Voters have preferences over parties and candidates. Their ability to express these preferences in an election is constrained by the ballot choices that are presented to them and in the way in which ballots are aggregated into outcomes. In most cases voters do not sense the impact of these constraints, nor any tension between their preferences and their choices. In single member district systems with plurality rule (SMD), they simply vote for the candidate that they most prefer. In multi-candidate proportional representation (PR) elections, sincere voters’ rankings of candidates mirror their preference orderings. Sometimes, however, voters do not act sincerely, and make choices that do not follow from a straightforward mapping of their preferences.
The most heavily studied departure from sincere voting is that of tactical voting in SMD systems. This occurs when voters, believing that their most preferred candidate has no chance of winning, vote instead for a less preferred candidate so as to counter another candidate whom they dislike. In PR systems, voters might also conclude that they would be “wasting” their vote on a party that is unlikely to get enough votes to win a seat. Considerations of how different party coalitions are likely to form can also lead voters to engage in different forms of tactical voting. In this chapter we discuss tactical voting in SMD systems and in PR systems. We review and evaluate previous research, and identify the major findings that have emerged. We then suggest avenues for future research that are important and promising.
Farquharson (1964) provides a modern, game theoretic analysis of tactical voting, but, as he notes, it has a long and venerable heritage; Pliny the Younger describes an episode of tactical voting in the Roman Senate in the year 105. The most common formulation of tactical voting in SMD systems is associated with Duverger (1955). The winner-take-all nature of SMD contests strongly penalizes minor parties when votes are translated into seats. Voters are aware of the bias generated by this “mechanical factor,” which is thereby reinforced by a “psychological factor”: when voters perceive that their most preferred candidate is sure to lose, they often cast a tactical vote for a less preferred (but competitive and still acceptable) candidate in order to counter their least preferred candidate. Thus, according to Duverger’s Law, minor parties are suppressed, and elections in SMD systems are generally contested by two major parties.
Duverger’s Law is not a law, of course, but at most a general tendency. The factors he identifies apply to individual contests, and individual contests can aggregate into a wide variety of patterns. Britain, the US, Canada, and India all have SMD but very different party systems (Grofman, Blais, and Bowler 2009). In Britain the party system has become significantly more fractionalized over time even though SMD has remained in place. In any given election, moreover, tactical voting need not systematically help or hurt a particular party in all contests. Little about the nature and extent of tactical voting can be inferred from the nature of a country’s party system or from the outcome of national elections.
There are studies of tactical voting in countries besides Britain, but because of a pattern of party competition that persisted for decades – two major parties and a significant third party contesting most constituencies – research on tactical voting has focused primarily on British parliamentary elections. Early efforts to estimate the extent of tactical voting in Britain used aggregate-level constituency returns to track changes in vote totals across successive elections. The hypothesis they entertained is that a party that finishes third in one election suffers a loss of votes in the subsequent election, as discouraged supporters switch to voting for one of the two top parties. Parties that finish first or second in the first election correspondingly experience an increase in vote share in the next election due to the infusion of tactical votes. In his “flow of the vote” study of British elections held between 1892 and 1966, Shively (1970) concludes that tactical voting can be detected, but that it was insignificant in magnitude and impact. In their study of the 1983 and 1987 elections, Curtice and Steed (1988) also report that the extent of tactical voting was quite limited, and changed the outcomes of only a handful of individual constituency elections. Johnston and Pattie (1991) report finding more tactical voting in 1987 than in 1983, but even in 1987 the rate of tactical voting they detected was limited in extent and of minor consequence for election outcomes.
A problem with flow-of-the-vote analyses is that of establishing a proper baseline. A party that finished third in the first of two elections could well have had its totals depressed by tactical voting in the first election as well as the second. Parties that finished first or second could correspondingly have had their totals enhanced by tactical voting in both elections. Comparisons of constituency results across pairs of elections can also be compromised by omitted variable bias – for example, by changes in party support traceable to economic and demographic trends that are not accounted for in the analysis (Johnston 1981).
Many political scientists are also skeptical of flow-of-the-vote analyses because they are based upon constituency-level vote totals. As Campbell et al. (1966: 2) put it, “… As long as one has only aggregate data at hand one can only speculate as to what moved the individual members of the collectivity” (see also Russo in this volume). Subsequent studies of tactical voting have relied heavily upon survey data, and, specifically, upon a question posed in the British Election Study (BES) in every election since 1983: “Which one of the reasons comes closest to the main reason you voted for the Party you chose?” The responses offered vary from study to study, but in all elections respondents could report that “I really preferred another party, but it had no chance of winning in this constituency.” Those who chose this response are deemed to have voted tactically.
Analyses of BES survey data indicate that tactical voting is more prevalent than suggested by flow-of-the-vote studies (Niemi, Whitten, and Franklin 1992). Cain’s (1978) study of 1970 BES survey data, based instead upon differences in party feeling thermometer scores, reports a remarkably high rate of tactical voting – nearly 20 percent. Subsequent studies, however, point to a number of problems in survey-based estimates of tactical voting. According to Heath et al. (1991), some respondents said that they had voted tactically, but reported their favorite party to be the same party that they voted for. Others claim to have voted tactically but shifted their vote to another small party that had no better chance of winning than their favorite. The Heath et al. (1991) study, as well as Franklin, Niemi, and Whitten (1994), also find that nearly half the survey respondents who claim to have voted tactically most preferred a party that finished in first or second place. There is general agreement with Heath and Evans (1994) that “instrumental” tactical voting – that is, what Duverger had in mind – must be distinguished from an “expressive” version. In such cases voting decisions that respondents describe as tactical are motivated by idiosyncratic considerations that lead them to make a variety of often inexplicable choices.
These survey-based estimates of tactical voting, as well as estimates derived from flow-ofthe-votes, pertain to the entire electorate. According to Cox (1997), however, one should expect to see tactical voting only in constituencies in which the two leading parties are clearly identifiable, and not where uncertainty clouds the expected order of finish. Consider the following. In the first constituency the Liberal Democrat (LD) is widely expected to run well behind Labour and the Conservatives. Here it will be reasonable for LD supporters to consider casting a tactical vote. In the other constituency the Liberal Democrat finished third by only a few percentage points behind the second-place party in the last election, and the LD candidate is currently running a vigorous campaign. In this case LD supporters have reason to believe that they will finish second, or perhaps even first. The rationale for voting tactically thus disappears.
More generally, for tactical voting to occur, and to thus undermine support for minor parties and enhance the vote shares of the two major parties, voters must have common expectations as to which parties are viable – that is, running either first or second – and which are not (Palfrey 1989; Myatt and Fisher 2002). When common expectations about viability are present, supporters of the top two parties remain loyal and vote sincerely, while tactical supporters of nonviable parties abandon ship and vote for one of the top two. When common expectations about viability are not present, voters are uncertain as to whether the party they support is going to finish third or whether it is going to be one of the top two. Here tactical voting does not occur.
A common intuition concerning tactical voting, usually referred to as the marginality hypothesis, is that it should be more prevalent in close elections. Minor party supporters are seen to be more comfortable voting sincerely if one party or candidate has such a commanding lead that the outcome of the election is a foregone conclusion. It could also be that a close election makes the choice between the top two alternatives more salient and third parties less relevant. But tactical voting will not occur even in very close elections if voters are unable to distinguish between which parties are viable – that is, headed for a first- or second-place finish – and which are not. Votes can remain splintered among three or more parties, and a party can win with much less than an absolute majority. In Britain parliamentary elections, there are usually dozens of constituencies in which the winning party garners less than 40 percent of the vote.
A pair of studies reveals the importance of confining estimates of the rate of tactical voting to those constituencies where it makes sense for voters to consider voting tactically. In their analysis of BES survey data, Alvarez and Nagler (2000) regress individuals’ vote choices onto a wide range of variables, including demographics, issue distance measures, perceptions of economic conditions, and variables reflecting the pattern of party competition in the respondent’s constituency. They then counterfactually predict how voters would have voted absent tactical considerations and compare their prediction to how respondents reported voting. They report a tactical voting rate of 7.3 percent in 1987 – a little higher than that suggested by the standard tactical voting question, but in the same neighborhood. Alvarez, Boehmke, and Nagler (2005) then refine this analysis by confining their attention solely to supporters of parties who finished third (or worse) in their constituency in the previous election. Among this subset of respondents, they estimate that the rate of tactical voting was 43 percent.
Kiewiet (2013), similarly, stipulates that voters consider casting a tactical vote only when their most preferred party faces a “dismal outlook.” This occurs when their most preferred party (1) finished more than 10 percentage points behind the second-place party in their constituency in the previous election, and (2) fails to spend even half of the sum that could legally be spent on behalf of the individual candidate in the current election period. Research based upon the 1997–2001 British Election Panel Study strongly indicates that a poor third-place finish by a party in the previous election encourages its supporters to vote tactically (Fieldhouse, Shryane, and Pickles 2007) and that spending on behalf of individual candidates was consistent with tactical considerations by both parties and voters (Fieldhouse, Pattie, and Johnston 1996). Analyzing both NES survey data and constituency returns for all six parliamentary elections held in Britain between 1983 and 2005, Kiewiet finds that, in most elections, between one fourth and one third of the voters whose preferred party faced a dismal outlook cast tactical votes. Labour supporters appear to have voted tactically more frequently than LD supporters – in a couple of elections (1992 and 1997) at rates approaching 50 percent.
Kiewiet’s findings are also consistent with Myatt and Fisher (2002), who show that the pattern of party competition in individual constituencies is such that the mechanical and psychological components of Duverger’s Law can work in opposite directions. Liberal Democrats usually finished third and so were badly penalized by the translation of votes into seats. In many individual contests, however, they benefited from tactical votes that they received from supporters of Labour and other parties. Both of these effects result from the fact that they occupy the middle position on the ideological spectrum. Virtually all tactical votes cast by Labour and Conservative supporters went to the Liberal Democrats, while those cast by Liberal Democrats flowed both to the left (Labour) and to the right (Conservatives). In a recent, innovative study that utilizes Bayesian small sample estimation to infer district-level preferences from the inevitably small numbers of respondents sampled from each district, Hermann, Munzert, and Selb (2016) report finding similar patterns of tactical voting in the 1997 and 2001 British parliamentary elections.
As noted previously, several political scientists studying tactical voting have entertained the marginality hypothesis – that is, that it should be more prevalent in close elections. Early aggregate-level studies report some evidence favoring the marginality hypothesis in the 1966 and 1970 British general elections (Spafford 1972; Cain 1978). Since then, studies that have investigated this hypothesis with both BES and constituency-level data, with methodologies ranging from quite simple to highly sophisticated, have not found a consistent relationship between closeness and the rate of tactical voting (Lanoue and Bowler 1992; Fisher 2001; Alvarez, Boehmke, and Nagler 2005; Kiewiet 2013; Elff 2014).
In contrast to winner-take-all SMD systems, proportional representation (PR) systems award seats to parties in proportion to the votes they receive in multi-member districts. A voter’s most preferred party can finish third, fourth, or even worse, and still obtain a seat. For that reason Duverger (1955) doubted that voters would vote tactically in such systems. As Leys (1959) and Sartori (1968) note, however, if district magnitudes are low (i.e., only a few seats are allocated per district), it is common for small parties not to have enough votes to be awarded a seat. As in SMD systems, this could lead those who most prefer such a party to vote tactically for their most-preferred party that is likely to win at least one seat. The Leys–Sartori conjecture, then, is that the smaller the district magnitude in a PR election, the greater the extent of tactical voting that is likely to occur.
Analyzing the outcomes of small magnitude PR elections in Colombia and Japan, Cox and Shugart (1996) look for the presence of tactical voting by calculating “SF” ratios – that is, where F is the largest vote remainder and S the second largest vote remainder after assigning the final seat in the district. The SF ratio is a macro-level statistic, reflective of choices made by party elites and candidates as well as those of voters. An SF ratio approaching zero, however, signals the presence of tactical voting, and so is a useful diagnostic when more specific information about voters’ preferences and choices is not available. Their findings suggest that voters in both countries exhibited some amount of tactical voting by deserting parties unable to win a seat.
Cox and Shugart also report that tactical voting could not be detected in larger magnitude districts, and there are two reasons to accept this verdict. Mathematically, the larger district magnitude is, provided there is a low threshold, the smaller the share of votes a party needs to win at least one seat. Another consideration is informational. As in SMD systems, tactical desertion here also requires widely shared expectations regarding which parties are likely to obtain a seat and which are not. When district magnitudes are large – typically seen as more than five – voters may find it difficult to form such expectations and thus vote sincerely.
Gschwend, Stoiber, and Günther (2004) do not accept this line of argument. They argue instead that there are not insurmountable informational costs to tactical voting in PR systems, as voters can condition their decision upon what happened in the last election. Just as those looking for tactical voting in SMD systems hypothesize that supporters of parties that came in third in the previous election should be the most inclined to vote tactically, they posit that tactical voting should be most prevalent in PR systems among supporters of parties that either barely won a seat in the previous election or failed to win a seat. They call this an “election history heuristic” because it is not seen to require much from the voters in the way of information or thought. The findings of their analysis of elections in Finland between 1991 and 2003 indicate that marginal parties suffered less from tactical voting in large districts than in small ones (district magnitude in Finland during this period ranged from 6 to 33), but that non-negligible amounts of tactical voting occurred even in large magnitude districts.
Lago (2008) agrees that voters can use an electoral history heuristic regardless of district magnitude. Analyzing survey data collected in Spain in the 1970s and 1980s, he finds that voters have no more difficulty forming expectations about party viability in large magnitude districts than in small ones. There is less tactical desertion away from small parties in large magnitude districts simply because it is more likely for small parties to win a seat. Analyses of Spanish elections in 1977 and 1982 (Gunter 1989) and 2000 and 2008 (García-Viñuela, Artés, and Jurado 2015) report similar findings: supporters of small parties are more likely to cast tactical votes for larger parties when district magnitude is small, but even in large magnitude districts smaller party vote totals are suppressed by their supporters voting tactically for parties that win seats.
Another major reason why small parties fail to win seats in parliament, other than being squeezed by low district magnitude, is the presence of a threshold – that is, the requirement to win a certain percentage of votes (often 5 percent) in order to obtain a seat. Tavits and Annus (2006) find that voters in the new democracies of Eastern Europe and the former Soviet Union were sensitive to threshold requirements and tended to avoid casting votes for parties that fell below the threshold. Their findings are also supportive of the hypothesis that voters adopt an election history heuristic, in that there was clear evidence of voter learning over the course of successive elections. In consecutive elections in these countries, voters increasingly deserted parties whose support was either barely above the threshold of representation or fell below it. Additional evidence that voters in new democracies have come to adopt Duvergerian tactical reasoning comes from Duch and Palmer’s (2002) study of tactical voting in Hungary, which utilizes a complex three-layer, mixed member electoral system. Duch and Palmer ask voters whether they would be willing to switch their vote to another party if their most preferred party would get too few votes to get into parliament. About 14 percent of respondents answered this question positively, which they then use as an indicator of tactical voting. A probit analysis of these data shows that voters with stronger partisan attachments are less likely to behave tactically, and that those who are ideologically closer to their alternative party are more likely to vote tactically than those who are ideologically farther from it.
Political scientists have also sought to estimate the extent of tactical voting by comparing the choices voters make in mixed member district (MMD) systems, which have become common throughout the world. In the German MMD system, for example, about half the members of parliament are selected from party lists, with seats awarded in proportion to their national vote shares, and the other half from single member districts. Voters thus cast two votes, participating simultaneously in parallel PR and SMD elections. Assuming that voters vote sincerely when casting PR votes, votes cast for other parties’ candidates at the district level are deemed to be tactical in nature. Fisher (1973), Jesse (1987), and Bawn (1999) find that, in Germany, major party candidates garner larger percentages of votes in the SMD elections than their parties receive in the PR election. Reed (1999), similarly, finds that in the 1996 MMD election in Japan, voters who supported smaller parties cast larger percentages of votes for them in the PR election than in the district election.
According to Gschwend, Johnston, and Pattie (2003), German parties that had done poorly in the SMD constituency contest in the previous general election were disproportionately hurt by tactical voting – more evidence that voters use an election history heuristic in deciding whether or not to cast a tactical vote. Unlike most studies of tactical voting in pure SMD systems, Bawn finds some support for the marginality hypothesis: the extent of tactical voting was positively correlated with how close the race was between the top two candidates in districtlevel elections.
Both Bawn (1999) and Gschwend, Johnston, and Pattie (2003) report that, in the German MMD system, remarkably large percentages of small party supporters split their votes across the two elections. In the 1983 election over two thirds of those who voted for the minor party Free Democrats in the PR election, as well as nearly half of those who voted for the Greens, voted for a CDU/CSU or SPD candidate in the SMD contest. One reason why these percentages are so high is because they likely reflect more than just Duvergerian tactical voting. Some of the votes won by the SMD winners in excess of their party’s PR totals are “personal” votes reflecting their popularity with the local electorate (Moser and Scheiner 2005), or are due to an incumbency advantage (Bawn 1999).
SMD and PR systems aggregate votes into seats in a very different manner, but PR systems generally differ from SMD systems in another fundamental way: the election does not in and of itself determine who will and who will not be in the next government. This is determined instead by the nature of the party coalitions that form after the election. According to Hobolt and Karp (2010), in 17 Western European countries about two thirds of all governments formed since World War II have been coalitions. Voters presumably care about which parties end up in government, and not just about who gets elected. If so, they might well cast tactical votes based upon their expectations concerning which parties will form coalition governments in the postelection bargaining process.
Just as he had doubted that there would be tactical voting in PR systems motivated by the desire to avoid wasting a vote on a party unlikely to win a seat, Downs (1957) did not believe that voters’ decisions would be influenced by considerations of which post-election coalitions would emerge. Tactical voting of this nature would require voters to assess which coalitions were likely to form given different election outcomes, the probability that a party would enter a coalition, and the policy compromises parties would be willing to accept to enter each possible coalition. Few voters, it seemed to Downs (1957: 300), would be able to work out this complicated chain of inferences: “… in systems usually governed by coalitions, most citizens do not vote as though elections were government-selection mechanisms.”
Downs notwithstanding, Austen-Smith and Banks (1988) formally characterize the decision environment that voters (and parties) face in situations where coalitional governments are likely to occur, and define conditions under which some voters will engage in tactical voting motivated by coalitional considerations. It is also the case that voters can make use of informational shortcuts to reduce the potentially complex considerations involving election outcomes and coalition formation. Some possible coalitions can be discarded as practical impossibilities – for example, a partnership between the extreme right and extreme left parties. More generally, coalitions of parties that are close ideologically are more likely than ones in which the parties are further apart. There is the electoral history heuristic. Parties that formed a coalition in the past are more likely to form a coalition again. Armstrong and Duch (2010) document historical patterns of coalition governments in 34 countries and find that the composition of governing coalitions (and the prime minister’s party in that coalition) is quite stable over time. Finally, in countries where coalition governments are common, party elites can provide information about which coalitions they are willing to enter. In their study of the 2002 election in the Netherlands, Irwin and van Holsteyn (2012) find that voters had reasonably accurate beliefs about the number of seats each party would get and about which coalitions were likely to emerge. A significant number of voters assigned a high likelihood to the coalition government that actually ended up forming. Depending on the party, between 8 and 17 percent of the voters who were in a position to cast a tactical vote based upon coalitional considerations did so. These findings are impressive, given the complexity of the Dutch party system and elite reluctance to make preelectoral coalition arrangements (Blais et al. 2006).
Predicting what voters will do in response to coalitional expectations, however, is not as simple and straightforward as the Duvergerian hypothesis that they will seek to avoid wasting their votes on parties not in contention. In their study of Austrian elections, Meffert and Gschwend (2010) reason that tactical voting may take three different forms: (1) the standard “wasted vote” hypothesis that supporters of small parties not expected to exceed the electoral threshold defect to a larger party; (2) the “rental vote” hypothesis that some major party supporters vote for a junior coalition party in order to help it pass the minimum threshold for representation; and (3) tactical behavior aimed at influencing (balancing) the composition of the next coalition government. Their study relies on a pre-election survey that asked respondents about their party preferences, their vote intentions, and coalition preferences, as well as their expectations. Their findings show support for all three types of tactical behavior considered. Around 10 percent of small party supporters voted so as to not waste their vote, another 5–10 percent “rented” their vote out to a potential junior coalition partner, and a significant proportion was motivated by concerns about the composition of the governing coalition.
The results of Gschwend’s (2007) multinomial logit analysis of survey data from the 1998 MMD German election also support the rental vote hypothesis, in that some CDU/CSU supporters cast votes in the PR contest for their junior coalition partner, the FDP. It does not appear, however, that SPD voters rented out votes to their junior partners, the Greens. The results of a highly sophisticated econometric analysis of data from the 1994 German election by Shikano, Hermann, and Thurner (2009) are similar. About a quarter of CDU/CSU supporters who feared the FDP’s entry to parliament was at risk voted FDP, but they also failed to observe this type of behavior among supporters of the other large party, the SPD, which had a preelectoral coalition arrangement with the Greens. The authors speculate that this difference in behavior among the supporters of the two major parties is due to the fact that the 1994 election was the first time the Green party made pre-election coalition agreements with the SPD. The FDP, in contrast, had often made pre-election coalition agreements, and thus once again electoral history informed voters’ tactical behavior.
Bowler, Karp, and Donovan (2010) study the case of New Zealand during the 2002 general election. For this election, almost every New Zealander expected the Labour Party to be in the governing coalition, but there was uncertainty as to whom the junior coalition member would be, if any. Here, around 20 percent of the voters made tactical choices: if they believed their preferred party had little chance of being in government they were more likely to vote for their second most preferred party. If, on the other hand, they believed that their most disliked party had a good chance of being in government, they switched their vote to their second preference so as to block that outcome.
Blais et al. (2006) study tactical voting based upon coalitional considerations in what is probably the most challenging environment of all as far as information is concerned – the nationwide party list elections to the 120-member Israeli Knesset, where there is now a 3.25 percent threshold for representation (prior to 1988 the threshold was only 1 percent). In a national survey conducted prior to the 2003 election, respondents were asked to rate parties, to assess potential coalitions, and to indicate their vote preference. They defined tactical voting to occur in this situation when, instead of voting for their most preferred party, they voted instead for a party they believed would be a member of their most preferred coalition. Their analyses indicate that about 10 percent of Israeli voters voted tactically due to coalitional considerations. Individuals with stronger preferences for a particular party were less likely to cast tactical coalition votes, while those with stronger preferences for a particular coalition were more likely to do so.
Many political scientists have seen tactical voting as evidence that voters are rational and instrumental; rather than utterly forego the chance of affecting the election, they instead narrow their choice to candidates where their vote might be pivotal. In large-scale elections, however, a voter has far less probability of affecting the outcome than they do of getting hit by lightning several times. The standard rational choice model of tactical voting based on potential pivotality thus hardly seems consistent with a theory that purports to be based upon principles of rationality. The zero likelihood of being pivotal has led some to argue that it makes no sense to cast a tactical vote, but not casting a tactical vote, or not voting at all, has the same, utterly inconsequential impact on election outcomes.
It is important, therefore, to learn more about what exactly voters are thinking when they consider casting a tactical or protest vote. They may indeed harbor the illusion of possibly being pivotal, but the consistent evidence that rates of tactical voting are not related to the closeness of the elections suggests they do not. Alvarez and Kiewiet (2009) characterize tactical voters as “rationalistic”: they have well-behaved preference orderings, make choices that are consistent with these orderings, and sometimes vote for a lesser preferred candidate as if they were the pivotal voter. Another possibility is that tactical voters may see themselves as contributing a vote to a potentially pivotal voting bloc. But this of course does not lessen the problem that their single vote is inconsequential. Finally, tactical voters may not even be thinking in tactical terms at all, but may be simply restricting their voting choices to what they consider to be the feasible set of candidates (Elff 2014).
Although estimates vary considerably, available evidence indicates that in most cases around 15 percent to 40 percent of the voters who are in a position to cast a tactical vote actually do so. Those who strongly support their most preferred party or candidate are less likely to vote tactically and those who are more favorably disposed to a party in contention are more likely, but most potential tactical voters end up voting sincerely. Voters thus display what might be described as a sincerity bias. Why? Is it because they are simply unaware that tactical voting is an option? Or are they aware that they could vote tactically, but, like Pliny the Younger nearly 2,000 years ago, believe that it is morally wrong to misrepresent their preferences? What is needed, then, is more research on the psychology of tactical voting. It is an area of inquiry that is ripe for theoretical and empirical development.
Alvarez, R. M. and Kiewiet, D. R. (2009) “Rationality and Rationalistic Choice in the California Recall,” British Journal of Political Science, vol. 39, no. 2, April, 267–290.
Alvarez, R. M. and Nagler, J. (2000) “A New Approach for Modeling Strategic Voting in Multiparty Elections,” British Journal of Political Science, vol. 30, no. 1, January, 57–75.
Alvarez, R. M., Boehmke, F., and Nagler, J. (2005) “Strategic Voting in British Elections,” Electoral Studies, vol. 25, no. 1, March, 1–19.
Armstrong, D. A. and Duch, R. M. (2010) “Why Can Voters Anticipate Post-Election Coalition Formation Likelihoods,” Electoral Studies, vol. 29, no. 3, September, 308–315.
Austen-Smith, D. and Banks, J. (1988) “Elections, Coalitions, and Legislative Outcomes,” American Political Science Review, vol. 82, no. 2, June, 405–422.
Bawn, K. (1999) “Voter Responses to Electoral Complexity: Ticket Splitting, Rational Voters, and Representation in the Federal Republic of Germany,” British Journal of Political Science, vol. 29, no. 3, June, 487–505.
Blais, A., Aldrich, J. H., Indridason, I. H., and Levine, R. (2006) “Do Voters Vote for Government Coalitions? Testing Downs’ Pessimistic Conclusion,” Party Politics, vol. 12, no. 6, November, 691–705.
Bowler, S., Karp, J., and Donovan, T. (2010) “Strategic Coalition Voting: Evidence from New Zealand,” Electoral Studies, vol. 29, no. 3, September, 350–357.
Cain, Bruce (1978) “Strategic Voting in Britain,” American Journal of Political Science, vol. 22, no. 3, August, 639–655.
Campbell, A., Converse, P., Miller, W., and Stokes, D. (1966) Elections and the Political Order, New York: John Wiley and Sons.
Cox, G. W. (1997) Making Votes Count: Strategic Coordination in the World’s Electoral Systems, Cambridge: Cambridge University Press.
Cox, G. W. and Shugart, M. S. (1996) “Strategic Voting Under Proportional Representation,” Journal of Law, Economics, and Organization, vol. 12, no. 2, October, 299–324.
Curtice, J. and Steed, M. (1988) “Appendix 2: Analysis,” in Butler, D. and Kavanagh, D. (eds.) The British General Election of 1987, New York: St. Martin’s Press: 316–362.
Downs, A. (1957) An Economic Theory of Democracy, New York: Harper and Row.
Duch, R. M. and Palmer, H. D. (2002) “Strategic Voting in Post-Communist Democracy?,” British Journal of Political Science, vol. 32, no. 1, January, 63–91.
Duverger, M. (1955) Political Parties, New York: Wiley.
Elff, M. (2014) “Separating Tactical from Sincere Voting: A Finite-Mixture Discrete-Choice Modelling Approach to Disentangling Voting Calculi,” Paper presented at the 2014 Annual Meeting of the Midwest Political Science Association, Chicago, April 3–6.
Farquharson, R. (1964) Theory of Voting, New Haven: Yale University Press.
Fieldhouse, E., Pattie, C., and Johnston, R. (1996) “Tactical Voting and Constituency Campaigning at the 1992 General Election in England,” British Journal of Political Science, vol. 26, no. 3, July, 403–439.
Fieldhouse, E., Shryane, N., and Pickles, A. (2007) “Strategic Voting and Constituency Context: Modelling Party Preference and Vote in Multiparty Elections,” Political Geography, vol. 26, no. 2, February, 159–178.
Fisher, S. D. (2001) “Intuition Versus Formal Theory: Tactical Voting in England 1987–1997,” Paper presented at the 2001 Annual Meeting of the American Political Science Association, Washington DC, August 30–September 2.
Fisher, S. L. (1973) “The Wasted Vote Thesis: West German Evidence,” Comparative Politics, vol. 5, no. 2, January, 293–299.
Franklin, M., Niemi, R., and Whitten, G. (1994) “The Two Faces of Tactical Voting,” British Journal of Political Science, vol. 24, no. 4, October, 549–557.
García-Viñuela, E., Artés, J., and Jurado, I. (2015) “Strategic Voting and Non-Voting in Spanish Elections,” Party Politics, vol. 21, no. 5, September, 738–749.
Grofman, B., Blais, A., and Bowler, S. (2009) Duverger’s Law of Plurality Voting, New York: Springer.
Gschwend, T. (2007) “Ticket-Splitting and Strategic Voting under Mixed Electoral Rules: Evidence from Germany,” European Journal of Political Research, vol. 46, no. 1, January, 1–23.
Gschwend, T., Johnston, R., and Pattie, C. (2003) “Split-Ticket Patterns in Mixed-Member Proportional Election Systems: Estimates and Analyses of Their Spatial Variation at the German Federal Election, 1998,” British Journal of Political Science, vol. 33, no. 1, January, 109–127.
Gschwend, T., Stoiber, M., and Günther, M. (2004) “Strategic Voting in Proportional Systems: The Case of Finland,” Paper presented at the 2004 Annual Meeting of the Midwest Political Science Association, Chicago, April 15–18.
Gunther, R. (1989) “Electoral Laws, Party Systems, and Elites: The Case of Spain,” American Political Science Review, vol. 83, no. 3, September, 835–858.
Heath, A. and Evans, G. (1994) “Tactical Voting: Concepts, Measurement and Findings,” British Journal of Political Science, vol. 24, no. 4, October, 557–561.
Heath, A., Jowell, R., Curtice, J., Evans, G., Field, J., and Witherspoon, S. (1991) Understanding Political Change: The British Voter 1964–1987, Oxford: Pergamon Press.
Hermann, M., Munzert, S., and Selb, P. (2016) “Determining the Effect of Strategic Voting on Election Results,” Journal of Royal Statistical Society, vol. 179, no. 2, February, 583–605.
Hobolt, S. B. and Karp, J. A. (2010) “Voters and Coalition Governments,” Electoral Studies, vol. 29, no. 3, September, 299–307.
Irwin, G. A. and van Holsteyn, J. J. M. (2012) “Strategic Electoral Considerations Under Proportional Representation,” Electoral Studies, vol. 31, no. 1, September, 184–191.
Jesse, E. (1987) “Split-Voting in the Federal Republic of Germany: An Analysis of the Federal Elections from 1953 to 1987,” Electoral Studies, vol. 7, no. 2, August, 109–124.
Johnston, R. (1981) “Regional Variations in British Voting Trends – 1966–1979: Tests of an Ecological Model,” Regional Studies, vol. 15, no. 1, January, 23–32.
Johnston, R. and Pattie, C. (1991) “Tactical Voting in Britain in 1983 and 1987: An Alternative Approach,” British Journal of Political Science, vol. 21, no. 1, January, 95–128.
Kiewiet, D. R. (2013) “The Ecology of Tactical Voting in Britain,” Journal of Elections, Public Opinion and Parties, vol. 23, no. 1, January, 86–110.
Lago, I. (2008) “Rational Expectations or Heuristics? Strategic Voting in Proportional Representation Systems,” Party Politics, vol. 14, no. 1, January, 31–49.
Lanoue, D. and Bowler, S. (1992) “The Sources of Tactical Voting in British Parliamentary Elections, 1983–1987,” Political Behavior, vol. 14, no. 2, June, 141–157.
Leys, C. (1959) “Models, Theories and the Theory of Political Parties,” Political Studies, vol. 7, no. 2, June, 127–146.
Meffert, M. F. and Gschwend, T. (2010) “Strategic Coalition Voting: Evidence from Austria,” Electoral Studies, vol. 29, no. 3, September, 339–349.
Moser, R. G. and Scheiner, E. (2005) “Strategic Ticket Splitting and the Personal Vote in Mixed-Member Electoral Systems,” Legislative Studies Quarterly, vol. 30, no. 2, May, 259–276.
Myatt, D. and Fisher, S. D. (2002) “Tactical Coordination in Plurality Electoral Systems,” Oxford Review of Economic Policy, vol. 18, no. 4, Winter, 504–522.
Niemi, R., Whitten, G., and Franklin, M. (1992) “Constituency Characteristics, Individual Characteristics and Tactical Voting in the 1987 British General Election,” British Journal of Political Science, vol. 22, no. 2, April, 229–240.
Palfrey, T. (1989) “A Mathematical Proof of Duverger’s Law,” in Ordeshook, P. (ed.) Models of Strategic Choice in Politics, Ann Arbor: University of Michigan Press: 69–72.
Reed, S. R. (1999) “Strategic Voting in the 1996 Japanese General Election,” Comparative Political Studies, vol. 32, no. 2, April, 257–270.
Sartori, G. (1968) “Political Development and Political Engineering,” in Montgomery, J. D. and Hirschman, A. O. (eds.) Public Policy, Cambridge: Cambridge University Press: 261–298.
Shikano, S., Herrmann, M., and Thurner, P. W. (2009) “Strategic Voting Under Proportional Representation: Threshold Insurance in German Elections,” West European Politics, vol. 32, no. 3, April, 634–656.
Shively, W. P. (1970) “The Elusive ‘Psychological’ Factor: A Test for the Impact of Electoral Systems on Voters’ Behavior,” Comparative Politics, vol. 3, no. 1, October, 115–125.
Spafford, D. (1972) “Electoral Systems and Voters’ Behavior: Comment and a Further Test,” Comparative Politics, vol. 5, no. 1, October, 129–134.
Tavits, M. and Annus, A. (2006) “Learning to Make Votes Count: The Role of Democratic Experience,” Electoral Studies, vol. 25, no. 1, March, 72–90.