JURNAL ILMU PEMERINTAHAN: Kajian Ilmu Pemerintahan dan Politik Daerah Volume 9 Nomor 2, 2024, (Hlm 128-. Available online at: http://e-journal. id/index. php/jip How Do Informal Actors Work? Between Voter Lists and Social Pressure in Elections George Towar Ikbal Tawakkala* aUniversitas Brawijaya. Jl. Veteran. Malang 65145. Indonesi Information Article ABSTRACT History Article Submission : 30 Oct 2023 Revision : 8 Nov 2024 Published : 9 Nov 2024 Middleman in election, some scholars called them as vote broker, is an actor who mobilize voter. Their works would have a clear and measurable commodity. Candidates will not hire middleman when commodity can not be measured. Similarly, a middleman will not get job if there is no clear commodity that can be offered to candidate. That logic have began this study. Facts, on field indicated the debate about whether commodities . actually were measured. Some middlemen registered voters, and others did not, destabilizing the logic. Is this activity measured and unmeasured? Do both have different measurement logic? In what did the differeneces work?. To answer these questions. I made observations and interviews in Pati at 2011-2024, where the middleman called as Sabet, and in Demak at 2015-2024, where the middleman called as Gapit. Findings, this activity always measurable on two methods. It was based on list of voters name, and other was based on influence that a middleman will have over voters in his respective social network. First method led to distributing money or valuable goods to voters on the list. Second method led to using social pressure to influence voters. Different type of voters made candidates assign such divergent strategies DOI Article: https://doi. org/10. 5/jip. Key word: Middleman. Sabet. Gapit. Measurable. Commodity A 2024 Published by Governmental Studies. Selection and/or peer-review under responsibility of JIP. * Corresponding author E-mail address: george. ikbal@ub. Copyright A 2018. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 129 George Towar Ikbal Tawakkal INTRODUCTION In emerging democracy countries, some variations arise in the practice of democracy. One of them is the middleman in election. At least, the middlemen can be found in Brazil (Gingerich & Medina, 2013. Cooperman, 2. Argentina (Zarazaga, 2. Egypt (Blaydes, 2. Paraguay (Finan & Schechter, 2. Taiwan. Thailand (Hellmann, 2. , even in developed countries, e. Russia (Frye et al, 2. and China (Kennedy, 2010. Ma et al. , 2. In Indonesia, the literature showed the middlemen also be found (Noak, 2024. Tawakkal et al. , 2017. Aspinall, 2014. Aspinall & Sukmajati, 2015. Triantini, 2015. Noor Rahman, 2. We need to make clear about this term. Few scholars called them as the middlemen, while most scholars called them as vote In this article. I want to make fair by calling them as the middlemen. It is important to start the article. Why do we need to do this? By putting them as the middleman, we can capture all possibility of them, not only transactional values who attached in brokerAos term. Literature provided an understanding that the actors played a role as an intermediary, which bridges between candidates and voters. For example. Finan and Schechter . , described it with the phrase ". to interact with voters . ", which shows their role as an actor who helps candidates for connecting with voters indirectly. Other scholars, such as Aspinall . more explicitly called it a material distribution channel to voters, so that he divided into three characteristics related to their loyalty as a distribution channel. The actors were interpreted as people who bridge between candidates and voters, to collect votes for candidates (Blaydes, 2006. Hellmann, 2004. Gonzalez-Ocantos et al. , 2. Considering their position as an intermediary, sometimes vote brokers were also called as a middleman (Tawakkal, 2022. Finan & Schechter. Scott, 1. I prefer to use the last. The main point is that we call them in different names, but refer to one actor. They are people or group of people who have no party affiliation and provide votes for candidates. The scholars did not highlight that they have no party affiliations but easy to be understood from their article that the middlemen have no party affiliation. So we need to make clear that the middlemen are not political party leaders or members or part of party In some case, we will get difficultness to distinguish them, but we will get easy by seeing official administrative such registered as a party member or not. Many kinds of literature put the middlemen as distributing channel. Hellmann . shown the candidates mainly relied on vote-buying, based on dense networks of Hua Khanaen who would serve as the link between the politician and the voter at the local level. In Paraguay. Finan and Schecter . found the politicians hire respected community leaders in each village to interact with voters to offer them money and other forms of aid in exchange for the promise of their vote. In Nicaragua. Gonzalez-Ocantos and friends . found they play a key role in distributing goods. Kennedy . explained the middlemen have also appeared in mainland China village elections. The terms used are agents or middle people who personally dispense the gifts or cash to voters. Blaydes . explained how the vote seller submitted the ballot in the polling station and returned a blank ballot to the middleman. The vote seller was paid after coming out of the station with this blank ballot. From those, we can conclude that the middlemen distributed money or gifts from candidates to voters. But, how if the middlemen did not distribute anything? We found them in 2019 Indonesian elections. As institutions that provide voters, candidates did not build them without considerations. One of the considerations is how many voters can be provided by an intermediary institution. Previous literature did not explain much about how to estimate the voters. Abadeer et al. , . and Blaydes . , for example, described how intermediary institutions supervise and control Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 130 George Towar Ikbal Tawakkal voters during Egyptian elections, to ensure that voters fulfill their commitment to vote for their He did not talk about the methods in estimating the number of voters they can provide. Likewise, other scholars (Mwonzora, 2024. Darwin, 2017. Baldwin, 2. explain the brokerAos background as a community leader, then Finan and Schechter . analyze the relationship between the middleman and norms of reciprocity. The use of social networks explained in Finan and SchechterAos research, however, did not explore how the candidates or middlemen estimate the size and the influence of the social networks, including the number of voters within their social Similarly, most previous studies have explored the middleman-candidate relationship, but not about how voters were calculated. A book by Aspinall and Sukmajati . demonstrated that brokers provide a list of voters within their social networks that had committed to vote for their candidate. The lists were submitted to candidates and used as the basis of distributing It means, literature demonstrated that the middlemen have the list for measuring voters. But, is it the only one? We found many middlemen in 2019 Indonesian elections, who did not make the list. How to explain that? This article will answer those questions. Exploring how the candidates have variations to estimate voters who can be resulted by a middleman, and how it will implicate in influencing This article consists of four sessions. The first presents the social background of the middlemen, it refers to the social status of the middlemen in their social environment, before they work in the elections. The second presents how voters were estimated by candidates during an election, including several methods that measured the voters. The third presents how the middlemen got commitment by voters through each methods, including how they implemented the methods to increase voterAos commitment for their candidate. The fifth, as a conclusion, interrelate social status, estimating methods, and influencing methods. METHOD We conducted interviews with more than forty middlemen in every level, five local party leaders, and more than fifty voters during 2011-2024, in Central Java, especially in Demak Regency and Pati Regency. Indonesia. Both have characteristics of rural communities. Reasoning for selecting rural communities, not urban, was because the intermediary activities occurred dominantly in village communities (Hirseland, 2024. Brierley & Nathan, 2022. Asmawati, 2021. Fadillah, 2020. Hellmann, 2014. Aspinall, 2014. Triantini, 2015. Noor Rahman, 2015. GonzalezOcantos et al. , 2011. Aidt et al. , 2020. Stokes et al. , 2013. Schaffer & Schedler, 2008. Wu & Huang. Rigger, 2. Both also have sociological backgrounds that represent the Javanese society. Refers to Geertz . , he divided Javanese society into three, namely santri, priyayi, and He, simply, defined Santri as community or person who practices Islamic values, while Abangan as community or person who practices Javanese traditional values and less Islamic. Then Priyayi is a community of civil servant or aristocrat or other high social positions. Santri and abangan are communities who dominates Java society, so I used both and skip the priyayi. Demak is dominated by santri community, while Pati is dominated abangan society. In both locations, the intermediary activities have been entrenched and presented a long time before Indonesia moved to be a democratic country. That can be seen by they have a local name for describing the In Pati, the middleman commonly was called as Sabet or Pecut, while in Demak was called as Gapit. Sabet. Pecut, and Gapit have the same meaning, refer to people who intermediate between candidates and voters in providing votes for the candidates. It was not easy to get and interact with the middlemen and voters. They tent to keep the information when a stranger sees and interviews them. It can be understood because of the Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 131 George Towar Ikbal Tawakkal intermediary activities often closely related to vote-buying. Vote-buying is an illegal practice in Indonesia, which is known as money politics. Fortunately, we created trust around them well. With this position, we had a channel to interact, talk with them, and observed some events, activities, or social interactions. In both locations, we also have access to communicate with some local party leaders. FINDINGS AND DISCUSSION Voters as a Commodity: Estimating Voters The candidates need voters to win the offices. The middlemen served the intermediaries between the candidates and the voters, and therefore attempt to provide as much of the voters to the candidate as possible. Yet there is uncertainty about how many voters a middleman can offer to the candidate and what cost per voter. One way in which the middlemen attempted to quantify the voters was by providing data to the candidates about the number of possible voters. The middlemen made lists of voters whom they can influence on behalf of the candidate and present the lists to candidates. This information was important because it allowed the candidate to estimate the cost of influencing voters and also provided an estimate of the number of votes the candidate can expect in that area on election day. Aspinall and Sukmajati . showed about how they used the lists. Our findings showed that the use of lists was not the only method. Some middlemen made the list and the others did not. Generally, we found two methods of estimating voters. One, as mentioned in Aspinall and SukmajatiAos book that the middlemen made the lists. In this article, we will discuss more that. The candidates asked the middlemen to make the lists by asking the voters. Voters who said that their votes will not for the candidate who hires the middlemen so be kicked out of the list. Some middlemen explained that they talked with voters for accepting their name on the list. When we talked with the voters, they confirm that the middlemen came to them for the list. But, some middlemen did not do it. We went back to the middlemen to make sure if they made the list. Also, the upper-level middlemen confirmed that they made the lists. They allowed us to see the list and found the voters who told us that the middlemen did not come to them for that. Some middlemen made the list with confirmation, and others made the list without confirmation. At this point, some candidates have other teams to check the validity of the list. The teams came down to the area and investigated the validity. Commonly, they will talk with the voters randomly. The interesting thing is they did not ask the middlemen to revise and remake the list. They use the validation to be considered for the next steps. Some candidates explained have no team to do that. It was about the resource to generate them. What they did is choosing the best middlemen so can avoid the After they made the list, they submit them to the middlemen in the upper level and go-ahead to the candidates. The candidates will use it to estimate their potency and their ability. The potency refers to estimating how many voters they can get, while the ability refers to calculating how many resources they have and can spend them for influencing voters. All considerations were on the table for the potency. Some candidates put the list which can not be confirmed to the trash, but others still used it as second options or additional voters. The candidates calculated their resources then. By number potential voters that have been decided before, the candidates calculated how much money they need to mobilize or influence This is the main point if the candidates will go or not. Some candidates explained that they have to be careful at this point. For candidates without big funds, they chose the lists which have Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 132 George Towar Ikbal Tawakkal trust more. If the number of voters will be enough for the office, they shared the gifts. If the number of voters will not be enough for the office, they will stop at this point. Unfortunately, some candidates still run even have no strong confidence about the lists. They said they have no options. Sure, the lists were not the only one variable for stopping or not. But, the lists became the first For a rich candidate, the unconfirmed lists were considered as additional voters. The candidate still shared the gifts but less in quantity. For example, if voters in the confirmed lists got 5 kilograms of rice, so the voters in the unconfirmed list got 2. 5 kilograms. They hoped the voters in the list are swing voters, so have opportunity to get the voters. The second one is, the middlemen did not made the list. Some candidates confirmed that they did not asked some middlemen to made the lists. We talked to the middlemen whom the candidates mentioned, and they confirmed that they were not asked to made the lists. The candidates explained that they will not place the middlemen as gift distribution channel. It means, the candidates will not distribute the gift through the middlemen. In vote buying logic, what can they did with that? No lists, no gifts. Why did the candidates hire them? The candidates tried to use patronage that existed between the middlemen and the candidates did not asked the middlemen to made the lists. So the middlemen did not made it. How did the candidates estimate the voters? The candidates estimate the influence that a broker will have over voters in his respective social network. Here, the broker is not expected to provide candidates with a specific list of names of supporters, but instead presents the candidate with estimates about the size of his social network generally and his ranking within society. In this situation, brokers rely less on cash payments and more on their social influence within their villages as community leaders, religious leaders, and other social elites. For example, a former village head . n elected offic. in Juwana was hired by a candidate because of the respect he had among the voters in his village. The campaign knew the population of the village and did not ask this person to provide a list of likely supporters, but instead relied upon his influence with voters in his village in order to gain support on Election Day. Measuring this social and personal influence is obviously more difficult for campaigns than merely acquiring a list of names of people committed to support a candidate. Yet if the campaign were intending to distribute money to voters, they would either need to provide money to all voters . hich is unlikel. or they would need a list of likely supporters, which they did not ask this broker to provide. Curiously, other brokers hired by the campaign to work in this same area were asked to provide lists of voter names to the candidates. The difference in how campaigns ask brokers to measure likely support in their respective area raises interesting questions about the nature of brokerage activity in Indonesian elections. Specifically, if brokerage activity is merely about vote buying then why are all brokers not asked to provide voter lists to the campaigns? That list is not only the best available measure . cientific polling is rare in these kinds of local election. for the campaign to determine likely support on Election Day, but is also critical for determining which voters receive gifts and which are ignored by the campaign. The different methods of calculating likely support suggests a more complex system of brokerage activity that includes voter contact strategies beyond simple vote buying. Voter Persuasion and Voter Lists The two different ways in which brokers estimate likely support among voters in their respective areas is related to the various methods used by campaigns attain commitments from voters for their candidates. As noted earlier, candidates ultimately care about acquiring votes for the election and brokers are hired to obtain commitments from voters to support their candidates. Brokers therefore provide various incentives to voters to gain their support for a given candidate. Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 133 George Towar Ikbal Tawakkal Most of the literature on brokerage activity focuses on vote buying practices such as giving cash or other gifts directly to voters. Blaydes explains how brokers gain support from voters in Egypt through such cash offerings, for example. Blaydes also describes additional methods of acquiring support that include opening a cheap market for voters that offers goods for sale, gift vouchers, and also clientele-based relationships (Blaydes, 2. In Paraguay. Finan and Schechter . explained how campaigns gain commitments from voters by offering gifts such as cash and ten asking voters to promise they will support the candidate (Finan & Schechter. Kennedy . describes the combination of policy benefits that the candidate will provide to citizens in addition to offering cash directly to the voters in exchange for their support. In Indonesia. AspinallAos influential theory of broker betrayal . is based on the notion that brokers have a financial incentive in some cases to keep the money that is intended for voters. Observing brokerage activity across several election cycles in both Juwana and Mranggen, however, demonstrated a remarkable variety of ways in which brokers tried to gain voter commitments that go beyond cash distribution. Vote buying was indeed prominent in these elections, and Joko was hired by a candidate to distribute money to voters during the 2011 and 2012 elections in Pati. Likewise, in Mranggen, several brokers were hired to distribute money to voters in their respective neighborhoods. In another village, a broker coordinated the construction of public facilities . inanced by the candidate, of cours. in various locations as a way to gain public support. It should be noted that voters had requested these public facilities and the candidate financed them with the obvious understanding that the voters would support him in the election. Yet other, often subtle, methods were employed by brokers to influence voters. Suroso, for example, held social activities for members of his community and told attendees individually about his support for his candidate. Unstated was that Suroso had been hired by the campaign to contact voters on its behalf. One broker surnamed Kelik, who is a prominent religious leader and owner of an educational institution in his village, made a statement to worshippers in his mosque as well as the students at his institution expressing his support for the candidate. Though Kelik did not mention the candidate by name, his statement was clear enough that his students and members of his mosque understood precisely which candidate he was supporting. None of these various activities involved either giving money to voters or even promising money or gifts by the candidate on Election Day. The obvious difference between these persuasion efforts has to do with the underlying mechanism used to influence voting decisions. With direct cash payments, there is a quid pro quo of casting a vote for a candidate in direct exchange of material goods. It is a transactional arrangement where brokers deliver such goods to voters on behalf of candidates and supervise the voters to ensure they make good on their commitment to support their candidate. The other methods, however, are social or societal in nature. A well-respected member of the community provides citizens with a public event or a public statement at their place of worship, schools, and/or workplaces. Within the context of these events and statements, there is a more subtle social pressure applied to citizens to support a candidate for religious, moral, or social reasons. One candidate, for example, explained that an important trait for hiring some brokers was that he be. Aua good man in the communityAy, meaning someone who voters respect and to whom they look for guidance when making important social . nd also politica. With the transactional arrangement Ae giving money directly to voters Ae there is something of a moral pressure that is based on honoring the transaction. The candidate gave them money and the brokers try to ensure that voters uphold their end of the economic agreement. This is the type of brokerage activity that requires extensive voter lists collected by brokers and provided to Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 134 George Towar Ikbal Tawakkal This is also true with the more collective distribution of goods such as the construction of public facilities, which tends to be more rare than direct personal payments to Yet in both, the pressure to support the candidate is based on honoring the voterAos end of the transaction. With the social arrangement, however, the pressure is based on trust of the community or religious leaderAos judgment. These leaders serve as cues for citizens, many of who do not even know the candidates or their policies. The persuasion attempt is based on subtle pressure creating with social networks, where a respected person in the community provides leadership and guidance to citizens of his community . ften in exchange for payment by the This type of influence is much harder to measure and therefore the broker is not required to create a list of voters precisely because the broker is hired for his social status and the influence that has over the community. Characteristics of Voters as Explanation Why would candidates assign such divergent strategies and brokerage activities to different brokers in their efforts to win elections? Follow up questions with the brokers revealed that the different types of activities Ae making lists for distributing cash versus working social networks with community events Ae are targeted at two very different types of voters. Interviews with the brokers as well as my own discussions with voters revealed that some voters are transactions while others are persuaded more by social factors. Transaction voters largely base their vote on the money they receive from candidates through their brokers. A common slogan made by voters is Auora ono duit ora nyoblosAy, which commonly translates to Auno money, no vote!Ay The implied threat is that the voter will stay home for the election if he does not receive cash or goods from the candidate. Transaction voters view the interaction with candidates similar to trading in a market and they are willing . ften eage. to make a firm commitment if money is promised to Tabulating a list of names of voters who have committed to the candidate . n exchange for mone. is required in order to distribute the money back to them. The list of names further allows candidates to estimate the number of votes they are likely to receive among this population of By contrast, non-transactional voters tend to base their choice on other considerations such as the cues they receive from community leaders based on factors such as the candidateAos personal morality . ccording to the broke. , religion, and so forth. Some of this influence is based on social dynamics within the community and can be related to a voterAos occupation, status as a student, and other similar circumstances. Business owners, for example, can be hired as a broker for a candidate and will attempt to persuade their workers to support that candidate. The workers can take cues from the business owner about which candidate to support and vote accordingly. This is also common among mosques and other religious institutions, educational institutions, and similar social or occupational settings. Because the broker has personal and social influence over those in his network, distributing cash is not necessary for persuading voters to support his A list of supporters in this case would be more difficult to acquire because much of this attempted influence is intended to be subtle. It is not as blunt as a religious leader commanding followers to vote for the candidate. Instead, the religious leader will make a statement about the types of characteristics to look for in a candidate and might informally tell members of his mosque that a particular candidate is a good, religious man who will take care of the people. These cues are enough to communicate clearly, if discretely, the preferences of the community leader. This type of influence, along with the number of voters that the broker can deliver to the candidate, is difficult for campaigns to measure and therefore voter lists are rarely used when candidates are targeting non-transactional voters. Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 135 George Towar Ikbal Tawakkal CONCLUSION This manuscript has examined differences in brokerage activity and explored the underlying reasons for this difference. What might first appear to be a trivial difference Ae whether a broker makes a list of voters or does not Ae in reality signals important distinctions about the roles of various brokers in a campaign as well as differences in the types of voters targeted by The basis for this understanding begins with the recognition that voters are viewed as a commodity that candidates wish to acquire. Ultimately, a candidate wants to gain votes in order to win the election, and will hire intermediaries known as brokers to acquire the votes he desires. Yet citizens differ in the factors that motivate their voting behavior, with some voters influenced by money and others influenced by social pressure created when a trusted community leader speaks favorably about a candidate. The existence of these differences creates the need for distinct strategies for voter persuasion, which leads to specialization and differentiation in the role and tasks of the brokers that are hired by the candidate. Specifically, some brokers are hired by candidates to provide monetary or other valuable goods to transactional voters in the form of cash payment or, somewhat rarely, collective goods such as public facilities. Other brokers are hired because they are community leaders, religious leaders, or other respected individuals within the In ways that are sometimes explicit and sometimes subtle, these brokers use their social status to signal cues to their community to persuade voters in favor of their candidate. While separate and distinct tactics targeting different types of voters, however. I found that often there is significant overlap and synthesis in campaigns with these two persuasion strategies. In a given village there will be voters who are mostly influenced by the money and others who are mostly influenced by the social and societal factors. These voters live, work, and interact in the same communities and delineating them can be difficult for candidates. Thus, in practice, candidates often combine both methods of brokerage activity to maximize their chances of persuading the largest number of voters. Uncertainty also exists about which factors best influence specific voters and candidates will often use both tactics on the same voters This is especially true in the smallest geographic electoral units in Indonesia . he village or the rukun tetangga or AuRTA. , where populations are often not homogenous. Second, the brokers and the candidates believe that some voters require both forms of persuasion to influence their vote. Several candidates running for an office will offer gifts to voters and there is little that candidates can do to prevent a voter from taking the money and voting for the opposing candidate. Thus, delivering money to the voter and reinforcing that commitment through assurances by community leaders that the candidate is a Augood manAy that is supported by respected individuals in the village or RT is believed to maximize the commitment of the voter. Vote buying, in other words, may not be enough to elicit a strong commitment by the voter and social pressure is believed by brokers to be effective at firming up their support for the candidate. In conclusion, brokerage activity in Indonesian elections in Central Java is more complex and nuanced than has been portrayed by the extant literature. Past research has tended to focus primarily on vote buying practices by brokers and the brokerAos individual decision to betray or remain loyal to the candidate that hired them. This research has also focused on the methods used by brokers to deliver cash payments directly to voters. Less studied is the role of other, more subtle strategies that employ social pressure to persuade voters. If the candidates view the voters as a commodity and hiring brokers is a way to acquire more of this commodity, then it makes sense that candidates and brokers would employ multiple strategies to enhance the effectiveness of their efforts. Underlying voting behavior in Indonesia are differences in the factors that influence voters, and these underlying differences help explain the often vast differences in how Copyright A 2024. JIP. ISSN: 2503-4685 (Prin. ISSN: 2528-0724 (Onlin. Jurnal Ilmu Pemerintahan: Kajian Ilmu Pemerintahan dan Politik Daerah. Vol. , 2024 - 136 George Towar Ikbal Tawakkal brokers try to persuade voters to support their candidates. Transactional voters view elections as a AutradeAy where they provide their votes in exchange for cash or other valuable goods. Brokers are required to create lists of names of voters committed to their candidate so that candidates can calculate how much money is required to AupurchaseAy the AucommodityAy and also to estimate their likely support on Election Day within the brokerAos area. Non-transactional voters are more influenced by social and societal factors such as cues from religious leaders, business owners, and other community leaders in their villages. These leaders can signal their support for a candidate with the presumption that many voters will follow that cue. Collecting data on these supporters is far more difficult and thus brokers do not attempt to collect a list of names of supporters, but instead campaigns attempt to calculate likely support based on the size of the brokerAos social network as well as the brokerAos likely influence within that network. Finally, campaigns often combine these strategies by hiring brokers to create lists and distribute cash as well as hiring brokers who use their social status as community leaders to persuade those in their social The reason for employing these different strategies simultaneously is because of the uncertainty about which voters are influenced most by which factors and also because the candidates and the brokers believe that some voters require both kinds of persuasion efforts before a voter will make a firm commitment to the candidate on Election Day REFERENCES