klaus M. Schmidt. 1999. "A Theory of Fairness, Competition, and Cooperation." Quarterly Journal of Economics 114 (3): 817-68. Feldman, Stanley, and Marco Steenbergen. 2001. "The Humanitarian Foundation of Public Support for Social Welfare." American Journal of Political Science 45(3):658-77. Forsythe, Robert, Joel L. Horowitz, N. E. Savin, and Martin Sefton. 1994. "Fairness in Simple Bargaining Experiments." Games and Economic Behavior 6 (3): 347-69. Fowler, James H. 2006. "Altruism and Turnout." Journal of Politics 68 (3): 674-83. Hibbing, John R., and John R. Alford. 2004. "Accepting Authoritative Decisions: Humans as Wary Cooperators." American Journal of Political Science 48 (1): 62-76. Hoffman, Elizabeth, Kevin A. McCabe, Keith Shachat, and Vernon L. Smith. 1994. "Preferences, Property-Rights, and Anonymity in Bargaining Games." Games and Economic Behavior 7 (3): 346-80. Hoffman, Elizabeth, Kevin A. McCabe, and Vernon L. Smith. 1996a. "On Expectations and the Monetary Stakes in Ultimatum Games." International Journal of Game Theory 25 (3): 289-301. Hoffman, Elizabeth, Kevin A. McCabe, and Vernon L. Smith. 1996b. "Social Distance and Other-Regarding Behavior in Dictator Games." American Economic Review 86 (3): 653-60. Huddy, Leonie. 2003. "Group Identity and Political Cohesion." In Oxford Handbook of Political Psychology, ed. David O. Sears, Leonie Huddy, and Robert Jervis. New York: Oxford University Press, pp. 511-58. Hume, David. [1742] 1991. "Of the Rise and Progress of the Arts and Sciences." In Essays: Moral, Political, and Literary, ed. Eugene F. Miller. Indianapolis: Liberty Classics, pp. 111-137. Jackman, Robert W. 1993. "Rationality and Political Participation." American Journal of Political Science 37 (1): 279-90. Jankowski, Richard. 2002. "Buying a Lottery Ticket to Help the Poor: Altruism, Civic Duty, and Self-interest in the Decision to Vote." Rationality and Society 14 (1): 55-77. Jankowski, Richard. 2007. "Altruism and the Decision to Vote: Explaining and Testing High Voter Turnout." Rationality and Society 19 (10): 5-34. Johannesson, Magnus, and Björn Persson. 2000. "Non-Reciprocal Altruism in Dictator Games." Economics Letters 69 (2): 137-42. Kinder, Donald R., and Nicholas Winter. 2001. "Exploring the Racial Divide: Blacks, Whites, and Opinion on National Policy." American Journal of Political Science 45 (2): 439-56. Knack, Stephen. 1992. Social altruism and voter turnout: Evidence from the 1991 NES Pilot Study. NES Pilot Study Report, No. nes002294. Ledyard, John O. 1982. "The Paradox of Voting and Candidate Competition." In Essays in Contemporary Fields of Economics: In Honor of Emanuel T. Weiler, eds. George Horwich and James P. Quirk. West Lafayette, IN: Purdue University Press, pp. 54-80. Mansbridge, Jane J. 1990. "The Rise and Fall of Self-interest in the Explanation of Political Life." In Beyond Self-Interest, ed. Jane J. Mansbridge. Chicago: University of Chicago Press, 3-22. BEYOND THE SELF: SOCIAL IDENTITY, ALTRUISM, AND POLITICAL PARTICIPATION 827 Miller, Arthur H., Christopher Wlezien, and Anne Hildreth. 1991. "A Reference Group Theory of Partisan Coalitions." Journal of Politics 53 (4): 1134-49. Monroe, Kristen Renwick. 1996. The Heart of Altruism. Princeton, NJ: Princeton University Press. Palfrey, Thomas R., and Howard Rosenthal. 1985. "Voter Participation and Strategic Uncertainty." American Political Science Review 79 (1): 62-78. Piliavin, Jane Allyn, and Hon-Wen Charng. 1990. "Altruism: A Review of Recent Theory and Research." Annual Review of Sociology 16: 27-65. Price, Vincent. 1989. "Social Identification and Public Opinion: Effects of Communicating Group Conflict." Public Opinion Quarterly 53 (2): 197-224. Riker, William H., and Peter C. Ordeshook. 1968. "A Theory of the Calculus of Voting." American Political Science Review 62 (1): 25-42. Saad, Gad, and Tripat Gill. 2001. "The Effects of a Recipient's Gender in a Modified Dictator Game." Applied Economics Letters 8 (7): 463-66. Samuelson, Paul A. 1993. "Altruism as a Problem Involving Group Versus Individual Selection in Economics and Biology." American Economic Review 83 (2): 143-48. Schelling, Thomas C. 1968. "The Life You Save May Be Your Own." In Problems in Public Expenditure Analysis, ed. S. B. Chase. Washington, DC: Brookings Institute, 127-62. Schlozman, Kay Lehman, Sidney Verba, and Henry E. Brady. 1995. "Participation's Not a Paradox: The View from American Activists." British Journal of Political Science 25 (1): 1-36. Sears, David O. 1986. "College Sophomores in the Laboratory: Influences of a Narrow Data Base on Social Psychology's View of Human Nature." Journal of Personality and Social Psychology 51 (3): 515-30. Sears, David O., and Carolyn L. Funk. 1991. "The Role of Self-interest in Social and Political Attitudes." Advances in Experimental Social Psychology 24: 1-91. Smith, Vernon L. 2000. Bargaining and Market Behavior. Cambridge: Cambridge University Press. Tajfel, H. 1981. Human Groups and Social Categories. Cambridge: Cambridge University Press. Uhlaner, Carole J. 1989. "Rational Turnout: The Neglected Role of Groups." American Journal of Political Science 33 (2): 390-422. Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. 1995. Voice and Equality: Civic Voluntarism in American Politics. Cambridge: Harvard University Press. James H. Fowler is associate professor of political science, University of California—San Diego, San Diego, CA 92093. Cindy D. Kam is assistant professor of political science, University of California—Davis, Davis, CA 95616.