Skip to main content
Log in

Cue-Taking, Satisficing, or Both? Quasi-experimental Evidence for Ballot Position Effects

  • Original Paper
  • Published:
Political Behavior Aims and scope Submit manuscript

Abstract

Ballot position effects have been documented across a variety of political and electoral systems. In general, knowledge of the underlying mechanisms is limited. There is also little research on such effects in preferential-list PR systems, in which parties typically present ranked lists and thus signaling is important. This study addresses both gaps. Theoretically, we formalize four models of voter decision-making: pure appeal-based utility maximization, implying no position effects; rank-taking, where voters take cues from ballot position per se; satisficing, where choice is a function of appeal, but voters consider the options in the order of their appearance; and a hybrid “satisficing-with-rank-taking” variant. From these, we derive differential observable implications. Empirically, we exploit a quasi-experiment, created by the mixed-member electoral system that is used in the state of Bavaria, Germany. Particular electoral rules induce variation in both the observed rank and the set of competitors, and allow for estimating effects at all ranks. We find clear evidence for substantial position effects, which are strongest near the top, but discernible even for the 15th list position. In addition, a candidate’s vote increases when the average appeal of higher-placed (but not that of lower-placed) competitors is lower. Overall, the evidence is most compatible with the hybrid satisficing-with-rank-taking model. Ballot position thus affects both judgment and choice of candidates.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1

Similar content being viewed by others

Notes

  1. To avoid confusion, note that throughout the text we use the words “high” and “low” in the context of list ranks in a positional way, i.e. higher means further up.

  2. In 2011, from 30 European countries covered by Pilet et al. (2016), 60% used a PLPR system (ten with flexible lists, six with open lists and two with open list panachage systems).

  3. With semi-open/flexible lists, pre-electoral list ranks can matter for intra-party seat allocation after the votes are cast, since candidates need to reach a certain number of preference votes to qualify for being moved to the top of the post-electoral ranking (Marsh 1985; Renwick and Pilet 2016). However, even in fully open list PR (OLPR) systems, where this is not the case, lists may be intentionally ordered. In elections to the European Parliament, five out of ten OLPR cases (Estonia, Italy, Latvia, Lithuania and Poland) use ranked lists although the pre-electoral ranks have a direct influence on post-electoral vote–seat translation only in the case of a tie (Däubler and Hix 2018 , p. 3).

  4. Throughout the paper, we use the terms list position, list rank and ballot position interchangeably.

  5. As discussed by Rudolph and Däubler (2016, pp. 748–749) in more detail, this is a plausible approach in the context of OLPR. Many citizens can be expected to choose first a party and then a specific candidate, and even if they do not proceed that way, differences between the most preferred candidates within lists are unlikely to affect the choice between lists (as long as party-level factors weigh strong in voters’ utility function, which is typically the case in the parliamentary systems of Europe).

  6. Empirically, list position and the candidate vote would be correlated, of course – this, however, is only a consequence of unobserved confounders (e.g. media attention after the ranking is made, compare van Erkel and Thijssen 2016) that may add to the appeal of a candidate.

  7. Through a related mechanism, list rank may signal the viability of a candidate, which would matter to voters who choose strategically and may also anticipate that many other voters take list position as a cue. We do not elaborate on this, since strategic voting should not be relevant to the ballot position effect that we identify in our context and with our research design. In addition to our doubts that many voters cast their OLPR vote instrumentally, very sophisticated voters understanding the consequences of the complex Bavarian electoral system would also grasp that viability in the OLPR tier must be assessed on a region-wide basis, rather than using the list observed in the given SMD.

  8. Meredith and Salant (2013), on the other hand, report that simple satisficing cannot explain the patterns found in multi-winner races in California city council and school board elections, which employ random rotation.

  9. As of 2013; 2008: 91; 2003: 92.

  10. Although not doing so, i.e. indicating a party preference by markings on the list, has been deemed lawful by the constitutional court. Marcinkiewicz and Stegmaier (2015) suggested that obligatory preference voting is associated with stronger placement effects compared to systems in which the candidate vote is optional, and less informed voters may vote for the list “as a whole”. However, whether this argument extends to placement effects beyond the first list position is theoretically unclear.

  11. Thus, this model also rules out any anchoring (Ariely et al. 2003), contrast (Simonson and Tversky 1992), or other effects between the alternatives. We do by no means claim that they are irrelevant in the context of candidate choice but leave an incorporation of such effects for further research.

  12. As pointed out there, an equivalent way of interpreting the model is that candidate utility is fixed, but voters are uncertain about their reservation value.

  13. In case the value of a certain list position to a voter does not vary across parties or districts, it would actually be the same for any list of the same length.

  14. This is because such candidates are more readily accepted due to the value of their list position alone, and the assumed acceptance function has a sigmoid rather than linear shape (so additional gains decrease for baseline acceptance probabilities above .5).

  15. See http://landtagswahl2008.bayern.de/ for 2008 and http://landtagswahl2013.bayern.de/ for 2013. Additionally, we use information on individual candidates from Haus der Bayerischen Geschichte (http://www.hdbg.de/parlament/content/index.html).

  16. List length C varies between 13 and 60 candidates.

  17. We include zeros with half the observed value, i.e. we count zeros as half a vote.

  18. Put differently, our theoretical framework provides a theoretical rationale for the log-transformation, which is often motivated from a data perspective (Blom-Hansen et al. 2016; van Erkel and Thijssen 2016; Marcinkiewicz 2014).

  19. Even if there is correlation in candidate performance across tiers, there is no obvious bias, since the counterfactual focuses on another candidate: if we expect that some SMD-candidate A would do better in the OLPR tier of her SMD, this leads to a comparison of some other candidate B’s performance in SMDs in which A competes (and has not got the advantage) with B’s performance in the SMD of A (in which A would do better but does not appear on the list).

  20. Binary variables: district with 2013 CSU relatives affair (Rudolph and Däubler 2016; Kauder and Potrafke 2015), urban district (SMD district predominantly is a city or town); continuous variables, binarized at the 50th percentile: CSU first vote, employment share, immigrant share, population density, population influx, number of constructions, number of farmers, taxes per capita.

  21. Additionally, Online Appendix Figures A.2 and A.3 provide a graphical overview of the places of residence of candidates on general list position 1 to 3 and SMDs with a candidate on first, second or third OLPR list rank. As can be seen, there is some clustering in larger cities (especially Munich and Nuremberg; note, though, that these as well have 8 and respectively 4 local SMD districts). Overall, however, spatial patterns do not seem to point to systematic clustering of candidates with higher OLPR list rank in specific types of districts.

  22. Which they do, cf. Tables 2, 3 and A.5.

  23. Cf. Online Appendix Section A.2.

  24. When estimating our models for long lists, our results show similar patterns (see Table A.7).

  25. As such covariates, we use direct indicators of candidate appeal (incumbent MP, prominent politician) or indirect ones, such as the rank of the dropped candidate.

  26. Clustering at a higher level (we considered clustering at the district–year–party level) did not change our substantive conclusions; the changes to the standard errors were marginal. As clustering at a higher level is not the per se more conservative approach (standard errors are smaller for some coefficients), we cluster at the candidate level given the bias–variance trade-off inherent in choosing larger clusters (Cameron and Miller 2015).

  27. Online Appendix Table A.2 shows that our results hold for the full sample, and also with one single treatment dummy providing an average effect over all candidates (see discussion in Online Appendix Section A.4).

  28. With the exception of the coefficient for moving up from 7th to 10th as compared to 11th to 15th, although this difference is significant with controls in Model 2-2.

  29. Models A.2-1 and A.3-2 in the Online Appendix show that the argument of a diminishing return to list rank is mirrored in the (correlational) pattern that higher pre-election list rank is associated with increasingly higher vote shares.

  30. Also note that the candidate fixed effects capture the denominator of the expected vote share, so the model does not predict values out of range, see Online Appendix Section A.1. In addition, Table A.4 provides empirical evidence that the marginal benefit of appeal is higher in lower list positions, which anticipates the evidence for the satisficing-with-rank-taking model further discussed below.

  31. See Online Appendix Table A.3 for results displaying pre-election rank coefficients and including controls.

  32. Political attention during campaigns is usually focused on one candidate (“Spitzenkandidat”) in each party, who leads the campaign across the whole state. Since these candidates, who lead (no more than) one list, typically also run as SMD candidates, one may wonder to which extent their removal drives the observed differences in treatment effects shown in Fig. 1. Models 2 and 3 of Online Appendix Table A.8 contrast lists with and without Spitzenkandidat. While the difference in effects depending on which list position is dropped is much larger where a Spitzenkandidat takes baseline rank 1, it is still considerable in the remaining cases (In the 3\(\rightarrow\)2 case, the two treatment coefficients imply vote gains of 63% versus 21% even without a Spitzenkandidat being removed). Note, however, that Spitzenkandidaten tend to cluster in one (and the largest) OLPR district, Upper Bavaria (10 out of 14 observed cases).

  33. The indicator variable for being a prominent politician refers to the time just before the election and equals one for ministers (excluding junior ministers), party leaders, parliamentary party group leaders, party general secretaries, and state-wide campaign frontrunners (if these positions exist in the party).

  34. Note that only 20 candidates are removed below baseline list position two and only 12 below rank three. The latter are mainly CSU candidates in 2003. This could lead to statistical anomalies, which is also why we refrain from estimating effects for baseline list positions four or further down the list.

  35. We also assumed there is only one type of “average” voter, whom the empirical results characterize as deciding on the basis of a mixture of satisficing and rank-taking. On the other hand, we may think of the electorate as being composed of different types of voters, perhaps with variation in composition across districts and parties.

  36. Also note that parties may do this only because lists are not closed, and having the original order overturned may imply reputation costs. In this case, a correlation between list rank and candidate vote share must not lead to the wrong conclusion that PLPR is redundant.

  37. Parties may also strategically anticipate that ballot placement affects voters’ choices. The Bavarian Green Party, for example, places frontrunner candidates from time to time not at a top rank. This gives them de-facto-control over the entry of at least one other candidate via OLPR list votes. Similarly, the CSU regularly places SMD candidates at the bottom of lists in some regional districts, for other candidates to gain visibility.

References

  • Adams, J. (1999). Policy divergence in multicandidate probabilistic spatial voting. Public Choice, 100(1–2), 103–122.

    Article  Google Scholar 

  • André, A., Depauw, S., Shugart, M. S., & Chytilek, R. (2015). Party nomination strategies in flexible-list systems: Do preference votes matter? Party Politics. https://doi.org/10.1177/1354068815610974.

  • Angrist, J. D., & Pischke, J. S. (2008). Mostly harmless econometrics: An empiricist’s companion. Princeton University Press. https://doi.org/10.1057/be.2009.37.

  • Ariely, D., Loewenstein, G., & Prelec, D. (2003). “Coherent arbitrariness”: Stable demand curves without stable preferences. The Quarterly Journal of Economics, 118(1), 73–106.

    Article  Google Scholar 

  • Blom-Hansen, J., Elklit, J., Serritzlew, S., & Riis Villadsen, L. (2016). Ballot position and election results: Evidence from a natural experiment. Electoral Studies, 44, 172–183. https://doi.org/10.1016/j.electstud.2016.06.019.

    Article  Google Scholar 

  • Blumenau, J., Eggers, A. C., Hangartner, D., & Hix, S. (2017). Open/closed list and party choice: Experimental evidence from the UK. British Journal of Political Science, 47(4), 809–827. https://doi.org/10.1017/S0007123415000629.

    Article  Google Scholar 

  • Brockington, D. (2003). A low information theory of ballot position effect. Political Behavior, 25(1), 1–27.

    Article  Google Scholar 

  • Brüderl, J., & Ludwig, V. (2015). Fixed-effects panel regression. In H. Best, & C. Wolf (Eds.) Regression analysis and causal inference, pp. 327–357.

  • Cameron, A. C., & Miller, D. L. (2015). A practitioner’s guide to cluster-robust inference. Journal of Human Resources, 50(2), 317–372.

    Article  Google Scholar 

  • Caplin, A., Dean, M., & Martin, D. (2011). Search and satisficing. The American Economic Review, 101(7), 2899–2922.

    Article  Google Scholar 

  • Chen, E., Simonovits, G., Krosnick, J. A., & Pasek, J. (2014). The impact of candidate name order on election outcomes in North Dakota. Electoral Studies, 35, 115–122. https://doi.org/10.1016/j.electstud.2014.04.018.

    Article  Google Scholar 

  • Crisp, B. F., Olivella, S., Malecki, M., & Sher, M. (2013). Vote-earning strategies in flexible list systems: Seats at the price of unity. Electoral Studies. https://doi.org/10.1016/j.electstud.2013.02.007.

  • Dahlgaard, J. O. (2016). You just made it: Individual incumbency advantage under proportional representation. Electoral Studies, 44, 319–328. https://doi.org/10.1016/j.electstud.2016.09.004.

    Article  Google Scholar 

  • Däubler, T., & Hix, S. (2018). Ballot structure, list flexibility and policy representation. Journal of European Public Policy, 25(12), 1798–1816. https://doi.org/10.1080/13501763.2017.1361465

    Article  Google Scholar 

  • Einhorn, H. J., & Hogarth, R. M. (1981). Behavioral decision theory: Processes of judgement and choice. Annual Review of Psychology, 32(1), 53–88.

    Article  Google Scholar 

  • Faas, T., & Schoen, H. (2006). The importance of being first: Effects of candidates’ list positions in the 2003 Bavarian state election. Electoral Studies, 25(1), 91–102. https://doi.org/10.1016/j.electstud.2005.04.001.

    Article  Google Scholar 

  • Folke, O., Persson, T., & Rickne, J. (2015). The primary effect: Preference votes and political promotions. American Political Science Review, 110(3), 559–578.

    Article  Google Scholar 

  • Geys, B., & Heyndels, B. (2003). Influence of ‘cognitive sophistication’ on ballot layout effects. Acta Politica, 38(4), 295–311.

    Article  Google Scholar 

  • Gigerenzer, G., & Gaissmaier, W. (2011). Heuristic decision making. Annual Review of Psychology, 62, 451–482.

    Article  Google Scholar 

  • Górecki, M. A., & Kukołowicz, P. (2014). Gender quotas, candidate background and the election of women: A paradox of gender quotas in open-list proportional representation systems. Electoral Studies, 36, 65–80.

    Article  Google Scholar 

  • Górecki, M. A., & Marsh, M. (2012). Not just ‘friends and neighbours’: Canvassing, geographic proximity and voter choice. European Journal of Political Research, 51(5), 563–582.

    Article  Google Scholar 

  • Hernán, M. A., & VanderWeele, T. J. (2011). Compound treatments and transportability of causal inference. Epidemiology, 22(3), 368–77. https://doi.org/10.1097/EDE.0b013e3182109296.

    Article  Google Scholar 

  • Ho, D. E., & Imai, K. (2008). Estimating causal effects of ballot order from a randomized natural experiment: The California alphabet lottery, 1978–2002. Public Opinion Quarterly, 72(2), 216–240. https://doi.org/10.1093/poq/nfn018.

    Article  Google Scholar 

  • Jankowski, M. (2016). Voting for locals: Voters’ information processing strategies in open-list PR systems. Electoral Studies, 43, 72–84.

    Article  Google Scholar 

  • Katz, R. S., & Bardi, L. (1980). Preference voting and turnover in Italian parliamentary elections. American Journal of Political Science, 24(1), 97–114.

    Article  Google Scholar 

  • Kauder, B., & Potrafke, N. (2015). Just hire your spouse! Evidence from a political scandal in Bavaria. European Journal of Political Economy, 38, 42–54.

    Article  Google Scholar 

  • Kim, N., Krosnick, J., & Casasanto, D. (2015). Moderators of candidate name-order effects in elections: An experiment. Political Psychology, 36(5), 525–542. https://doi.org/10.1111/pops.12178.

    Article  Google Scholar 

  • Krosnick, J. A., & Alwin, D. F. (1987). An evaluation of a cognitive theory of response-order effects in survey measurement. Public Opinion Quarterly, 51(2), 201–219.

    Article  Google Scholar 

  • Lutz, G. (2010). First come, first served: The effect of ballot position on electoral success in open ballot PR elections. Representation, 46(2), 167–181.

    Article  Google Scholar 

  • Marcinkiewicz, K. (2014). Electoral contexts that assist voter coordination: Ballot position effects in Poland. Electoral Studies, 33, 322–334. https://doi.org/10.1016/j.electstud.2013.07.018.

    Article  Google Scholar 

  • Marcinkiewicz, K., & Stegmaier, M. (2015). Ballot position effects under compulsory and optional preferential-list PR electoral systems. Political Behavior, 37(2), 465–486.

    Article  Google Scholar 

  • Marsh, M. (1985). The voters decide? Preferential voting in european list systems. European Journal of Political Research, 13(4), 365–378.

    Article  Google Scholar 

  • Massicotte, L. (2011). Mixed systems. In J. M. Colomer (Ed.), Personal representation. The neglected dimension of electoral systems (pp. 97–115). Colchester: ECPR Press.

    Google Scholar 

  • Meier, B. P., & Robinson, M. D. (2004). Why the sunny side is up. Associations between affect and vertical position. Psychological Science, 15(4), 243–247.

    Article  Google Scholar 

  • Meredith, M., & Salant, Y. (2013). On the causes and consequences of ballot order effects. Political Behavior, 35(1), 175–197. https://doi.org/10.1007/s11109-011-9189-2.

    Article  Google Scholar 

  • Miller, J. M., & Krosnick, J. (1998). The impact of name order on election outcomes. Public Opinion Quarterly, 62(3), 291–330. https://doi.org/10.1086/297848.

    Article  Google Scholar 

  • Pasek, J., Schneider, D., Krosnick, J. A., Tahk, A., Ophir, E., & Milligan, C. (2014). Prevalence and moderators of the candidate name-order effect evidence from statewide general elections in California. Public Opinion Quarterly, 78(2), 416–439. https://doi.org/10.1093/poq/nfu013.

    Article  Google Scholar 

  • Pilet, J.B., Renwick, A., Núñez, L., Reimink, E., & Simón, P. (2016). Database of electoral systems. 2016. http://www.electoralsystemchanges.eu.

  • Redlawsk, D. P. (2004). What voters do: Information search during election campaigns. Political Psychology, 25(4), 595–610.

    Article  Google Scholar 

  • Renwick, A., & Pilet, J. B. (2016). Faces on the ballot. The personalization of electoral systems in Europe. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Rubinstein, A., & Salant, Y. (2006). A model of choice from lists. Theoretical Economics, 1(1), 3–17.

    Google Scholar 

  • Rudolph, L., & Däubler, T. (2016). Holding individual representatives accountable: The role of electoral systems. Journal of Politics, 78(3), 746–762. https://doi.org/10.1086/685378.

    Article  Google Scholar 

  • Shugart, M. S. (2005). Comparative electoral systems research: The maturation of a field and new challenges ahead. In M. Gallagher & P. Mitchell (Eds.), The politics of electoral systems (pp. 25–56). Oxford: Oxford University Press.

    Chapter  Google Scholar 

  • Shugart, M. S., Valdini, M. E., & Suominen, K. (2005). Looking for locals: Voter information demands and personal vote-earning attributes of legislators under proportional representation. American Journal of Political Science, 49(2), 437–449.

    Article  Google Scholar 

  • Simon, H. A. (1955). A behavioral model of rational choice. The Quarterly Journal of Economics, 69(1), 99–118.

    Article  Google Scholar 

  • Simonson, I., & Tversky, A. (1992). Choice in context: Tradeoff contrast and extremeness aversion. Journal of Marketing Research, 29(3), 281–295.

    Article  Google Scholar 

  • Sniderman, P. M. (2000). Taking sides: A fixed choice theory of political reasoning. In A. Lupia, M. D. McCubbins, & S. L. Popkin (Eds.), Elements of reason: Cognition, choice, and the bounds of rationality (pp. 67–84). Cambrdige: Cambridge University Press.

    Chapter  Google Scholar 

  • Stegmaier, M., Tosun, J., & Vlachová, K. (2014). Women’s parliamentary representation in the Czech Republic: Does preference voting matter? East European Politics & Societies, 28, 187–204.

    Article  Google Scholar 

  • Tavits, M. (2010). Effect of local ties on electoral success and parliamentary behaviour. The case of Estonia. Party Politics, 16(2), 215–235.

    Article  Google Scholar 

  • Thurner, P. W. (1998). Wählen als rationale Entscheidung. München: Oldenbourg.

    Book  Google Scholar 

  • Tourangeau, R., Couper, M. P., & Conrad, F. G. (2013). “Up means good”. The effect of screen position on evaluative ratings in web surveys. Public Opinion Quarterly, 77(S1), 69–88.

    Article  Google Scholar 

  • van Erkel, P. F., & Thijssen, P. (2016). The first one wins: Distilling the primacy effect. Electoral Studies, 44, 245–254. https://doi.org/10.1016/j.electstud.2016.09.002.

    Article  Google Scholar 

  • Villodres, C. O. (2003). Intra-party competition under preferential list systems: The case of Finland. Representation, 40(1), 55–66.

    Article  Google Scholar 

  • Wauters, B., Weekers, K., & Maddens, B. (2010). Explaining the number of preferential votes for women in an open-list PR system: An investigation of the 2003 federal elections in Flanders (Belgium). Acta Politica, 45(4), 468–490.

    Article  Google Scholar 

  • Zittlau, S., Gschwend, T., Blais, A., & Stephenson, L. (2017). Making Electoral Democracy Work (MEDW) 2013-2014 Bavaria Panel Study. https://doi.org/10.7910/DVN/YAHN5S.

Download references

Acknowledgements

We would like to thank Shaun Bowler, Alejandro Ecker, Anthony McGann, André Klima, Jon Krosnick, Moritz Marbach, Oliver Pamp and conference participants at EPSA and APSA 2017 for helpful comments. We are also grateful to Harald Schoen for valuable suggestions, to Harald Schoen and Thorsten Faas for sharing data, and to Ertan Bat for research assistance. Thomas Däubler acknowledges funding from the German Research Foundation, grant DA 1692/1-1. Lukas Rudolph acknowledges funding from the German Academic Scholarship Foundation. Replication materials are provided on https://doi.org/10.7910/DVN/SWORPS.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Thomas Däubler.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 1974 KB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Däubler, T., Rudolph, L. Cue-Taking, Satisficing, or Both? Quasi-experimental Evidence for Ballot Position Effects. Polit Behav 42, 625–652 (2020). https://doi.org/10.1007/s11109-018-9513-1

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11109-018-9513-1

Keywords

Navigation